Automatica
Journal Prestige (SJR): 3.896 Citation Impact (citeScore): 7 Number of Followers: 13 Hybrid journal (It can contain Open Access articles) ISSN (Print) 00051098 Published by Elsevier [3162 journals] 

K step+opacity+of+stochastic+discreteevent+systems&rft.title=Automatica&rft.issn=00051098&rft.date=&rft.volume=">Infinitestep opacity and K step opacity of stochastic discreteevent
systems Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Xiang Yin, Zhaojian Li, Weilin Wang, Shaoyuan Li Opacity is an important informationflow property that arises in security and privacy analysis of cyber–physical systems. Among many different notions of opacity, Kstep opacity requires that the intruder can never determine unambiguously that the system was at a secret state for any specific instant within K steps prior to that particular instant. This notion becomes infinitystep opacity when K goes to infinity. Existing works on the analysis of infinitestep opacity and Kstep opacity only provide a binary characterization, i.e., a system is either opaque or nonopaque. To analyze infinitestep and Kstep opacity more quantitatively, in this paper, we investigate the verification of infinitestep and Kstep opacity in the context of stochastic discreteevent systems. A new notion of opacity, called almost infinitestep opacity (respectively, almost Kstep opacity), is proposed to capture whether or not the probability of violating infinitestep opacity (respectively, Kstep opacity) is smaller than a given threshold. We also provide effective algorithms for the verification of almost infinitestep opacity and almost Kstep opacity.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Xiang Yin, Zhaojian Li, Weilin Wang, Shaoyuan Li Opacity is an important informationflow property that arises in security and privacy analysis of cyber–physical systems. Among many different notions of opacity, Kstep opacity requires that the intruder can never determine unambiguously that the system was at a secret state for any specific instant within K steps prior to that particular instant. This notion becomes infinitystep opacity when K goes to infinity. Existing works on the analysis of infinitestep opacity and Kstep opacity only provide a binary characterization, i.e., a system is either opaque or nonopaque. To analyze infinitestep and Kstep opacity more quantitatively, in this paper, we investigate the verification of infinitestep and Kstep opacity in the context of stochastic discreteevent systems. A new notion of opacity, called almost infinitestep opacity (respectively, almost Kstep opacity), is proposed to capture whether or not the probability of violating infinitestep opacity (respectively, Kstep opacity) is smaller than a given threshold. We also provide effective algorithms for the verification of almost infinitestep opacity and almost Kstep opacity.
 Distributed Kalman filtering for timevarying discrete sequential systems
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Bo Chen, Guoqiang Hu, Daniel W.C. Ho, Li Yu Discrete sequential system (DSS) consisting of different dynamical subsystems is a sequentiallyconnected dynamical system, and has found applications in many fields such as automation processes and series systems. However, few results are focused on the state estimation of DSSs. In this paper, the distributed Kalman filtering problem is studied for timevarying DSSs with Gaussian white noises. A locally optimal distributed estimator is designed in the linear minimum variance sense, and a stability condition is derived such that the mean square error of the distributed estimator is bounded. An illustrative example is given to demonstrate the effectiveness of the proposed methods.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Bo Chen, Guoqiang Hu, Daniel W.C. Ho, Li Yu Discrete sequential system (DSS) consisting of different dynamical subsystems is a sequentiallyconnected dynamical system, and has found applications in many fields such as automation processes and series systems. However, few results are focused on the state estimation of DSSs. In this paper, the distributed Kalman filtering problem is studied for timevarying DSSs with Gaussian white noises. A locally optimal distributed estimator is designed in the linear minimum variance sense, and a stability condition is derived such that the mean square error of the distributed estimator is bounded. An illustrative example is given to demonstrate the effectiveness of the proposed methods.
 Reinforcement learning for a class of continuoustime input constrained
optimal control problems Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Farnaz Adib Yaghmaie, David J. Braun In this paper, we identify a class of input constrained optimal control problems which can be approximately solved using Reinforcement Learning (RL) approaches. We start with a general class of problems which do not admit the theoretical assumptions used to derive RL frameworks. We then restrict this class by extra conditions on the dynamics and the objective function as deemed necessary. Our attention concerns two assumptions: (i) the smoothness of the value function which is typically not satisfied in input constrained problems, and (ii) the form of the objective function which can be more general than what has been proposed in previous formulations. For the first assumption, we use the method of vanishing viscosity to derive the conditions under which RL approaches can be used to find an approximate solution. These conditions relax a differentiability assumption to a continuity assumption of the value function thereby extending the applicability of RL frameworks. For the second assumption, we generalize the specific integrand form of the control cost used in previous formations to a more general class of cost functions that guarantee continuity of the control policy. Using these results, we present a new partially modelfree RL framework for optimal control of input constrained continuoustime systems. Our RL framework requires an initial stabilizing policy and guarantees uniformly ultimate boundedness of the state variables. We demonstrate our results by simulation examples.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Farnaz Adib Yaghmaie, David J. Braun In this paper, we identify a class of input constrained optimal control problems which can be approximately solved using Reinforcement Learning (RL) approaches. We start with a general class of problems which do not admit the theoretical assumptions used to derive RL frameworks. We then restrict this class by extra conditions on the dynamics and the objective function as deemed necessary. Our attention concerns two assumptions: (i) the smoothness of the value function which is typically not satisfied in input constrained problems, and (ii) the form of the objective function which can be more general than what has been proposed in previous formulations. For the first assumption, we use the method of vanishing viscosity to derive the conditions under which RL approaches can be used to find an approximate solution. These conditions relax a differentiability assumption to a continuity assumption of the value function thereby extending the applicability of RL frameworks. For the second assumption, we generalize the specific integrand form of the control cost used in previous formations to a more general class of cost functions that guarantee continuity of the control policy. Using these results, we present a new partially modelfree RL framework for optimal control of input constrained continuoustime systems. Our RL framework requires an initial stabilizing policy and guarantees uniformly ultimate boundedness of the state variables. We demonstrate our results by simulation examples.
 Networked stabilization of multiinput systems over shared channels with
scheduling/control codesign Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Wei Chen, Jing Yao, Li Qiu In this paper, we study the networked stabilization of a continuoustime multiinput system wherein the multiple control inputs are transmitted through a small number of shared channels with stochastic multiplicative uncertainties. Transmission scheduling over the shared channels needs to be performed so that at any time instant, each channel transmits only one control input. The main novelty of this work lies in the idea of scheduling/control codesign which suggests that the design of the transmission scheduling and the controller should be treated jointly rather than separately. By virtue of such a codesign, a sufficient condition is obtained for the channels’ overall quality of service required for stabilization given in terms of twice of the topological entropy of the plant. A numerical example is provided to illustrate our result.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Wei Chen, Jing Yao, Li Qiu In this paper, we study the networked stabilization of a continuoustime multiinput system wherein the multiple control inputs are transmitted through a small number of shared channels with stochastic multiplicative uncertainties. Transmission scheduling over the shared channels needs to be performed so that at any time instant, each channel transmits only one control input. The main novelty of this work lies in the idea of scheduling/control codesign which suggests that the design of the transmission scheduling and the controller should be treated jointly rather than separately. By virtue of such a codesign, a sufficient condition is obtained for the channels’ overall quality of service required for stabilization given in terms of twice of the topological entropy of the plant. A numerical example is provided to illustrate our result.
 An interpretation of the Schrödinger equation in quantum mechanics from
the controltheoretic point of view Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Akira Ohsumi In this paper, for a given (timedependent) Schrödinger equation in quantum mechanics, an interpretation of it is investigated from the perspective of stochastic control theory with the help of Nelson stochastic mechanics.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Akira Ohsumi In this paper, for a given (timedependent) Schrödinger equation in quantum mechanics, an interpretation of it is investigated from the perspective of stochastic control theory with the help of Nelson stochastic mechanics.
 Observerbased consensus for secondorder multiagent systems with
arbitrary asynchronous and aperiodic sampling periods Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Tomas Ménard, Emmanuel Moulay, Patrick Coirault, Michael Defoort A novel distributed consensus protocol, where only sampled position information is exchanged between neighboring agents, is designed for secondorder multiagent systems under a directed communication topology. This protocol allows to reach the consensus for asynchronous and aperiodic sampling periods, which means that every agent can send its measurements independently from its neighbors. Furthermore, the upper bound on the sampling periods can be chosen arbitrarily long by adapting the tuning parameters. This result is obtained by using a continuous–discrete time observer which allows to reconstruct the system state in real time from only discretetime measurements. The feedback control gain is set according to the observer gain which is itself set according to the maximum sampling period.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Tomas Ménard, Emmanuel Moulay, Patrick Coirault, Michael Defoort A novel distributed consensus protocol, where only sampled position information is exchanged between neighboring agents, is designed for secondorder multiagent systems under a directed communication topology. This protocol allows to reach the consensus for asynchronous and aperiodic sampling periods, which means that every agent can send its measurements independently from its neighbors. Furthermore, the upper bound on the sampling periods can be chosen arbitrarily long by adapting the tuning parameters. This result is obtained by using a continuous–discrete time observer which allows to reconstruct the system state in real time from only discretetime measurements. The feedback control gain is set according to the observer gain which is itself set according to the maximum sampling period.
 Detectability and observer design for switched
differential–algebraic equations Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Aneel Tanwani, Stephan Trenn This paper studies detectability for switched linear differential–algebraic equations (DAEs) and its application to the synthesis of observers, which generate asymptotically converging state estimates. Equating detectability to asymptotic stability of zerooutputconstrained state trajectories, and building on our work on intervalwise observability, we propose the notion of intervalwise detectability: If the output of the system is constrained to be identically zero over an interval, then the norm of the corresponding state trajectories scales down by a certain factor at the end of that interval. Conditions are provided under which the intervalwise detectability leads to asymptotic stability of zerooutputconstrained state trajectories. An application is demonstrated in designing state estimators. Decomposing the state into observable and unobservable components, we show that if the observable component of the system is reset appropriately and persistently, then the estimation error converges to zero asymptotically under the intervalwise detectability assumption.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Aneel Tanwani, Stephan Trenn This paper studies detectability for switched linear differential–algebraic equations (DAEs) and its application to the synthesis of observers, which generate asymptotically converging state estimates. Equating detectability to asymptotic stability of zerooutputconstrained state trajectories, and building on our work on intervalwise observability, we propose the notion of intervalwise detectability: If the output of the system is constrained to be identically zero over an interval, then the norm of the corresponding state trajectories scales down by a certain factor at the end of that interval. Conditions are provided under which the intervalwise detectability leads to asymptotic stability of zerooutputconstrained state trajectories. An application is demonstrated in designing state estimators. Decomposing the state into observable and unobservable components, we show that if the observable component of the system is reset appropriately and persistently, then the estimation error converges to zero asymptotically under the intervalwise detectability assumption.
 Ensuring privacy with constrained additive noise by minimizing Fisher
information Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Farhad Farokhi, Henrik Sandberg The problem of preserving the privacy of individual entries of a database when responding to linear or nonlinear queries with constrained additive noise is considered. For privacy protection, the response to the query is systematically corrupted with an additive random noise whose support is a subset or equal to a predefined constraint set. A measure of privacy using the inverse of the trace of the Fisher information matrix is developed. The Cramér–Rao bound relates the variance of any estimator of the database entries to the introduced privacy measure. The probability density that minimizes the trace of the Fisher information (as a proxy for maximizing the measure of privacy) is computed. An extension to dynamic problems is also presented. Finally, the results are compared to the differential privacy methodology.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Farhad Farokhi, Henrik Sandberg The problem of preserving the privacy of individual entries of a database when responding to linear or nonlinear queries with constrained additive noise is considered. For privacy protection, the response to the query is systematically corrupted with an additive random noise whose support is a subset or equal to a predefined constraint set. A measure of privacy using the inverse of the trace of the Fisher information matrix is developed. The Cramér–Rao bound relates the variance of any estimator of the database entries to the introduced privacy measure. The probability density that minimizes the trace of the Fisher information (as a proxy for maximizing the measure of privacy) is computed. An extension to dynamic problems is also presented. Finally, the results are compared to the differential privacy methodology.
 Robust eventtriggered state estimation: A risksensitive approach
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Jiarao Huang, Dawei Shi, Tongwen Chen In this work, we investigate a robust eventtriggered remote state estimation problem for linear Gaussian systems with a stochastic eventtriggering condition. The reference measure approach is used to obtain a robust eventtriggered estimate that minimizes the socalled risksensitive criterion, which refers to the expectation of the exponential of the sum of the squared estimation error. We introduce the reference measure, under which, the measurements are identically independently distributed (i.i.d.) and independent of the states, and propose a map to link the “realworld” measure to the reference measure so that the recursions of the information states under the reference measure can be obtained. Based on these results, the risksensitive criteria are reformulated under the reference measure and closedform expressions of the risksensitive eventtriggered posterior and prior estimates are presented, which are shown to evolve in simple recursive Kalmanlike structures. Moreover, two sufficient stability conditions for the proposed estimators are given, where the first requires the solution of a timevarying Riccati equation to be positivedefinite and satisfy a specific inequality, which can be further extended to the scenario when the weighting matrix in the risksensitive criterion is timevariant; the second gives the range of values of the risksensitive parameter and covariance of the initial state for which the proposed estimators are stable. Comparative simulation results demonstrate that the proposed risksensitive eventtriggered estimator is more robust to model uncertainties compared with a typical minimum mean squared error (MMSE) estimator with stochastic eventtriggered sensor scheduling.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Jiarao Huang, Dawei Shi, Tongwen Chen In this work, we investigate a robust eventtriggered remote state estimation problem for linear Gaussian systems with a stochastic eventtriggering condition. The reference measure approach is used to obtain a robust eventtriggered estimate that minimizes the socalled risksensitive criterion, which refers to the expectation of the exponential of the sum of the squared estimation error. We introduce the reference measure, under which, the measurements are identically independently distributed (i.i.d.) and independent of the states, and propose a map to link the “realworld” measure to the reference measure so that the recursions of the information states under the reference measure can be obtained. Based on these results, the risksensitive criteria are reformulated under the reference measure and closedform expressions of the risksensitive eventtriggered posterior and prior estimates are presented, which are shown to evolve in simple recursive Kalmanlike structures. Moreover, two sufficient stability conditions for the proposed estimators are given, where the first requires the solution of a timevarying Riccati equation to be positivedefinite and satisfy a specific inequality, which can be further extended to the scenario when the weighting matrix in the risksensitive criterion is timevariant; the second gives the range of values of the risksensitive parameter and covariance of the initial state for which the proposed estimators are stable. Comparative simulation results demonstrate that the proposed risksensitive eventtriggered estimator is more robust to model uncertainties compared with a typical minimum mean squared error (MMSE) estimator with stochastic eventtriggered sensor scheduling.
 Distributed estimation based on multihop subspace decomposition
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Álvaro Rodríguez del Nozal, Pablo Millán, Luis Orihuela, Alexandre Seuret, Luca Zaccarian This paper deals with the problem of distributedly estimating the state of an LTI plant through an interconnected network of agents. The proposed approach results in an observer structure that incorporates consensus among the agents and that can be distributedly designed, achieving a robust solution with a good estimation performance. The developed solution is based on an iterative decomposition of the plant in the local observable staircase forms. The proposed observer has several positive features compared to recent results in the literature, which include milder assumptions on the network connectivity and the ability to set the convergence rate.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Álvaro Rodríguez del Nozal, Pablo Millán, Luis Orihuela, Alexandre Seuret, Luca Zaccarian This paper deals with the problem of distributedly estimating the state of an LTI plant through an interconnected network of agents. The proposed approach results in an observer structure that incorporates consensus among the agents and that can be distributedly designed, achieving a robust solution with a good estimation performance. The developed solution is based on an iterative decomposition of the plant in the local observable staircase forms. The proposed observer has several positive features compared to recent results in the literature, which include milder assumptions on the network connectivity and the ability to set the convergence rate.
 Distributed algorithms for aggregative games of multiple heterogeneous
Euler–Lagrange systems Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Zhenhua Deng, Shu Liang In this paper, an aggregative game of Euler–Lagrange (EL) systems is investigated, where the cost functions of all players depend on not only their own decisions but also the aggregate of all decisions. Two distributed algorithms are designed for these heterogeneous EL players to reach the Nash equilibrium of aggregative games. By constructing suitable Lyapunov functions, the convergence of the two algorithms are analyzed. The first algorithm achieves globally exponential convergence without parameter uncertainty, and the other achieves globally asymptotic convergence, even in the presence of uncertain parameters. Numerical examples are given to illustrate the effectiveness of the methods.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Zhenhua Deng, Shu Liang In this paper, an aggregative game of Euler–Lagrange (EL) systems is investigated, where the cost functions of all players depend on not only their own decisions but also the aggregate of all decisions. Two distributed algorithms are designed for these heterogeneous EL players to reach the Nash equilibrium of aggregative games. By constructing suitable Lyapunov functions, the convergence of the two algorithms are analyzed. The first algorithm achieves globally exponential convergence without parameter uncertainty, and the other achieves globally asymptotic convergence, even in the presence of uncertain parameters. Numerical examples are given to illustrate the effectiveness of the methods.
 On robust Kalman filter for twodimensional uncertain linear discrete
timevarying systems: A least squares method Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Dong Zhao, Steven X. Ding, Hamid Reza Karimi, Yueyang Li, Youqing Wang The robust Kalman filter design problem for twodimensional uncertain linear discrete timevarying systems with stochastic noises is investigated in this study. First, we prove that the solution to a certain deterministic regularized least squares problem constrained by the nominal twodimensional system model is equivalent to the generalized twodimensional Kalman filter. Then, based on this relationship, the robust state estimation problem for twodimensional uncertain systems with stochastic noises is interpreted as a deterministic robust regularized least squares problem subject to twodimensional dynamic constraint. Finally, by solving the robust regularized least squares problem and using a simple approximation, a recursive robust twodimensional Kalman filter is determined. A heat transfer process serves as an example to show the properties and efficacy of the proposed filter.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Dong Zhao, Steven X. Ding, Hamid Reza Karimi, Yueyang Li, Youqing Wang The robust Kalman filter design problem for twodimensional uncertain linear discrete timevarying systems with stochastic noises is investigated in this study. First, we prove that the solution to a certain deterministic regularized least squares problem constrained by the nominal twodimensional system model is equivalent to the generalized twodimensional Kalman filter. Then, based on this relationship, the robust state estimation problem for twodimensional uncertain systems with stochastic noises is interpreted as a deterministic robust regularized least squares problem subject to twodimensional dynamic constraint. Finally, by solving the robust regularized least squares problem and using a simple approximation, a recursive robust twodimensional Kalman filter is determined. A heat transfer process serves as an example to show the properties and efficacy of the proposed filter.
 Stability analysis of a system coupled to a heat equation
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Lucie Baudouin, Alexandre Seuret, Frédéric Gouaisbaut As a first approach to the study of systems coupling finite and infinite dimensional natures, this article addresses the stability of a system of ordinary differential equations coupled with a classic heat equation using a Lyapunov functional technique. Inspired from recent developments in the area of time delay systems, a new methodology to study the stability of such a class of distributed parameter systems is presented here. The idea is to use a polynomial approximation of the infinite dimensional state of the heat equation in order to build an enriched energy functional. A well known efficient integral inequality (Bessel inequality) will allow to obtain stability conditions expressed in terms of linear matrix inequalities. We will eventually test our approach on academic examples in order to illustrate the efficiency of our theoretical results.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Lucie Baudouin, Alexandre Seuret, Frédéric Gouaisbaut As a first approach to the study of systems coupling finite and infinite dimensional natures, this article addresses the stability of a system of ordinary differential equations coupled with a classic heat equation using a Lyapunov functional technique. Inspired from recent developments in the area of time delay systems, a new methodology to study the stability of such a class of distributed parameter systems is presented here. The idea is to use a polynomial approximation of the infinite dimensional state of the heat equation in order to build an enriched energy functional. A well known efficient integral inequality (Bessel inequality) will allow to obtain stability conditions expressed in terms of linear matrix inequalities. We will eventually test our approach on academic examples in order to illustrate the efficiency of our theoretical results.
 Generalization of a Result of Fabian on the Asymptotic Normality of
Stochastic Approximation Abstract: Publication date: Available online 10 November 2018Source: AutomaticaAuthor(s): Karla Hernández, James C. Spall Stochastic approximation (SA) is a general framework for analyzing the convergence of a large collection of stochastic rootfinding algorithms. The Kiefer–Wolfowitz and stochastic gradient algorithms are two wellknown (and widely used) examples of SA. Because of their applicability to a wide range of problems, many results have been obtained regarding the convergence properties of SA procedures. One important reference in the literature, Fabian (1968), derives general conditions for the asymptotic normality of the SA iterates. Since then, many results regarding asymptotic normality of SA procedures have relied heavily on Fabian’s theorem. Unfortunately, some of the assumptions of Fabian’s result are not applicable to some modern implementations of SA in control and learning. In this paper we explain the nature of this incompatibility and show how Fabian’s theorem can be generalized to address the issue.
 Abstract: Publication date: Available online 10 November 2018Source: AutomaticaAuthor(s): Karla Hernández, James C. Spall Stochastic approximation (SA) is a general framework for analyzing the convergence of a large collection of stochastic rootfinding algorithms. The Kiefer–Wolfowitz and stochastic gradient algorithms are two wellknown (and widely used) examples of SA. Because of their applicability to a wide range of problems, many results have been obtained regarding the convergence properties of SA procedures. One important reference in the literature, Fabian (1968), derives general conditions for the asymptotic normality of the SA iterates. Since then, many results regarding asymptotic normality of SA procedures have relied heavily on Fabian’s theorem. Unfortunately, some of the assumptions of Fabian’s result are not applicable to some modern implementations of SA in control and learning. In this paper we explain the nature of this incompatibility and show how Fabian’s theorem can be generalized to address the issue.
 On periodic optimal solutions of persistent sensor planning for
continuoustime linear systems Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): JungSu Ha, HanLim Choi This work investigates informative planning of sensing agents over infinite time horizon when the system of interest is expressed as a continuoustime linear system. The objective of this planning problem, termed persistent monitoring problem, is to maintain the monitoring uncertainty at the minimum. The method reduces the persistent planning problem into a periodic planning problem; it is formulated as a periodic optimal control or optimization problem to determine the optimal periodic sensor plan as well as the period. The plan induces the periodic Riccati equation and is proven to lead an arbitrary initial uncertainty state to the optimized periodic trajectory. It is also proven that any infinitehorizon (nonperiodic) sensor plan is able to be approximated arbitrary well by a periodic sensor plan. A suboptimal filtering mechanism is proposed by using the resulting optimal periodic solution. Two numerical examples on (a) a relaxed periodic sensor scheduling for a two dimensional linear system, and (b) persistent monitoring by a mobile sensor of twodimensional diffusion dynamics show the validity of the proposed approach.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): JungSu Ha, HanLim Choi This work investigates informative planning of sensing agents over infinite time horizon when the system of interest is expressed as a continuoustime linear system. The objective of this planning problem, termed persistent monitoring problem, is to maintain the monitoring uncertainty at the minimum. The method reduces the persistent planning problem into a periodic planning problem; it is formulated as a periodic optimal control or optimization problem to determine the optimal periodic sensor plan as well as the period. The plan induces the periodic Riccati equation and is proven to lead an arbitrary initial uncertainty state to the optimized periodic trajectory. It is also proven that any infinitehorizon (nonperiodic) sensor plan is able to be approximated arbitrary well by a periodic sensor plan. A suboptimal filtering mechanism is proposed by using the resulting optimal periodic solution. Two numerical examples on (a) a relaxed periodic sensor scheduling for a two dimensional linear system, and (b) persistent monitoring by a mobile sensor of twodimensional diffusion dynamics show the validity of the proposed approach.
 Design of supertwisting control gains: A describing function based
methodology Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Ulises PérezVentura, Leonid Fridman The Describing Function approach is used to adjust the parameters of fastoscillations (chattering) caused by the presence of fastactuators in SuperTwisting control loops. Estimated parameters, amplitude and frequency of selfexcited oscillations, allow to compute the average power needed to maintain the trajectories of the system into real slidingmodes. Through the parametrization of the actuator dynamics by a critically damped secondorder system or by a constant delay, sets of STA gains are provided to minimize the amplitude of oscillations or the average power. The results are confirmed by simulations.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Ulises PérezVentura, Leonid Fridman The Describing Function approach is used to adjust the parameters of fastoscillations (chattering) caused by the presence of fastactuators in SuperTwisting control loops. Estimated parameters, amplitude and frequency of selfexcited oscillations, allow to compute the average power needed to maintain the trajectories of the system into real slidingmodes. Through the parametrization of the actuator dynamics by a critically damped secondorder system or by a constant delay, sets of STA gains are provided to minimize the amplitude of oscillations or the average power. The results are confirmed by simulations.
 Switching and information exchange in compressed estimation of coupled
high dimensional processes Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Karan Narula, Jose E. Guivant Compressed Estimation approaches, such as the Generalised Compressed Kalman Filter (GCKF), reduce the computational cost and complexity of highdimensional and highfrequency data assimilation problems, usually without sacrificing optimality. Configured using adequate cores, such as the Unscented Kalman Filter (UKF), the GCKF could also treat certain highdimensional nonlinear cases. However, the application of a compressed estimation process is limited to a class of problems which inherently allow the estimation process to be divided, at certain intervals of time, into a set of lowerdimensional problems. This limitation prohibits applying the compressing techniques for estimating coupled highdimensional processes. However, those limitations can be overcome by applying proper techniques. In this paper, the concepts of subsystem switching and information exchange architecture, namely ‘Exploiting Local Statistical Dependency’ (ELSD), have been derived and explored, allowing compressed estimators to mimic optimal fullGaussian estimators. The performances of the methods have been verified through applications in solving usual types of Stochastic Partial Differential Equations (SPDEs). The computational advantages of using the proposed techniques have also been highlighted with a recommendation for its usage over the full filter when dealing with highdimensional and highfrequency data assimilation.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Karan Narula, Jose E. Guivant Compressed Estimation approaches, such as the Generalised Compressed Kalman Filter (GCKF), reduce the computational cost and complexity of highdimensional and highfrequency data assimilation problems, usually without sacrificing optimality. Configured using adequate cores, such as the Unscented Kalman Filter (UKF), the GCKF could also treat certain highdimensional nonlinear cases. However, the application of a compressed estimation process is limited to a class of problems which inherently allow the estimation process to be divided, at certain intervals of time, into a set of lowerdimensional problems. This limitation prohibits applying the compressing techniques for estimating coupled highdimensional processes. However, those limitations can be overcome by applying proper techniques. In this paper, the concepts of subsystem switching and information exchange architecture, namely ‘Exploiting Local Statistical Dependency’ (ELSD), have been derived and explored, allowing compressed estimators to mimic optimal fullGaussian estimators. The performances of the methods have been verified through applications in solving usual types of Stochastic Partial Differential Equations (SPDEs). The computational advantages of using the proposed techniques have also been highlighted with a recommendation for its usage over the full filter when dealing with highdimensional and highfrequency data assimilation.
 Decentralised estimation with correlation limited by optimal processing of
independent data Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Jiří Ajgl, Ondřej Straka Decentralised estimation aims at providing the best combination of multiple estimates. Since the exact solutions are expensive in terms of computation and communication requirements, the mean square error optimality is traded for the bound optimality. Fusion under unknown correlations has been inspected for two decades and the research now focuses on a partial knowledge of the correlations. This paper focuses on the assumption that the estimates to be fused were obtained by the optimal processing of local data with independent measurement errors. A generalisation of a recent solution to such a problem is proposed. In particular, nonuniqueness of the optimal fusion weights is discovered. The relation of the generalised and existing solutions is discussed and illustrative examples are given.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Jiří Ajgl, Ondřej Straka Decentralised estimation aims at providing the best combination of multiple estimates. Since the exact solutions are expensive in terms of computation and communication requirements, the mean square error optimality is traded for the bound optimality. Fusion under unknown correlations has been inspected for two decades and the research now focuses on a partial knowledge of the correlations. This paper focuses on the assumption that the estimates to be fused were obtained by the optimal processing of local data with independent measurement errors. A generalisation of a recent solution to such a problem is proposed. In particular, nonuniqueness of the optimal fusion weights is discovered. The relation of the generalised and existing solutions is discussed and illustrative examples are given.
 Innovative fractional derivative estimation of the pseudostate for a
class of fractional order linear systems Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): YanQiao Wei, DaYan Liu, Driss Boutat In this paper, a nonasymptotic and robust method is proposed to estimate the fractional integral and derivative of the pseudostate for a class of fractional order linear systems in noisy environment with unknown initial conditions. To the best of our knowledge, no method has been developed for such estimation. Firstly, the estimation problem is transformed into estimating the fractional integral and derivative of the output and a set of fractional derivative initial values. Then, algebraic integral formulas are exactly derived for the sought estimators by applying different modulating functions with specified properties. In particular, a design parameter is introduced in the formulas of fractional derivative initial values, which can improve the robustness by choosing appropriate values. Secondly, it is shown how to construct the required modulating functions in an efficient way, where another design parameter is involved. Moreover, some error analysis is given to choose the design parameters. Finally, numerical simulations are provided to demonstrate the efficiency and the robustness to noises of the proposed method.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): YanQiao Wei, DaYan Liu, Driss Boutat In this paper, a nonasymptotic and robust method is proposed to estimate the fractional integral and derivative of the pseudostate for a class of fractional order linear systems in noisy environment with unknown initial conditions. To the best of our knowledge, no method has been developed for such estimation. Firstly, the estimation problem is transformed into estimating the fractional integral and derivative of the output and a set of fractional derivative initial values. Then, algebraic integral formulas are exactly derived for the sought estimators by applying different modulating functions with specified properties. In particular, a design parameter is introduced in the formulas of fractional derivative initial values, which can improve the robustness by choosing appropriate values. Secondly, it is shown how to construct the required modulating functions in an efficient way, where another design parameter is involved. Moreover, some error analysis is given to choose the design parameters. Finally, numerical simulations are provided to demonstrate the efficiency and the robustness to noises of the proposed method.
 Dynamics of semilattice networks with strongly connected dependency graph
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Alan VelizCuba, Reinhard Laubenbacher Discretetime dynamical systems on a finite state space have been used to model natural and engineered systems such as biological networks, social networks, and engineered control systems. They have the advantage of being intuitive and the models can be easily simulated on a computer in most cases; however, few analytical tools beyond simulation are available. The motivation for this paper is to develop such tools. It identifies a broad class of discrete dynamical systems with a finite phase space for which one can derive strong results about their longterm dynamics in terms of properties of their dependency graphs. The paper contains a complete classification of the periodic orbits of semilattice networks with strongly connected dependency graph, by finding analytically the exact number of periodic orbits of any given period.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Alan VelizCuba, Reinhard Laubenbacher Discretetime dynamical systems on a finite state space have been used to model natural and engineered systems such as biological networks, social networks, and engineered control systems. They have the advantage of being intuitive and the models can be easily simulated on a computer in most cases; however, few analytical tools beyond simulation are available. The motivation for this paper is to develop such tools. It identifies a broad class of discrete dynamical systems with a finite phase space for which one can derive strong results about their longterm dynamics in terms of properties of their dependency graphs. The paper contains a complete classification of the periodic orbits of semilattice networks with strongly connected dependency graph, by finding analytically the exact number of periodic orbits of any given period.
 Output feedback control for unknown LTI systems driven by unknown periodic
disturbances Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Cemal Tugrul Yilmaz, Halil Ibrahim Basturk This paper considers unknown minimumphase LTI systems with known relative degree and system order. The main aim is to reject the unknown, unmatched sinusoidal disturbances and make the output track a given trajectory with the output feedback. The essence of the control design is composed of disturbance parametrization, Kfilter technique and adaptive backstepping procedure. Firstly, the unmeasured system states are represented in terms of filtered input and output signals. Then, the disturbance information in the output signal is parametrized and the problem is converted to an adaptive control problem. After that, the Kfilter approach is employed to redefine the system states that enable to use a backstepping technique. An adaptive output feedback controller is designed recursively. It is proven that the equilibrium at the origin is globally uniformly stable and the output signal tracks the reference signal asymptotically. Finally, the effectiveness of the controller is illustrated with a numerical simulation. The robustness of the closed loop system with respect to an additive unmodelled noise is also discussed.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Cemal Tugrul Yilmaz, Halil Ibrahim Basturk This paper considers unknown minimumphase LTI systems with known relative degree and system order. The main aim is to reject the unknown, unmatched sinusoidal disturbances and make the output track a given trajectory with the output feedback. The essence of the control design is composed of disturbance parametrization, Kfilter technique and adaptive backstepping procedure. Firstly, the unmeasured system states are represented in terms of filtered input and output signals. Then, the disturbance information in the output signal is parametrized and the problem is converted to an adaptive control problem. After that, the Kfilter approach is employed to redefine the system states that enable to use a backstepping technique. An adaptive output feedback controller is designed recursively. It is proven that the equilibrium at the origin is globally uniformly stable and the output signal tracks the reference signal asymptotically. Finally, the effectiveness of the controller is illustrated with a numerical simulation. The robustness of the closed loop system with respect to an additive unmodelled noise is also discussed.
 Sampleddata emulation of dynamic output feedback controllers for
nonlinear timedelay systems Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Mario Di Ferdinando, Pierdomenico Pepe In this paper we deal with the problem of the stabilization in the sampleandhold sense by emulation of continuoustime dynamic output feedback controllers. Nonlinear timedelay systems not necessarily affine in the control input are studied. Sufficient conditions are provided such that the emulation of continuoustime dynamic output feedback controllers yields stabilization in the sampleandhold sense. The intersampling system behavior as well as timevarying sampling intervals are taken into account. The case of nonlinear delayfree systems is addressed as a special case. In this case, it is shown that the sufficient conditions, ensuring the stabilization in the sampleandhold sense, are satisfied if the continuoustime dynamic output feedback controller is a global stabilizer. An example is presented which validates the theoretical results.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Mario Di Ferdinando, Pierdomenico Pepe In this paper we deal with the problem of the stabilization in the sampleandhold sense by emulation of continuoustime dynamic output feedback controllers. Nonlinear timedelay systems not necessarily affine in the control input are studied. Sufficient conditions are provided such that the emulation of continuoustime dynamic output feedback controllers yields stabilization in the sampleandhold sense. The intersampling system behavior as well as timevarying sampling intervals are taken into account. The case of nonlinear delayfree systems is addressed as a special case. In this case, it is shown that the sufficient conditions, ensuring the stabilization in the sampleandhold sense, are satisfied if the continuoustime dynamic output feedback controller is a global stabilizer. An example is presented which validates the theoretical results.
 Analysis and synthesis for a class of stochastic switching systems against
delayed mode switching: A framework of integrating mode weights Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Lixian Zhang, Zepeng Ning, Yang Shi This paper is concerned with the issues of stability analysis and control synthesis for a class of linear stochastic switching systems in discretetime domain. The switching dynamics are considered to be governed by a semiMarkov process and the sojourn time for each system mode is deemed to be finite. A modedependent control scheme is employed with an adaptation sense in the presence of time delays in the mode switching of controller, which is manifested as a constant lag between the system mode and the controller mode. A novel form of Lyapunov function is adopted, in which the Lyapunov matrix depends on the modes of both the system and the controller as well as the time since the occurrence of the last mode switching. On the basis of the new proposed σerror meansquare stability that integrates the weights of all the system modes, numerically testable stability criteria are developed via the semiMarkov kernel approach. In virtue of certain techniques that can eliminate the terms containing powers or products of matrices, a desired modedependent stabilizing controller is designed such that the closedloop system is σerror meansquare stable by allowing a modeunmatched controller to perform before the controller switches to a modematched one. Finally, the theoretical results are applied to a practical example of one joint of a space robot manipulator to demonstrate the effectiveness, applicability and superiority of the proposed control strategy as well as the necessity of considering the modeswitching delays in the designed controller.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Lixian Zhang, Zepeng Ning, Yang Shi This paper is concerned with the issues of stability analysis and control synthesis for a class of linear stochastic switching systems in discretetime domain. The switching dynamics are considered to be governed by a semiMarkov process and the sojourn time for each system mode is deemed to be finite. A modedependent control scheme is employed with an adaptation sense in the presence of time delays in the mode switching of controller, which is manifested as a constant lag between the system mode and the controller mode. A novel form of Lyapunov function is adopted, in which the Lyapunov matrix depends on the modes of both the system and the controller as well as the time since the occurrence of the last mode switching. On the basis of the new proposed σerror meansquare stability that integrates the weights of all the system modes, numerically testable stability criteria are developed via the semiMarkov kernel approach. In virtue of certain techniques that can eliminate the terms containing powers or products of matrices, a desired modedependent stabilizing controller is designed such that the closedloop system is σerror meansquare stable by allowing a modeunmatched controller to perform before the controller switches to a modematched one. Finally, the theoretical results are applied to a practical example of one joint of a space robot manipulator to demonstrate the effectiveness, applicability and superiority of the proposed control strategy as well as the necessity of considering the modeswitching delays in the designed controller.
 Consensus conditions of continuoustime multiagent systems with
timedelays and measurement noises Abstract: Publication date: Available online 4 November 2018Source: AutomaticaAuthor(s): Xiaofeng Zong, Tao Li, JiFeng Zhang This work is concerned with stochastic consensus conditions of multiagent systems with both timedelays and measurement noises. For the case of additive noises, we develop some necessary conditions and sufficient conditions for stochastic weak consensus by estimating the differential resolvent function for delay equations. By the martingale convergence theorem, we obtain necessary conditions and sufficient conditions for stochastic strong consensus. For the case of multiplicative noises, we consider two kinds of timedelays, appeared in the measurement term and the noise term, respectively. We first show that stochastic weak consensus with the exponential convergence rate implies stochastic strong consensus. Then by constructing degenerate Lyapunov functional, we find the sufficient consensus conditions and show that stochastic consensus can be achieved by carefully choosing the control gain according to the noise intensities and the timedelay in the measurement term.
 Abstract: Publication date: Available online 4 November 2018Source: AutomaticaAuthor(s): Xiaofeng Zong, Tao Li, JiFeng Zhang This work is concerned with stochastic consensus conditions of multiagent systems with both timedelays and measurement noises. For the case of additive noises, we develop some necessary conditions and sufficient conditions for stochastic weak consensus by estimating the differential resolvent function for delay equations. By the martingale convergence theorem, we obtain necessary conditions and sufficient conditions for stochastic strong consensus. For the case of multiplicative noises, we consider two kinds of timedelays, appeared in the measurement term and the noise term, respectively. We first show that stochastic weak consensus with the exponential convergence rate implies stochastic strong consensus. Then by constructing degenerate Lyapunov functional, we find the sufficient consensus conditions and show that stochastic consensus can be achieved by carefully choosing the control gain according to the noise intensities and the timedelay in the measurement term.
 Eventtriggered distributed predictive control for asynchronous
coordination of multiagent systems Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Yuanyuan Zou, Xu Su, Shaoyuan Li, Yugang Niu, Dewei Li This paper investigates the eventtriggered distributed predictive control (DPC) problem for multiagent systems subject to bounded disturbances. A novel eventtriggering mechanism which involves the neighbours’ information is derived for each agent to achieve a tradeoff between resource usage and control performance. In such a framework, the DPC optimization problem is solved and information is exchanged only at triggering instants, thus achieving asynchronous coordination. To lower computation and communication consumption more significantly, a dynamic variable considering effects of neighbours is introduced to design a dynamic eventtriggering condition and we show that larger interexecution time can be obtained using the dynamic triggering mechanism. The theoretical conditions on ensuring feasibility and closedloop stability are developed for these two triggering mechanisms, respectively. Finally, numerical simulations are given to illustrate the effectiveness of the proposed control strategy.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Yuanyuan Zou, Xu Su, Shaoyuan Li, Yugang Niu, Dewei Li This paper investigates the eventtriggered distributed predictive control (DPC) problem for multiagent systems subject to bounded disturbances. A novel eventtriggering mechanism which involves the neighbours’ information is derived for each agent to achieve a tradeoff between resource usage and control performance. In such a framework, the DPC optimization problem is solved and information is exchanged only at triggering instants, thus achieving asynchronous coordination. To lower computation and communication consumption more significantly, a dynamic variable considering effects of neighbours is introduced to design a dynamic eventtriggering condition and we show that larger interexecution time can be obtained using the dynamic triggering mechanism. The theoretical conditions on ensuring feasibility and closedloop stability are developed for these two triggering mechanisms, respectively. Finally, numerical simulations are given to illustrate the effectiveness of the proposed control strategy.
 Coordinated trajectory tracking of multiple vertical takeoff and landing
UAVs Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Yao Zou, Ziyang Meng This paper investigates the coordinated trajectory tracking problem of multiple vertical takeoff and landing (VTOL) unmanned aerial vehicles (UAVs). The case of unidirectional information flow is considered and the objective is to drive all the follower VTOL UAVs to accurately track a desired trajectory associated with a leader. Firstly, a novel distributed estimator is developed for each VTOL UAV to obtain the leader’s desired information asymptotically. With the outputs of the estimators, the solution to the coordinated trajectory tracking problem of multiple VTOL UAVs is transformed to individually solving the tracking problem of each VTOL UAV. Due to the underactuated nature of the VTOL UAV, a hierarchical framework is introduced for each VTOL UAV such that a command force and an applied torque are exploited in sequence, then the position tracking to the estimated desired position and the attitude tracking to the command attitude are achieved. Moreover, an auxiliary system with proper parameters is implemented to guarantee the singularityfree command attitude extraction and to obviate the use of the unavailable desired information. The stability analysis and simulations effectively validate the achievement of the coordinated trajectory tracking of multiple VTOL UAVs with the proposed control approach.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Yao Zou, Ziyang Meng This paper investigates the coordinated trajectory tracking problem of multiple vertical takeoff and landing (VTOL) unmanned aerial vehicles (UAVs). The case of unidirectional information flow is considered and the objective is to drive all the follower VTOL UAVs to accurately track a desired trajectory associated with a leader. Firstly, a novel distributed estimator is developed for each VTOL UAV to obtain the leader’s desired information asymptotically. With the outputs of the estimators, the solution to the coordinated trajectory tracking problem of multiple VTOL UAVs is transformed to individually solving the tracking problem of each VTOL UAV. Due to the underactuated nature of the VTOL UAV, a hierarchical framework is introduced for each VTOL UAV such that a command force and an applied torque are exploited in sequence, then the position tracking to the estimated desired position and the attitude tracking to the command attitude are achieved. Moreover, an auxiliary system with proper parameters is implemented to guarantee the singularityfree command attitude extraction and to obviate the use of the unavailable desired information. The stability analysis and simulations effectively validate the achievement of the coordinated trajectory tracking of multiple VTOL UAVs with the proposed control approach.
 Adaptive stabilization of switched affine systems with unknown equilibrium
points: Application to power converters Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Gaëtan Beneux, Pierre Riedinger, Jamal Daafouz, Louis Grimaud The paper addresses the problem of designing a stabilizing control for switched affine systems with unknown parameters. We formulate the problem both in the case where the set of affine subsystems is finite and also in the case where the set of affine subsystems is not finite and given by a convex polytope, i.e., the convex hull of finitely many affine subsystems. The main contribution is a switched and adaptive control design methodology with a global asymptotic stability property. The difficulty is related to the fact that the equilibrium point is unknown a priori. We propose an observerbased control strategy that uses a parameter estimate to update the control law in real time. A DC/DC Flyback converter is considered to illustrate the effectiveness of the proposed method. We also show that the proposed strategy preserves the stability property when the Flyback converter works in the socalled discontinuous conduction mode (DCM).
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Gaëtan Beneux, Pierre Riedinger, Jamal Daafouz, Louis Grimaud The paper addresses the problem of designing a stabilizing control for switched affine systems with unknown parameters. We formulate the problem both in the case where the set of affine subsystems is finite and also in the case where the set of affine subsystems is not finite and given by a convex polytope, i.e., the convex hull of finitely many affine subsystems. The main contribution is a switched and adaptive control design methodology with a global asymptotic stability property. The difficulty is related to the fact that the equilibrium point is unknown a priori. We propose an observerbased control strategy that uses a parameter estimate to update the control law in real time. A DC/DC Flyback converter is considered to illustrate the effectiveness of the proposed method. We also show that the proposed strategy preserves the stability property when the Flyback converter works in the socalled discontinuous conduction mode (DCM).
 Multistage discrete time and randomized dynamic average consensus
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Mauro Franceschelli, Andrea Gasparri In this paper we propose a novel local interaction protocol which solves the discrete time dynamic average consensus problem, i.e., the consensus problem on the average value of a set of timevarying input signals in an undirected graph. The proposed interaction protocol is based on a multistage cascade of dynamic consensus filters which tracks the average value of the inputs with small and bounded error. We characterize its convergence properties for timevarying discretetime inputs with bounded variations. The main novelty of the proposed algorithm is that, with respect to other dynamic average consensus protocols, we obtain the next unique set of advantages: i) The protocol, inspired by proportional dynamic consensus, does not exploit integral control actions or input derivatives, thus exhibits robustness to reinitialization errors, changes in the network size and noise in the input signals; ii) The proposed design allows to tradeoff the quantity of information locally exchanged by the agents, i.e., the number of stages, with steadystate error, tracking error and convergence time; iii) The protocol can be implemented with randomized and asynchronous local state updates and keep in expectation the performance of the discretetime version. Numerical examples are given to corroborate the theoretical findings, including the case where a new agent joins and leaves the network during the algorithm execution to show robustness to reinitialization errors during runtime.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Mauro Franceschelli, Andrea Gasparri In this paper we propose a novel local interaction protocol which solves the discrete time dynamic average consensus problem, i.e., the consensus problem on the average value of a set of timevarying input signals in an undirected graph. The proposed interaction protocol is based on a multistage cascade of dynamic consensus filters which tracks the average value of the inputs with small and bounded error. We characterize its convergence properties for timevarying discretetime inputs with bounded variations. The main novelty of the proposed algorithm is that, with respect to other dynamic average consensus protocols, we obtain the next unique set of advantages: i) The protocol, inspired by proportional dynamic consensus, does not exploit integral control actions or input derivatives, thus exhibits robustness to reinitialization errors, changes in the network size and noise in the input signals; ii) The proposed design allows to tradeoff the quantity of information locally exchanged by the agents, i.e., the number of stages, with steadystate error, tracking error and convergence time; iii) The protocol can be implemented with randomized and asynchronous local state updates and keep in expectation the performance of the discretetime version. Numerical examples are given to corroborate the theoretical findings, including the case where a new agent joins and leaves the network during the algorithm execution to show robustness to reinitialization errors during runtime.
 An exponential quantum projection filter for open quantum systems
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Qing Gao, Guofeng Zhang, Ian R. Petersen An approximate exponential quantum projection filtering scheme is developed for a class of open quantum systems described by Hudson–Parthasarathy quantum stochastic differential equations, aiming to reduce the computational burden associated with online calculation of the quantum filter. By using a differential geometric approach, the quantum trajectory is constrained in a finitedimensional differentiable manifold consisting of an unnormalized exponential family of quantum density operators, and an exponential quantum projection filter is then formulated as a number of stochastic differential equations satisfied by the finitedimensional coordinate system of this manifold. A convenient design of the differentiable manifold is also presented through reduction of the local approximation errors, which yields a simplification of the quantum projection filter equations. It is shown that the computational cost can be significantly reduced by using the quantum projection filter instead of the quantum filter. It is also shown that when the quantum projection filtering approach is applied to a class of open quantum systems that asymptotically converge to a pure state, the inputtostate stability of the corresponding exponential quantum projection filter can be established. Simulation results from an atomic ensemble system example are provided to illustrate the performance of the projection filtering scheme. It is expected that the proposed approach can be used in developing more efficient quantum control methods.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Qing Gao, Guofeng Zhang, Ian R. Petersen An approximate exponential quantum projection filtering scheme is developed for a class of open quantum systems described by Hudson–Parthasarathy quantum stochastic differential equations, aiming to reduce the computational burden associated with online calculation of the quantum filter. By using a differential geometric approach, the quantum trajectory is constrained in a finitedimensional differentiable manifold consisting of an unnormalized exponential family of quantum density operators, and an exponential quantum projection filter is then formulated as a number of stochastic differential equations satisfied by the finitedimensional coordinate system of this manifold. A convenient design of the differentiable manifold is also presented through reduction of the local approximation errors, which yields a simplification of the quantum projection filter equations. It is shown that the computational cost can be significantly reduced by using the quantum projection filter instead of the quantum filter. It is also shown that when the quantum projection filtering approach is applied to a class of open quantum systems that asymptotically converge to a pure state, the inputtostate stability of the corresponding exponential quantum projection filter can be established. Simulation results from an atomic ensemble system example are provided to illustrate the performance of the projection filtering scheme. It is expected that the proposed approach can be used in developing more efficient quantum control methods.
 Hybrid mechanisms for robust synchronization and coordination of
multiagent networked sampleddata systems Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Jorge I. Poveda, Andrew R. Teel We present a novel approach for the design of robust feedback coordination and control mechanisms for networks of asynchronous nonlinear multiagent systems (MAS). Each agent corresponds to a sampleddata system characterized by a continuoustime plant, a discretetime controller with logic or integer states, and a sampler/zeroorder hold with a local clock. The goal is to robustly stabilize an applicationdependent compact set defined a priori for the MAS, taking into account the asynchronous nature of the triggering mechanisms of the agents, and the limited information in the network. To solve this problem, we propose an emulationlike approach, where the feedback mechanism for each agent is initially designed for a nominal ideal synchronous MAS with a single logic state. Unlike existing emulation results for networked sampleddata systems with a single triggering mechanism, the implementation of multiple triggering mechanisms in MAS requires additional decentralized coordination algorithms to guarantee that the implemented system robustly emulates the behavior of the nominal synchronous system. Therefore, we propose a decentralized synchronization and coordination mechanism that controls the triggering mechanisms of the agents, guaranteeing robust stabilization of the closedloop MAS. Our results are established by using Lyapunov tools, the invariance principle, and robustness corollaries for setvalued hybrid dynamical systems.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Jorge I. Poveda, Andrew R. Teel We present a novel approach for the design of robust feedback coordination and control mechanisms for networks of asynchronous nonlinear multiagent systems (MAS). Each agent corresponds to a sampleddata system characterized by a continuoustime plant, a discretetime controller with logic or integer states, and a sampler/zeroorder hold with a local clock. The goal is to robustly stabilize an applicationdependent compact set defined a priori for the MAS, taking into account the asynchronous nature of the triggering mechanisms of the agents, and the limited information in the network. To solve this problem, we propose an emulationlike approach, where the feedback mechanism for each agent is initially designed for a nominal ideal synchronous MAS with a single logic state. Unlike existing emulation results for networked sampleddata systems with a single triggering mechanism, the implementation of multiple triggering mechanisms in MAS requires additional decentralized coordination algorithms to guarantee that the implemented system robustly emulates the behavior of the nominal synchronous system. Therefore, we propose a decentralized synchronization and coordination mechanism that controls the triggering mechanisms of the agents, guaranteeing robust stabilization of the closedloop MAS. Our results are established by using Lyapunov tools, the invariance principle, and robustness corollaries for setvalued hybrid dynamical systems.
 Subspace identification with moment matching
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Masaki Inoue We propose a deterministic identification method that involves a priori information characterized as moments of a transfer function. The moments are expressed in terms of the solution to a Sylvester matrix equation. The Sylvester equation is incorporated with a conventional subspace identification method, and a problem for momentconstrained subspace identification is formulated. Since the identification problem is in a class of nonlinear optimization problems, it cannot be efficiently solved in numerical computation. Application of a changeofvariable technique reduces the problem to least squares optimization, and the solution provides a statespace model that involves the prespecified moments. Finally, the effectiveness of the proposed method is illustrated in numerical simulations.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Masaki Inoue We propose a deterministic identification method that involves a priori information characterized as moments of a transfer function. The moments are expressed in terms of the solution to a Sylvester matrix equation. The Sylvester equation is incorporated with a conventional subspace identification method, and a problem for momentconstrained subspace identification is formulated. Since the identification problem is in a class of nonlinear optimization problems, it cannot be efficiently solved in numerical computation. Application of a changeofvariable technique reduces the problem to least squares optimization, and the solution provides a statespace model that involves the prespecified moments. Finally, the effectiveness of the proposed method is illustrated in numerical simulations.
 Timeoptimal handsoff control for linear timeinvariant systems
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Takuya Ikeda, Masaaki Nagahara In this article, we investigate theoretical properties of the timeoptimal handsoff control for linear timeinvariant systems. The purpose of the control is to maximize the time duration on which the control value is exactly zero (maximum handsoff control) and also to minimize the response time to achieve a given state transition (timeoptimal control). For this, we introduce a cost function described by a linear combination of the L0 measure and the response time of the control. Since the L0 measure is nonconvex and discontinuous, we adopt the L1 relaxation technique for the analysis of the optimal control. By using this relaxation, we show the existence of the timeoptimal handsoff control, and the equivalence between L0 andL1 optimal controls under the normality assumption.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Takuya Ikeda, Masaaki Nagahara In this article, we investigate theoretical properties of the timeoptimal handsoff control for linear timeinvariant systems. The purpose of the control is to maximize the time duration on which the control value is exactly zero (maximum handsoff control) and also to minimize the response time to achieve a given state transition (timeoptimal control). For this, we introduce a cost function described by a linear combination of the L0 measure and the response time of the control. Since the L0 measure is nonconvex and discontinuous, we adopt the L1 relaxation technique for the analysis of the optimal control. By using this relaxation, we show the existence of the timeoptimal handsoff control, and the equivalence between L0 andL1 optimal controls under the normality assumption.
 Comments on “Adaptive tracking control of uncertain MIMO nonlinear
systems with input constraints” Abstract: Publication date: Available online 31 October 2018Source: AutomaticaAuthor(s): AnMin Zou, Anton H.J. de Ruiter, Krishna Dev Kumar
 Abstract: Publication date: Available online 31 October 2018Source: AutomaticaAuthor(s): AnMin Zou, Anton H.J. de Ruiter, Krishna Dev Kumar
 Stabilization of a class of slow–fast control systems at
nonhyperbolic points Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Hildeberto JardónKojakhmetov, Jacquelien M.A. Scherpen, Dunstano del PuertoFlores In this document, we deal with the local asymptotic stabilization problem of a class of slow–fastsystems (or singularly perturbed Ordinary Differential Equations). The systems studied here have the following properties: (1) they have one fast and an arbitrary number of slow variables, and (2) they have a nonhyperbolic singularity at the origin of arbitrary degeneracy. Our goal is to stabilize such a point. The presence of the aforementioned singularity complicates the analysis and the controller design. In particular, the classical theory of singular perturbations cannot be used. We propose a novel design based on geometric desingularization, which allows the stabilization of a nonhyperbolic point of singularly perturbed control systems. Our results are exemplified on a didactic example and on an electric circuit.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Hildeberto JardónKojakhmetov, Jacquelien M.A. Scherpen, Dunstano del PuertoFlores In this document, we deal with the local asymptotic stabilization problem of a class of slow–fastsystems (or singularly perturbed Ordinary Differential Equations). The systems studied here have the following properties: (1) they have one fast and an arbitrary number of slow variables, and (2) they have a nonhyperbolic singularity at the origin of arbitrary degeneracy. Our goal is to stabilize such a point. The presence of the aforementioned singularity complicates the analysis and the controller design. In particular, the classical theory of singular perturbations cannot be used. We propose a novel design based on geometric desingularization, which allows the stabilization of a nonhyperbolic point of singularly perturbed control systems. Our results are exemplified on a didactic example and on an electric circuit.
 Stochastic learning in multiagent optimization: Communication and
payoffbased approaches Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Tatiana Tatarenko Game theory serves as a powerful tool for distributed optimization in multiagent systems in different applications. In this paper we consider multiagent systems that can be modeled by means of potential games whose potential function coincides with a global objective function to be maximized. In this approach, the agents correspond to the strategic decision makers and the optimization problem is equivalent to the problem of learning a potential function maximizer in the designed game. The paper deals with two different information settings in the system. Firstly, we consider systems, where agents have the access to the gradient of their utility functions. However, they do not possess the full information about the joint actions. Thus, to be able to move along the gradient toward a local optimum, they need to exchange the information with their neighbors by means of communication. The second setting refers to a payoffbased approach. Here, we assume that at each iteration agents can only observe their own played actions and experienced payoffs. In both cases, the paper studies unconstrained nonconcave optimization with a differentiable objective function. To develop the corresponding algorithms guaranteeing convergence to a local maximum of the potential function in absence of saddle points, we utilize the idea of the wellknown Robbins–Monro procedure based on the theory of stochastic approximation.
 Abstract: Publication date: January 2019Source: Automatica, Volume 99Author(s): Tatiana Tatarenko Game theory serves as a powerful tool for distributed optimization in multiagent systems in different applications. In this paper we consider multiagent systems that can be modeled by means of potential games whose potential function coincides with a global objective function to be maximized. In this approach, the agents correspond to the strategic decision makers and the optimization problem is equivalent to the problem of learning a potential function maximizer in the designed game. The paper deals with two different information settings in the system. Firstly, we consider systems, where agents have the access to the gradient of their utility functions. However, they do not possess the full information about the joint actions. Thus, to be able to move along the gradient toward a local optimum, they need to exchange the information with their neighbors by means of communication. The second setting refers to a payoffbased approach. Here, we assume that at each iteration agents can only observe their own played actions and experienced payoffs. In both cases, the paper studies unconstrained nonconcave optimization with a differentiable objective function. To develop the corresponding algorithms guaranteeing convergence to a local maximum of the potential function in absence of saddle points, we utilize the idea of the wellknown Robbins–Monro procedure based on the theory of stochastic approximation.
 On the construction of safe controllable regions for affine systems with
applications to robotics Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Mohamed K. Helwa, Angela P. Schoellig This paper studies the problem of constructing inblock controllable (IBC) regions for affine systems. That is, we are concerned with constructing regions in the state space of affine systems such that all the states in the interior of the region are mutually accessible within the region’s interior by applying uniformly bounded inputs. We first show that existing results for checking inblock controllability on given polytopic regions cannot be easily extended to address the question of constructing IBC regions. We then explore the geometry of the problem to provide a computationally efficient algorithm for constructing IBC regions. We also prove the soundness of the algorithm. We then use the proposed algorithm to construct safe speed profiles for robotic systems. As a case study, we present several experimental results on unmanned aerial vehicles (UAVs) to verify the effectiveness of the proposed algorithm; these results include using the proposed algorithm for realtime collision avoidance for UAVs.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Mohamed K. Helwa, Angela P. Schoellig This paper studies the problem of constructing inblock controllable (IBC) regions for affine systems. That is, we are concerned with constructing regions in the state space of affine systems such that all the states in the interior of the region are mutually accessible within the region’s interior by applying uniformly bounded inputs. We first show that existing results for checking inblock controllability on given polytopic regions cannot be easily extended to address the question of constructing IBC regions. We then explore the geometry of the problem to provide a computationally efficient algorithm for constructing IBC regions. We also prove the soundness of the algorithm. We then use the proposed algorithm to construct safe speed profiles for robotic systems. As a case study, we present several experimental results on unmanned aerial vehicles (UAVs) to verify the effectiveness of the proposed algorithm; these results include using the proposed algorithm for realtime collision avoidance for UAVs.

H ∞
average dwell time approach Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Rongni Yang, Wei Xing Zheng This paper studies the H∞ filtering problem for a class of discretetime twodimensional (2D) switched systems. The 2D systems under consideration are described by the wellknown Fornasini–Marchesini local statespace model. Our attention is focused on designing a fullorder filter such that the filtering error system is guaranteed to be asymptotically or exponentially stable with a prescribed H∞ disturbance attenuation level. Based on the switched quadratic Lyapunov function and the piecewise Lyapunov function approach, sufficient conditions are established to ensure the existence of the desired filters for such systems under the arbitrary switching and restricted switching signals, respectively. Furthermore, the notion of time instant for 2D switched systems is introduced and the extended average dwell time technique is utilized under the restricted switching signal. Then the corresponding filter is designed by dealing with a convex optimization problem that can be efficiently solved via standard numerical algorithms. Finally, two numerical examples are provided to demonstrate the effectiveness of the developed filter design algorithms.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Rongni Yang, Wei Xing Zheng This paper studies the H∞ filtering problem for a class of discretetime twodimensional (2D) switched systems. The 2D systems under consideration are described by the wellknown Fornasini–Marchesini local statespace model. Our attention is focused on designing a fullorder filter such that the filtering error system is guaranteed to be asymptotically or exponentially stable with a prescribed H∞ disturbance attenuation level. Based on the switched quadratic Lyapunov function and the piecewise Lyapunov function approach, sufficient conditions are established to ensure the existence of the desired filters for such systems under the arbitrary switching and restricted switching signals, respectively. Furthermore, the notion of time instant for 2D switched systems is introduced and the extended average dwell time technique is utilized under the restricted switching signal. Then the corresponding filter is designed by dealing with a convex optimization problem that can be efficiently solved via standard numerical algorithms. Finally, two numerical examples are provided to demonstrate the effectiveness of the developed filter design algorithms.
 Tracking control of uncertain nonlinear systems with deferred asymmetric
timevarying full state constraints Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): YongDuan Song, Shuyan Zhou In this paper, we investigate the tracking control problem of uncertain strictfeedback systems under deferred and asymmetric yet timevarying (DATV) constraints. We show that such type of constraints, occurring some time after (rather than from the beginning of) system operation, are frequently encountered in practice that have not been adequately addressed in existing works. By utilizing an errorshifting transformation, together with a new asymmetric Barrier Lyapunov Function with variational barrier bounds, we develop a tracking control method capable of dealing with DATV full state constraints under completely unknown initial tracking condition, leading to a control solution to the underlying problem. We also show that, with the proposed method, full state constraints being violated initially (rendering the previous methods inapplicable) can be made satisfied within a prespecified finite time. The benefits and effectiveness of the proposed control are theoretically authenticated and numerically validated.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): YongDuan Song, Shuyan Zhou In this paper, we investigate the tracking control problem of uncertain strictfeedback systems under deferred and asymmetric yet timevarying (DATV) constraints. We show that such type of constraints, occurring some time after (rather than from the beginning of) system operation, are frequently encountered in practice that have not been adequately addressed in existing works. By utilizing an errorshifting transformation, together with a new asymmetric Barrier Lyapunov Function with variational barrier bounds, we develop a tracking control method capable of dealing with DATV full state constraints under completely unknown initial tracking condition, leading to a control solution to the underlying problem. We also show that, with the proposed method, full state constraints being violated initially (rendering the previous methods inapplicable) can be made satisfied within a prespecified finite time. The benefits and effectiveness of the proposed control are theoretically authenticated and numerically validated.
 Permanent magnet synchronous motors are globally asymptotically
stabilizable with PI current control Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Romeo Ortega, Nima Monshizadeh, Pooya Monshizadeh, Dmitry Bazylev, Anton Pyrkin This note shows that the industry standard desired equilibrium for permanent magnet synchronous motors (i.e, maximum torque per Ampere) can be globally asymptotically stabilized with a PI control around the current errors, provided some viscous friction (possibly small) is present in the rotor dynamics and the proportional gain of the PI is suitably chosen. Instrumental to establish this surprising result is the proof that the map from voltages to currents of the incremental model of the motor satisfies some passivity properties. The analysis relies on basic Lyapunov theory making the result available to a wide audience.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Romeo Ortega, Nima Monshizadeh, Pooya Monshizadeh, Dmitry Bazylev, Anton Pyrkin This note shows that the industry standard desired equilibrium for permanent magnet synchronous motors (i.e, maximum torque per Ampere) can be globally asymptotically stabilized with a PI control around the current errors, provided some viscous friction (possibly small) is present in the rotor dynamics and the proportional gain of the PI is suitably chosen. Instrumental to establish this surprising result is the proof that the map from voltages to currents of the incremental model of the motor satisfies some passivity properties. The analysis relies on basic Lyapunov theory making the result available to a wide audience.
 Adaptive output feedback control of stochastic nonholonomic systems with
nonlinear parameterization Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Hui Wang, Quanxin Zhu The problem of adaptive outputfeedback control of nonlinearly parameterized stochastic nonholonomic systems is studied in this paper. Since many unknowns (e.g., unknown control coefficients and unknown nonlinear parameters in systems’ nonlinearities) occur into systems, we utilize an adaptive control method, together with a parameter separation technique, to construct an adaptive output feedback controller to regulate the whole systems. During the design procedure, a new form of reducedorder Kfilters is given to compensate the unmeasured states of the systems. A switching strategy is proposed explicitly to stabilize the entire systems in the control scheme. Finally, a bilinear model with stochastic disturbances is presented to demonstrate our theoretical results.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Hui Wang, Quanxin Zhu The problem of adaptive outputfeedback control of nonlinearly parameterized stochastic nonholonomic systems is studied in this paper. Since many unknowns (e.g., unknown control coefficients and unknown nonlinear parameters in systems’ nonlinearities) occur into systems, we utilize an adaptive control method, together with a parameter separation technique, to construct an adaptive output feedback controller to regulate the whole systems. During the design procedure, a new form of reducedorder Kfilters is given to compensate the unmeasured states of the systems. A switching strategy is proposed explicitly to stabilize the entire systems in the control scheme. Finally, a bilinear model with stochastic disturbances is presented to demonstrate our theoretical results.
 Finitetime stabilization of weak solutions for a class of nonlocal
Lipschitzian stochastic nonlinear systems with inverse dynamics Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): GuiHua Zhao, JianChao Li, ShuJun Liu In this paper, finitetime stabilization is investigated for a class of nonlocal Lipschitzian stochastic nonlinear systems with stochastic inverse dynamics. Different from the existing work about finitetime control, to guarantee the existence of the solution under mild conditions, we study the stabilization in the sense of weak solution. We first present a finitetime stability theory under the framework of weak solution. Then, for a class of stochastic nonlinear systems with stochastic inverse dynamics, a finitetime controller via state feedback is constructively designed under the assumption that the stochastic inverse dynamics is stochastic inputtostate stable. The trivial weak solution of the closedloop system is proved to be globally finitetime stable in probability. Finally, a simulation example is given to illustrate the efficiency of the proposed design procedure.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): GuiHua Zhao, JianChao Li, ShuJun Liu In this paper, finitetime stabilization is investigated for a class of nonlocal Lipschitzian stochastic nonlinear systems with stochastic inverse dynamics. Different from the existing work about finitetime control, to guarantee the existence of the solution under mild conditions, we study the stabilization in the sense of weak solution. We first present a finitetime stability theory under the framework of weak solution. Then, for a class of stochastic nonlinear systems with stochastic inverse dynamics, a finitetime controller via state feedback is constructively designed under the assumption that the stochastic inverse dynamics is stochastic inputtostate stable. The trivial weak solution of the closedloop system is proved to be globally finitetime stable in probability. Finally, a simulation example is given to illustrate the efficiency of the proposed design procedure.
 Asymptotic adaptive control of nonlinear systems with elimination of
overparametrization in a Nussbaumlike design Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Ci Chen, Zhi Liu, Kan Xie, Yun Zhang, C.L. Philip Chen Making a tradeoff between the control accuracy and computational reduction is a problem frequently encountered in the system control design. This is especially difficult when one designs adaptive fuzzy (or neural network) controls for nonlinear systems, in which fuzzy controls have to consume many computational resources to tune a sufficiently large number of adaptive parameters, meanwhile nonlinear uncertainties block the high demanding control accuracy. Current works usually face a dilemma that, either the computation is reduced but the control accuracy is degraded due to the use of the norm estimation, or the asymptotic control is resulted but the computation is increased due to the extra compensation controls. To address such dilemma, we propose an asymptotic adaptive fuzzy tracking control algorithm, whose main feature is that only two adaptive laws are needed throughout the control scheme. In particular, we allocate one adaptive law to achieve adaptive fuzzy backstepping control for nonlinear systems with a focus on stabilizing the closedloop system. We then allocate the other adaptive law not only to asymptotically drive the stabilization error to the zero, but also to achieve the elimination of overparametrization in a Nussbaumlike design, which is inspired by the tuning functionbased approach.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Ci Chen, Zhi Liu, Kan Xie, Yun Zhang, C.L. Philip Chen Making a tradeoff between the control accuracy and computational reduction is a problem frequently encountered in the system control design. This is especially difficult when one designs adaptive fuzzy (or neural network) controls for nonlinear systems, in which fuzzy controls have to consume many computational resources to tune a sufficiently large number of adaptive parameters, meanwhile nonlinear uncertainties block the high demanding control accuracy. Current works usually face a dilemma that, either the computation is reduced but the control accuracy is degraded due to the use of the norm estimation, or the asymptotic control is resulted but the computation is increased due to the extra compensation controls. To address such dilemma, we propose an asymptotic adaptive fuzzy tracking control algorithm, whose main feature is that only two adaptive laws are needed throughout the control scheme. In particular, we allocate one adaptive law to achieve adaptive fuzzy backstepping control for nonlinear systems with a focus on stabilizing the closedloop system. We then allocate the other adaptive law not only to asymptotically drive the stabilization error to the zero, but also to achieve the elimination of overparametrization in a Nussbaumlike design, which is inspired by the tuning functionbased approach.
 An improved timedelay implementation of derivativedependent feedback
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Anton Selivanov, Emilia Fridman We consider an LTI system of relative degree r≥2 that can be stabilized using r−1 output derivatives. The derivatives are approximated by finite differences leading to a timedelayed feedback. We present a new method of designing and analyzing such feedback under continuoustime and sampled measurements. This method admits essentially larger timedelay/sampling period compared to the existing results and, for the first time, allows to use consecutively sampled measurements in the sampleddata case. The main idea is to present the difference between the derivative and its approximation in a convenient integral form. The kernel of this integral is hard to express explicitly but we show that it satisfies certain properties. These properties are employed to construct the Lyapunov–Krasovskii functional that leads to LMIbased stability conditions. If the derivativedependent control exponentially stabilizes the system, then its timedelayed approximation stabilizes the system with the same decay rate provided the timedelay (for continuoustime measurements) or the sampling period (for sampled measurements) are small enough.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Anton Selivanov, Emilia Fridman We consider an LTI system of relative degree r≥2 that can be stabilized using r−1 output derivatives. The derivatives are approximated by finite differences leading to a timedelayed feedback. We present a new method of designing and analyzing such feedback under continuoustime and sampled measurements. This method admits essentially larger timedelay/sampling period compared to the existing results and, for the first time, allows to use consecutively sampled measurements in the sampleddata case. The main idea is to present the difference between the derivative and its approximation in a convenient integral form. The kernel of this integral is hard to express explicitly but we show that it satisfies certain properties. These properties are employed to construct the Lyapunov–Krasovskii functional that leads to LMIbased stability conditions. If the derivativedependent control exponentially stabilizes the system, then its timedelayed approximation stabilizes the system with the same decay rate provided the timedelay (for continuoustime measurements) or the sampling period (for sampled measurements) are small enough.
 Adaptive asymptotic tracking using barrier functions
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): YongHua Liu, Hongyi Li This paper studies the global output tracking problem for a class of unknown timevarying nonlinear systems in strictfeedback form. By utilizing the barrier functions, a universal adaptive statefeedback control strategy is proposed that achieves asymptotic tracking performance. Unlike the existing results in the literature, the proposed control scheme utilizes the barrier functions to ensure the unknown system nonlinearities to be the bounded “disturbancelike” terms, which are adaptively compensated at each step, this enables any approximation structures are not needed. Furthermore, the “explosion of complexity” issue in backsteppinglike approaches is avoided without using additional filtering. Simulation results are presented to demonstrate the effectiveness of the proposed methodology.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): YongHua Liu, Hongyi Li This paper studies the global output tracking problem for a class of unknown timevarying nonlinear systems in strictfeedback form. By utilizing the barrier functions, a universal adaptive statefeedback control strategy is proposed that achieves asymptotic tracking performance. Unlike the existing results in the literature, the proposed control scheme utilizes the barrier functions to ensure the unknown system nonlinearities to be the bounded “disturbancelike” terms, which are adaptively compensated at each step, this enables any approximation structures are not needed. Furthermore, the “explosion of complexity” issue in backsteppinglike approaches is avoided without using additional filtering. Simulation results are presented to demonstrate the effectiveness of the proposed methodology.
 Frequency domain identification of continuoustime outputerror models
with timedelay from relay feedback tests Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Fengwei Chen, Hugues Garnier, Marion Gilson, Xiangtao Zhuan This paper is concerned with identification of continuoustime outputerror models with timedelay from relay feedback tests. Conventional methods for solving this problem consist in deriving analytical limit cycle expressions and fitting them to measured shape factors. However, they may fail to handle different limit cycles uniformly, due to the structural differences in the analytical expressions. To overcome this problem, we consider a more general, databased, parametric identification framework using sampled limit cycle data. A frequency domain method that minimizes the sum of squared outputerrors is developed. The proposed method can be of high accuracy, thanks to the periodic input–output signals provided by sustained relay feedback oscillations, which can help to reduce leakage and aliasing errors. Besides, a distinctive merit of the proposed method is that identification of stable and unstable plants can be equally simple. The effectiveness and superiority of the proposed method are demonstrated by means of both theoretical analyses and simulation examples.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Fengwei Chen, Hugues Garnier, Marion Gilson, Xiangtao Zhuan This paper is concerned with identification of continuoustime outputerror models with timedelay from relay feedback tests. Conventional methods for solving this problem consist in deriving analytical limit cycle expressions and fitting them to measured shape factors. However, they may fail to handle different limit cycles uniformly, due to the structural differences in the analytical expressions. To overcome this problem, we consider a more general, databased, parametric identification framework using sampled limit cycle data. A frequency domain method that minimizes the sum of squared outputerrors is developed. The proposed method can be of high accuracy, thanks to the periodic input–output signals provided by sustained relay feedback oscillations, which can help to reduce leakage and aliasing errors. Besides, a distinctive merit of the proposed method is that identification of stable and unstable plants can be equally simple. The effectiveness and superiority of the proposed method are demonstrated by means of both theoretical analyses and simulation examples.
 Prediction error identification of linear dynamic networks with
rankreduced noise Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Harm H.M. Weerts, Paul M.J. Van den Hof, Arne G. Dankers Dynamic networks are interconnected dynamic systems with measured node signals and dynamic modules reflecting the links between the nodes. We address the problem of identifying a dynamic network with known topology, on the basis of measured signals, for the situation of additive process noise on the node signals that is spatially correlated and that is allowed to have a spectral density that is singular. A prediction error approach is followed in which all node signals in the network are jointly predicted. The resulting jointdirect identification method, generalizes the classical direct method for closedloop identification to handle situations of mutually correlated noise on inputs and outputs. When applied to general dynamic networks with rankreduced noise, it appears that the natural identification criterion becomes a weighted LS criterion that is subject to a constraint. This constrained criterion is shown to lead to maximum likelihood estimates of the dynamic network and therefore to minimum variance properties, reaching the Cramér–Rao lower bound in the case of Gaussian noise. In order to reduce technical complexity, the analysis is restricted to dynamic networks with strictly proper modules.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Harm H.M. Weerts, Paul M.J. Van den Hof, Arne G. Dankers Dynamic networks are interconnected dynamic systems with measured node signals and dynamic modules reflecting the links between the nodes. We address the problem of identifying a dynamic network with known topology, on the basis of measured signals, for the situation of additive process noise on the node signals that is spatially correlated and that is allowed to have a spectral density that is singular. A prediction error approach is followed in which all node signals in the network are jointly predicted. The resulting jointdirect identification method, generalizes the classical direct method for closedloop identification to handle situations of mutually correlated noise on inputs and outputs. When applied to general dynamic networks with rankreduced noise, it appears that the natural identification criterion becomes a weighted LS criterion that is subject to a constraint. This constrained criterion is shown to lead to maximum likelihood estimates of the dynamic network and therefore to minimum variance properties, reaching the Cramér–Rao lower bound in the case of Gaussian noise. In order to reduce technical complexity, the analysis is restricted to dynamic networks with strictly proper modules.
 Balancing and suppression of oscillations of tension and cage in
dualcable mining elevators Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Ji Wang, Yangjun Pi, Miroslav Krstic Dualcable mining elevator has advantages in the transportation of heavy load to a large depth over the single cable elevator. However challenges occur when lifting a cage via two parallel compliant cables, such as tension oscillation inconformity between two cables and the cage roll, which are important physical variables relating to the fatigue fracture of mining cables. Mining elevator vibration dynamics are modeled by two pairs of 2 × 2 heterodirectional coupled hyperbolic PDEs on a timevarying domain and all four PDE bottom boundaries are coupled at one ODE. We design an output feedback boundary control law via backstepping to exponentially stabilize the dynamic system including the tension oscillation states, tension oscillation error states and the cage roll states. The control law is constructed with the estimated states from the observer formed by available boundary measurements. The exponential stability of the closedloop system is proved via Lyapunov analysis. Effective suppression of tension oscillations, reduction of inconformity between tension oscillations in two cables, and balancing the cage roll under the proposed controller are verified via numerical simulation.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Ji Wang, Yangjun Pi, Miroslav Krstic Dualcable mining elevator has advantages in the transportation of heavy load to a large depth over the single cable elevator. However challenges occur when lifting a cage via two parallel compliant cables, such as tension oscillation inconformity between two cables and the cage roll, which are important physical variables relating to the fatigue fracture of mining cables. Mining elevator vibration dynamics are modeled by two pairs of 2 × 2 heterodirectional coupled hyperbolic PDEs on a timevarying domain and all four PDE bottom boundaries are coupled at one ODE. We design an output feedback boundary control law via backstepping to exponentially stabilize the dynamic system including the tension oscillation states, tension oscillation error states and the cage roll states. The control law is constructed with the estimated states from the observer formed by available boundary measurements. The exponential stability of the closedloop system is proved via Lyapunov analysis. Effective suppression of tension oscillations, reduction of inconformity between tension oscillations in two cables, and balancing the cage roll under the proposed controller are verified via numerical simulation.
 Cooperative and mobile manipulation of multiple microscopic objects based
on microhands and laserstage control Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Quang Minh Ta, Chien Chern Cheah While various techniques have been developed for manipulation of biological cells or microobjects using optical tweezers, the performance and feasibility of these techniques are mostly dependent on the physical properties of the target objects to be manipulated. In these existing techniques, direct trapping and manipulation of the manipulated objects using laser traps are performed, and therefore, existing techniques for optical manipulation are not capable of coordinating and manipulating various types of objects in the microworld, including untrappable microobjects, relatively large microobjects, and laser sensitive biological cells. In this paper, a cooperative control technique is proposed for coordinative and mobile manipulation of multiple microscopic objects using microhands with multiple laserdriven fingertips and robotassisted stage control. Several virtual microhands are formed by coordinating multiple optically trapped microparticles that serve as the laserdriven fingertips, and then utilized for individual and coordinative manipulation of the target microobjects. Simultaneously, global transportation of all the grasped target objects is performed by controlling the robotassisted stage. While it is difficult to design multifingered hands in microscale due to scaling effect, this paper presents the first result on cooperative and mobile manipulation of multiple microobjects using multiple microhands with laserdriven fingertips and robotassisted stage control. In this paper, a primary study on repositioning strategy of the laserdriven fingertips is also introduced to allow the fingertips in a grasping formation to be repositioned. Rigorous mathematical formulations and solutions are derived to achieve the control objective, and experimental results are presented to demonstrate the effectiveness of the proposed control technique.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Quang Minh Ta, Chien Chern Cheah While various techniques have been developed for manipulation of biological cells or microobjects using optical tweezers, the performance and feasibility of these techniques are mostly dependent on the physical properties of the target objects to be manipulated. In these existing techniques, direct trapping and manipulation of the manipulated objects using laser traps are performed, and therefore, existing techniques for optical manipulation are not capable of coordinating and manipulating various types of objects in the microworld, including untrappable microobjects, relatively large microobjects, and laser sensitive biological cells. In this paper, a cooperative control technique is proposed for coordinative and mobile manipulation of multiple microscopic objects using microhands with multiple laserdriven fingertips and robotassisted stage control. Several virtual microhands are formed by coordinating multiple optically trapped microparticles that serve as the laserdriven fingertips, and then utilized for individual and coordinative manipulation of the target microobjects. Simultaneously, global transportation of all the grasped target objects is performed by controlling the robotassisted stage. While it is difficult to design multifingered hands in microscale due to scaling effect, this paper presents the first result on cooperative and mobile manipulation of multiple microobjects using multiple microhands with laserdriven fingertips and robotassisted stage control. In this paper, a primary study on repositioning strategy of the laserdriven fingertips is also introduced to allow the fingertips in a grasping formation to be repositioned. Rigorous mathematical formulations and solutions are derived to achieve the control objective, and experimental results are presented to demonstrate the effectiveness of the proposed control technique.
 Decentralized control scheme for largescale systems defined over a graph
in presence of communication delays and random missing measurements Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Yan Wang, Junlin Xiong, Daniel W.C. Ho This paper studies the decentralized outputfeedback control of largescale systems defined over a directed connected graph with communication delay and random missing measurements. The nodes in the graph represent the subsystems, and the edges represent the communication connection. The information travels across an edge in the graph and suffers from one step communication delay. For saving the storage space, the information delayed more than D step times is discarded. In addition, to model the system in a more practical case, we assume that the observation for the subsystem output suffers random missing. Under this new information pattern, the optimal outputfeedback control problem is nonconvex, what is worse, the separation principle fails. This implies that the optimal control problem with the information pattern introduced above is difficult to solve. In this paper, a new decentralized control scheme is proposed. In particular, a new estimator structure and a new controller structure are constructed, and the gains of the estimator and the controller are designed simultaneously. An optimality condition with respect to the gains is established. Based on the optimality condition, an iterative algorithm is exploited to design the gains numerically. It is shown that the exploited algorithm converges to Nash optimum. Finally, the proposed theoretical results are illustrated by a physical system which is a heavy duty vehicles platoon.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Yan Wang, Junlin Xiong, Daniel W.C. Ho This paper studies the decentralized outputfeedback control of largescale systems defined over a directed connected graph with communication delay and random missing measurements. The nodes in the graph represent the subsystems, and the edges represent the communication connection. The information travels across an edge in the graph and suffers from one step communication delay. For saving the storage space, the information delayed more than D step times is discarded. In addition, to model the system in a more practical case, we assume that the observation for the subsystem output suffers random missing. Under this new information pattern, the optimal outputfeedback control problem is nonconvex, what is worse, the separation principle fails. This implies that the optimal control problem with the information pattern introduced above is difficult to solve. In this paper, a new decentralized control scheme is proposed. In particular, a new estimator structure and a new controller structure are constructed, and the gains of the estimator and the controller are designed simultaneously. An optimality condition with respect to the gains is established. Based on the optimality condition, an iterative algorithm is exploited to design the gains numerically. It is shown that the exploited algorithm converges to Nash optimum. Finally, the proposed theoretical results are illustrated by a physical system which is a heavy duty vehicles platoon.
 Lowpower peakingfree highgain observers
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Daniele Astolfi, Lorenzo Marconi, Laurent Praly, Andrew R. Teel We propose a peakingfree lowpower highgain observer that preserves the main feature of standard highgain observers in terms of arbitrarily fast convergence to zero of the estimation error, while overtaking their main drawbacks, namely the “peaking phenomenon” during the transient and the numerical implementation issue deriving from the highgain parameter that is powered up to the order of the system. Moreover, the new observer is proved to have superior features in terms of sensitivity of the estimation error to highfrequency measurement noise when compared with standard highgain observers. The proposed observer structure has a highgain parameter that is powered just up to two regardless the dimension of the observed system and adopts saturations to prevent the peaking of the estimates during the transient. As for the classical solution, the new observer is robust with respect to uncertainties in the observed system dynamics in the sense that practical estimation in the highgain parameter can be proved.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Daniele Astolfi, Lorenzo Marconi, Laurent Praly, Andrew R. Teel We propose a peakingfree lowpower highgain observer that preserves the main feature of standard highgain observers in terms of arbitrarily fast convergence to zero of the estimation error, while overtaking their main drawbacks, namely the “peaking phenomenon” during the transient and the numerical implementation issue deriving from the highgain parameter that is powered up to the order of the system. Moreover, the new observer is proved to have superior features in terms of sensitivity of the estimation error to highfrequency measurement noise when compared with standard highgain observers. The proposed observer structure has a highgain parameter that is powered just up to two regardless the dimension of the observed system and adopts saturations to prevent the peaking of the estimates during the transient. As for the classical solution, the new observer is robust with respect to uncertainties in the observed system dynamics in the sense that practical estimation in the highgain parameter can be proved.
 Low complexity constrained control using higher degree Lyapunov functions
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Sarmad Munir, Morten Hovd, Sorin Olaru Explicit Model Predictive Control often has a complex solution in terms of the number of regions required to define the solution and the corresponding memory requirement to represent the solution in the online implementation. An alternative approach to constrained control is based on the use of controlled contractive sets. However, polytopic controlled contractive sets may themselves be relatively complex, leading to a complex explicit solution, and the polytopic structure can limit the size of the controlled contractive set. This paper develops a method to obtain a larger controlled contractive set by allowing higher order functions in the definition of the contractive set, and explores the use of such higherorder contractive sets in controller design leading to a low complexity explicit control formulation.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Sarmad Munir, Morten Hovd, Sorin Olaru Explicit Model Predictive Control often has a complex solution in terms of the number of regions required to define the solution and the corresponding memory requirement to represent the solution in the online implementation. An alternative approach to constrained control is based on the use of controlled contractive sets. However, polytopic controlled contractive sets may themselves be relatively complex, leading to a complex explicit solution, and the polytopic structure can limit the size of the controlled contractive set. This paper develops a method to obtain a larger controlled contractive set by allowing higher order functions in the definition of the contractive set, and explores the use of such higherorder contractive sets in controller design leading to a low complexity explicit control formulation.
 Dissipative control for nonlinear Markovian jump systems with actuator
failures and mixed timedelays Abstract: Publication date: Available online 25 September 2018Source: AutomaticaAuthor(s): Lifeng Ma, Zidong Wang, QingLong Han, Yurong Liu This paper addresses the dissipative control problem for nonlinear Markovian jump systems subject to actuator failures and mixed timedelays, where the mixed timedelays consist of both discrete and distributed timedelays and are modedependent. The purpose of the problem under investigation is to design a state feedback controller such that, in the presence of actuator failures and mixed timedelays, the closedloop system is asymptotically stable in the mean square sense while achieving the prespecified dissipativity. By constructing a Lyapunov–Krasovskii functional and using a completing square approach, sufficient conditions are proposed for the existence of the desired controller in terms of the solvability of certain Hamilton–Jacobi inequalities. Finally, an illustrative numerical example is provided to demonstrate the effectiveness of the developed control scheme.
 Abstract: Publication date: Available online 25 September 2018Source: AutomaticaAuthor(s): Lifeng Ma, Zidong Wang, QingLong Han, Yurong Liu This paper addresses the dissipative control problem for nonlinear Markovian jump systems subject to actuator failures and mixed timedelays, where the mixed timedelays consist of both discrete and distributed timedelays and are modedependent. The purpose of the problem under investigation is to design a state feedback controller such that, in the presence of actuator failures and mixed timedelays, the closedloop system is asymptotically stable in the mean square sense while achieving the prespecified dissipativity. By constructing a Lyapunov–Krasovskii functional and using a completing square approach, sufficient conditions are proposed for the existence of the desired controller in terms of the solvability of certain Hamilton–Jacobi inequalities. Finally, an illustrative numerical example is provided to demonstrate the effectiveness of the developed control scheme.
 Nonlinear state estimation under bounded noises
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Bo Chen, Guoqiang Hu Most of the existing nonlinear state estimation methods require to know the statistical information of noises. However, the statistical information may not be accurately obtained or satisfied in practical applications. Actually, the noises are always bounded in a practical system. In this paper, we study the nonlinear state estimation problem under bounded noises, where the addressed noises do not provide any statistical information, and the bounds of noises are also unknown. By using matrix analysis and secondorder Taylor series expansion, a novel constructive method is proposed to find an upper bound of the square error of the nonlinear estimator. Then, a convex optimization problem on the design of an optimal estimator gain is established in terms of linear matrix inequalities, which can be solved by standard software packages. Moreover, stability conditions are derived such that the square error of the designed nonlinear estimator is asymptotically bounded. Finally, two illustrative examples are employed to show the advantages and effectiveness of the proposed methods.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Bo Chen, Guoqiang Hu Most of the existing nonlinear state estimation methods require to know the statistical information of noises. However, the statistical information may not be accurately obtained or satisfied in practical applications. Actually, the noises are always bounded in a practical system. In this paper, we study the nonlinear state estimation problem under bounded noises, where the addressed noises do not provide any statistical information, and the bounds of noises are also unknown. By using matrix analysis and secondorder Taylor series expansion, a novel constructive method is proposed to find an upper bound of the square error of the nonlinear estimator. Then, a convex optimization problem on the design of an optimal estimator gain is established in terms of linear matrix inequalities, which can be solved by standard software packages. Moreover, stability conditions are derived such that the square error of the designed nonlinear estimator is asymptotically bounded. Finally, two illustrative examples are employed to show the advantages and effectiveness of the proposed methods.
 Collaborative operational fault tolerant control for stochastic
distribution control system Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Yuwei Ren, Yixian Fang, Aiping Wang, Huaxiang Zhang, Hong Wang Based on a class of industrial processes, a new distributed fault diagnosis approach and a collaborative operational fault tolerant control law are proposed for an irreversible interconnected stochastic distribution control (SDC) system with boundary conditions. This control method is different from the existing collaborative fault tolerant controllers which enable the output probability density function (PDF) to track a desired PDF as close as possible. When fault occurs, a setpoint redesigned fault tolerant approach is adopted to accommodate the fault instead of reconstructing the controller. An augmented PID nominal controller and a setpoint compensation item with linear structure are used to obtain a collaborative operational fault tolerant controller via solution of linear matrix inequalities (LMIs). Simulations are included to show the effectiveness of the proposed algorithms where encouraging results have been obtained.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Yuwei Ren, Yixian Fang, Aiping Wang, Huaxiang Zhang, Hong Wang Based on a class of industrial processes, a new distributed fault diagnosis approach and a collaborative operational fault tolerant control law are proposed for an irreversible interconnected stochastic distribution control (SDC) system with boundary conditions. This control method is different from the existing collaborative fault tolerant controllers which enable the output probability density function (PDF) to track a desired PDF as close as possible. When fault occurs, a setpoint redesigned fault tolerant approach is adopted to accommodate the fault instead of reconstructing the controller. An augmented PID nominal controller and a setpoint compensation item with linear structure are used to obtain a collaborative operational fault tolerant controller via solution of linear matrix inequalities (LMIs). Simulations are included to show the effectiveness of the proposed algorithms where encouraging results have been obtained.
 Secure Luenbergerlike observers for cyber–physical systems under sparse
actuator and sensor attacks Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): AnYang Lu, GuangHong Yang This paper investigates the secure state estimation problem for cyber–physical systems (CPSs) under sparse actuator and sensor attacks. By introducing the notion of orthogonal complement matrix, a necessary and sufficient condition for the state observability is provided. Then, based on the least square technique, a new projection operator is proposed to reconstruct the state from a set of successive measurements. Besides, by constructing an augmented system where the attacks are seen as part of the augmented state vector, a novel secure Luenbergerlike observer is proposed, and sufficient conditions for the existence of the desired observer are proposed in terms of linear matrix inequalities (LMIs). It is shown that the proposed observability condition can be reduced to the sparse observability. A distinguishing point is that the attacks may be still unavailable even if the state is observable, and besides estimating the state, the attacks are also reconstructed by the proposed algorithm and observer according to their observability automatically.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): AnYang Lu, GuangHong Yang This paper investigates the secure state estimation problem for cyber–physical systems (CPSs) under sparse actuator and sensor attacks. By introducing the notion of orthogonal complement matrix, a necessary and sufficient condition for the state observability is provided. Then, based on the least square technique, a new projection operator is proposed to reconstruct the state from a set of successive measurements. Besides, by constructing an augmented system where the attacks are seen as part of the augmented state vector, a novel secure Luenbergerlike observer is proposed, and sufficient conditions for the existence of the desired observer are proposed in terms of linear matrix inequalities (LMIs). It is shown that the proposed observability condition can be reduced to the sparse observability. A distinguishing point is that the attacks may be still unavailable even if the state is observable, and besides estimating the state, the attacks are also reconstructed by the proposed algorithm and observer according to their observability automatically.
 Robust consensus of uncertain linear multiagent systems via dynamic
output feedback Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Xianwei Li, Yeng Chai Soh, Lihua Xie This paper systematically deals with robust consensus of uncertain linear multiagent systems via dynamic outputfeedback protocols. Agents are assumed to have identical nominal linear timeinvariant dynamics but are subject to heterogeneous additive stable perturbations. Dynamic outputfeedback protocols with or without controller state information exchange between neighboring controllers are studied in a unified framework. Two methods are proposed for protocol design, which need to solve an algebraic Riccati equation and some scalar/matrix inequalities. The first method characterizes some key parameters by scalar inequalities related to the nonzero eigenvalues of the Laplacian, which requires the diagonalizability of the Laplacian, while the second method characterizes the parameters by linear matrix inequalities, which circumvents the requirement of the diagonalizability of the Laplacian. Compared with existing results, the proposed approach can simultaneously cope with heterogeneous uncertainties and directed communication graphs. Numerical results verify the advantages of the proposed design methods.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Xianwei Li, Yeng Chai Soh, Lihua Xie This paper systematically deals with robust consensus of uncertain linear multiagent systems via dynamic outputfeedback protocols. Agents are assumed to have identical nominal linear timeinvariant dynamics but are subject to heterogeneous additive stable perturbations. Dynamic outputfeedback protocols with or without controller state information exchange between neighboring controllers are studied in a unified framework. Two methods are proposed for protocol design, which need to solve an algebraic Riccati equation and some scalar/matrix inequalities. The first method characterizes some key parameters by scalar inequalities related to the nonzero eigenvalues of the Laplacian, which requires the diagonalizability of the Laplacian, while the second method characterizes the parameters by linear matrix inequalities, which circumvents the requirement of the diagonalizability of the Laplacian. Compared with existing results, the proposed approach can simultaneously cope with heterogeneous uncertainties and directed communication graphs. Numerical results verify the advantages of the proposed design methods.
 Eventtriggered identification of FIR systems with binaryvalued output
observations Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): JingDong Diao, Jin Guo, ChangYin Sun This paper investigates the identification of FIR (finite impulse response) systems whose output observations are subject to both the binaryvalued quantization and the eventtriggered scheme. Based on the a priori information of the unknown parameters and the statistical property of the system noise, a recursive stochasticapproximationtype identification algorithm is proposed. Under a class of persistently exciting inputs, the algorithm is proved to be strongly convergent and the convergence rate of the estimation error is also established, where the corresponding eventtriggering conditions are provided. Moreover, the communication rate is discussed. A numerical example is included to verify the effectiveness of the results obtained.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): JingDong Diao, Jin Guo, ChangYin Sun This paper investigates the identification of FIR (finite impulse response) systems whose output observations are subject to both the binaryvalued quantization and the eventtriggered scheme. Based on the a priori information of the unknown parameters and the statistical property of the system noise, a recursive stochasticapproximationtype identification algorithm is proposed. Under a class of persistently exciting inputs, the algorithm is proved to be strongly convergent and the convergence rate of the estimation error is also established, where the corresponding eventtriggering conditions are provided. Moreover, the communication rate is discussed. A numerical example is included to verify the effectiveness of the results obtained.
 Modeling and iterative pulseshape control of optical chirped pulse
amplifiers Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Andreas Deutschmann, Pavel Malevich, Andrius Baltuška, Andreas Kugi In this paper, we present an iterative learning algorithm for pulseshape control applications of optical chirped pulse amplifiers for ultrashort, highenergy light pulses. For this, we first introduce a general nonlinear and infinitedimensional mathematical model of chirped pulse amplifiers. By reducing the complexity of this detailed model and reformulating the control task, we are subsequently able to apply inversionbased iterative learning control to track desired output pulses. Using the reduced model to estimate both internal states and unknown parameters yields a fast and simple way of consistently estimating the input–output behavior without relying on a calibrated system model. The effectiveness of the resulting adaptive algorithm is finally illustrated with simulation scenarios on an experimentally validated mathematical model.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Andreas Deutschmann, Pavel Malevich, Andrius Baltuška, Andreas Kugi In this paper, we present an iterative learning algorithm for pulseshape control applications of optical chirped pulse amplifiers for ultrashort, highenergy light pulses. For this, we first introduce a general nonlinear and infinitedimensional mathematical model of chirped pulse amplifiers. By reducing the complexity of this detailed model and reformulating the control task, we are subsequently able to apply inversionbased iterative learning control to track desired output pulses. Using the reduced model to estimate both internal states and unknown parameters yields a fast and simple way of consistently estimating the input–output behavior without relying on a calibrated system model. The effectiveness of the resulting adaptive algorithm is finally illustrated with simulation scenarios on an experimentally validated mathematical model.
 Combinatorial methods for invariance and safety of hybrid systems
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Nikolaos Athanasopoulos, Raphaël M. Jungers Inspired by Switching Systems and Automata theory, we investigate how combinatorial analysis techniques can be performed on a hybrid automaton in order to enhance its safety or invariance analysis. We focus on the particular case of Constrained Switching Systems, that is, hybrid automata with linear dynamics and no guards. We follow two opposite approaches, each with unique benefits: First, we construct invariant sets via the ‘Reduced’ system, induced by a smaller graph which consists of the essential nodes, called the unavoidable nodes. The computational amelioration of working with a smaller, and in certain cases the minimum necessary number of nodes, is significant. Second, we exploit graph liftings, in particular the Iterated Dynamics Lift (TLift) and the PathDependent Lift (PLift). For the former case, we show that invariant sets can be computed in a fraction of the iterations compared to the nonlifted case, while we show how the latter can be utilized to compute nonconvex approximations of invariant sets of a controlled complexity.We also revisit well studied problems, highlighting the potential benefits of the approach. In particular, we apply our framework to (i) invariant sets computations for systems with dwelltime restrictions, (ii) fast computations of the maximal invariant set for uncertain linear systems and (iii) nonconvex approximations of the minimal invariant set for arbitrary switching linear systems.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Nikolaos Athanasopoulos, Raphaël M. Jungers Inspired by Switching Systems and Automata theory, we investigate how combinatorial analysis techniques can be performed on a hybrid automaton in order to enhance its safety or invariance analysis. We focus on the particular case of Constrained Switching Systems, that is, hybrid automata with linear dynamics and no guards. We follow two opposite approaches, each with unique benefits: First, we construct invariant sets via the ‘Reduced’ system, induced by a smaller graph which consists of the essential nodes, called the unavoidable nodes. The computational amelioration of working with a smaller, and in certain cases the minimum necessary number of nodes, is significant. Second, we exploit graph liftings, in particular the Iterated Dynamics Lift (TLift) and the PathDependent Lift (PLift). For the former case, we show that invariant sets can be computed in a fraction of the iterations compared to the nonlifted case, while we show how the latter can be utilized to compute nonconvex approximations of invariant sets of a controlled complexity.We also revisit well studied problems, highlighting the potential benefits of the approach. In particular, we apply our framework to (i) invariant sets computations for systems with dwelltime restrictions, (ii) fast computations of the maximal invariant set for uncertain linear systems and (iii) nonconvex approximations of the minimal invariant set for arbitrary switching linear systems.
 Exponential convergence under distributed averaging integral frequency
control Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Erik Weitenberg, Claudio De Persis, Nima Monshizadeh We investigate the performance and robustness of distributed averaging integral controllers used in the optimal frequency regulation of power networks. We construct a strict Lyapunov function that allows us to quantify the exponential convergence rate of the closedloop system. As an application, we study the stability of the system in the presence of disruptions to the controllers’ communication network, and investigate how the convergence rate is affected by these disruptions.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Erik Weitenberg, Claudio De Persis, Nima Monshizadeh We investigate the performance and robustness of distributed averaging integral controllers used in the optimal frequency regulation of power networks. We construct a strict Lyapunov function that allows us to quantify the exponential convergence rate of the closedloop system. As an application, we study the stability of the system in the presence of disruptions to the controllers’ communication network, and investigate how the convergence rate is affected by these disruptions.
 Control for networked control systems with remote and local controllers
over unreliable communication channel Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Xiao Liang, Juanjuan Xu This paper is concerned with the problems of optimal control and stabilization for networked control systems (NCSs), where the remote controller and the local controller operate the linear plant simultaneously. The main contributions are twofold. Firstly, a necessary and sufficient condition for the finite horizon optimal control problem is given in terms of the two Riccati equations. Secondly, it is shown that the system without the additive noise is stabilizable in the mean square sense if and only if the two algebraic Riccati equations admit the unique solutions, and a sufficient condition is given for the boundedness in the mean square sense of the system with the additive noise. Numerical examples about unmanned aerial vehicles model are shown to illustrate the effectiveness of the proposed algorithm.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Xiao Liang, Juanjuan Xu This paper is concerned with the problems of optimal control and stabilization for networked control systems (NCSs), where the remote controller and the local controller operate the linear plant simultaneously. The main contributions are twofold. Firstly, a necessary and sufficient condition for the finite horizon optimal control problem is given in terms of the two Riccati equations. Secondly, it is shown that the system without the additive noise is stabilizable in the mean square sense if and only if the two algebraic Riccati equations admit the unique solutions, and a sufficient condition is given for the boundedness in the mean square sense of the system with the additive noise. Numerical examples about unmanned aerial vehicles model are shown to illustrate the effectiveness of the proposed algorithm.
 Suboptimal receding horizon estimation via noise blocking
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): He Kong, Salah Sukkarieh For discretetime linear systems, we propose a suboptimal approach to constrained estimation so that the associated computation burden is reduced. This is achieved by enforcing a move blocking (MB) structure in the estimated process noise sequence (PNS). We show that full information estimation (FIE) and receding horizon estimation (RHE) with MB are both stable in the sense of an observer. The techniques in proving stability are inspired by those that have been proposed for standard RHE. To be specific, stability results are mainly achieved by (i) carefully embellishing the general assumptions for standard RHE to accommodate the MB requirement; (ii) exploiting the principle of optimality, as well as convexity of the quadratic programs (QPs) associated with FIE and RHE; (iii) relying on the fact that the Kalman filter is the best linear estimator in the leastsquares sense. A crucial requirement in achieving stability for MB RHE is that the segment structure (SS) of the PNS of MB FIE for the optimization steps within the receding horizon (i.e., steps between T−N and T−1) has to be enforced in the MB RHE optimization. As a result, the MB RHE strategy becomes a dynamic estimator with a periodically varying computational complexity. The theoretical results have been illustrated with examples.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): He Kong, Salah Sukkarieh For discretetime linear systems, we propose a suboptimal approach to constrained estimation so that the associated computation burden is reduced. This is achieved by enforcing a move blocking (MB) structure in the estimated process noise sequence (PNS). We show that full information estimation (FIE) and receding horizon estimation (RHE) with MB are both stable in the sense of an observer. The techniques in proving stability are inspired by those that have been proposed for standard RHE. To be specific, stability results are mainly achieved by (i) carefully embellishing the general assumptions for standard RHE to accommodate the MB requirement; (ii) exploiting the principle of optimality, as well as convexity of the quadratic programs (QPs) associated with FIE and RHE; (iii) relying on the fact that the Kalman filter is the best linear estimator in the leastsquares sense. A crucial requirement in achieving stability for MB RHE is that the segment structure (SS) of the PNS of MB FIE for the optimization steps within the receding horizon (i.e., steps between T−N and T−1) has to be enforced in the MB RHE optimization. As a result, the MB RHE strategy becomes a dynamic estimator with a periodically varying computational complexity. The theoretical results have been illustrated with examples.
 Sliding mode observers for a network of thermal and hydroelectric power
plants Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Gianmario Rinaldi, Michele Cucuzzella, Antonella Ferrara This paper deals with the design of a novel sliding mode observerbased scheme to estimate and reconstruct the unmeasured state variables in power networks including hydroelectric power plants and thermal power plants. The proposed approach reveals to be flexible to topological changes to power networks and can be easily updated only where changes occur. The discussed numerical simulations validate the effectiveness of the proposed estimation scheme.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Gianmario Rinaldi, Michele Cucuzzella, Antonella Ferrara This paper deals with the design of a novel sliding mode observerbased scheme to estimate and reconstruct the unmeasured state variables in power networks including hydroelectric power plants and thermal power plants. The proposed approach reveals to be flexible to topological changes to power networks and can be easily updated only where changes occur. The discussed numerical simulations validate the effectiveness of the proposed estimation scheme.
 Robust adaptive fault tolerant control for a linear cascaded ODEbeam
system Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Zhijie Liu, Jinkun Liu, Wei He In this paper, we present fault tolerant control design for a class of cascaded systems described by ordinary differential equations (ODEs) and an Euler–Bernoulli beam (EBB). The objective of this study is to design a robust adaptive fault tolerant control such that the global stability of the resulting closedloop cascaded system is ensured and asymptotic tracking can be achieved subject to actuator failures, parameter uncertainty and external disturbances. The Lyapunov’s direct method is used to design the control schemes and prove the stability of the closedloop system. Finally, the results are illustrated using numerical simulations for control performance verification.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Zhijie Liu, Jinkun Liu, Wei He In this paper, we present fault tolerant control design for a class of cascaded systems described by ordinary differential equations (ODEs) and an Euler–Bernoulli beam (EBB). The objective of this study is to design a robust adaptive fault tolerant control such that the global stability of the resulting closedloop cascaded system is ensured and asymptotic tracking can be achieved subject to actuator failures, parameter uncertainty and external disturbances. The Lyapunov’s direct method is used to design the control schemes and prove the stability of the closedloop system. Finally, the results are illustrated using numerical simulations for control performance verification.
 Fault tolerant control for a class of interconnected asynchronous
sequential machines Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): JungMin Yang This paper considers fault tolerant control for parallel interconnected asynchronous sequential machines (ASMs) governed by a single corrective controller with output feedback. The control objective is to diagnose unauthorized state transitions and to recover the normal input/output behavior of the closedloop system in an asynchronous mechanism. The existence condition and design algorithm for a fault tolerant controller is addressed in the framework of corrective control. The proposed scheme is efficient in that it does not require complete modeling of parallel composition nor output bursts in the feedback channel. An illustrative example is provided to demonstrate the procedure of controller synthesis.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): JungMin Yang This paper considers fault tolerant control for parallel interconnected asynchronous sequential machines (ASMs) governed by a single corrective controller with output feedback. The control objective is to diagnose unauthorized state transitions and to recover the normal input/output behavior of the closedloop system in an asynchronous mechanism. The existence condition and design algorithm for a fault tolerant controller is addressed in the framework of corrective control. The proposed scheme is efficient in that it does not require complete modeling of parallel composition nor output bursts in the feedback channel. An illustrative example is provided to demonstrate the procedure of controller synthesis.
 Twostage information filters for single and multiple sensors, and their
squareroot versions Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Kumar Pakki Bharani Chandra, Mohamed Darouach Accurate states and unknown random bias estimation for well and illconditioned systems are crucial for several applications. In this paper, a fusion of a twostage Kalman filter and an information filter, and its extensions are considered to estimate the state variables and unknown random bias. Specifically, we propose four extensions of twostage Kalman filters: twostage information filter (TSIF), multisensor twostage information filter (MTSIF) and their squareroot versions. The TSIF deals with singlesensor systems whereas the MTSIF is capable to handle multisensor systems. For illconditioned systems, numerically stable squareroot versions of TSIF and MTSIF are developed. The performance of the proposed filters (along with the existing twostage Kalman filter), for well and illconditioned cases, is demonstrated on a quadrupletank model.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Kumar Pakki Bharani Chandra, Mohamed Darouach Accurate states and unknown random bias estimation for well and illconditioned systems are crucial for several applications. In this paper, a fusion of a twostage Kalman filter and an information filter, and its extensions are considered to estimate the state variables and unknown random bias. Specifically, we propose four extensions of twostage Kalman filters: twostage information filter (TSIF), multisensor twostage information filter (MTSIF) and their squareroot versions. The TSIF deals with singlesensor systems whereas the MTSIF is capable to handle multisensor systems. For illconditioned systems, numerically stable squareroot versions of TSIF and MTSIF are developed. The performance of the proposed filters (along with the existing twostage Kalman filter), for well and illconditioned cases, is demonstrated on a quadrupletank model.
 Analysis of Lur’e dominant systems in the frequency domain
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Félix A. MirandaVillatoro, Fulvio Forni, Rodolphe J. Sepulchre Frequency domain analysis of linear timeinvariant (LTI) systems in feedback with static nonlinearities is a classical and fruitful topic of nonlinear systems theory. We generalize this approach beyond equilibrium stability analysis with the aim of characterizing feedback systems whose asymptotic behavior is low dimensional. We illustrate the theory with a generalization of the circle criterion for the analysis of multistable and oscillatory Lur’e feedback systems.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Félix A. MirandaVillatoro, Fulvio Forni, Rodolphe J. Sepulchre Frequency domain analysis of linear timeinvariant (LTI) systems in feedback with static nonlinearities is a classical and fruitful topic of nonlinear systems theory. We generalize this approach beyond equilibrium stability analysis with the aim of characterizing feedback systems whose asymptotic behavior is low dimensional. We illustrate the theory with a generalization of the circle criterion for the analysis of multistable and oscillatory Lur’e feedback systems.
 Optimal multirate sampling in symbolic models for incrementally stable
switched systems Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Adnane Saoud, Antoine Girard Methods for computing approximately bisimilar symbolic models for incrementally stable switched systems are often based on discretization of time and space, where the value of time and space sampling parameters must be carefully chosen in order to achieve a desired precision. These approaches can result in symbolic models that have a very large number of transitions, especially when the time sampling, and thus the space sampling parameters are small. In this paper, we present an approach to the computation of symbolic models for switched systems with dwelltime constraints using multirate time sampling, where the period of symbolic transitions is a multiple of the control (i.e. switching) period. We show that all the multirate symbolic models, resulting from the proposed construction, are approximately bisimilar to the original incrementally stable switched system with the precision depending on the sampling parameters, and the sampling factor between transition and control periods. The main contribution of the paper is the explicit determination of the optimal sampling factor, which minimizes the number of transitions in the class of proposed symbolic models for a prescribed precision. Interestingly, we prove that this optimal sampling factor is mainly determined by the state space dimension and the number of modes of the switched system. Finally, an illustration of the proposed approach is shown on an example, which shows the benefit of multirate symbolic models in reducing the computational cost of abstractionbased controller synthesis.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Adnane Saoud, Antoine Girard Methods for computing approximately bisimilar symbolic models for incrementally stable switched systems are often based on discretization of time and space, where the value of time and space sampling parameters must be carefully chosen in order to achieve a desired precision. These approaches can result in symbolic models that have a very large number of transitions, especially when the time sampling, and thus the space sampling parameters are small. In this paper, we present an approach to the computation of symbolic models for switched systems with dwelltime constraints using multirate time sampling, where the period of symbolic transitions is a multiple of the control (i.e. switching) period. We show that all the multirate symbolic models, resulting from the proposed construction, are approximately bisimilar to the original incrementally stable switched system with the precision depending on the sampling parameters, and the sampling factor between transition and control periods. The main contribution of the paper is the explicit determination of the optimal sampling factor, which minimizes the number of transitions in the class of proposed symbolic models for a prescribed precision. Interestingly, we prove that this optimal sampling factor is mainly determined by the state space dimension and the number of modes of the switched system. Finally, an illustration of the proposed approach is shown on an example, which shows the benefit of multirate symbolic models in reducing the computational cost of abstractionbased controller synthesis.
 Boundary observability of wave equations with nonlinear van der Pol type
boundary conditions Abstract: Publication date: Available online 21 September 2018Source: AutomaticaAuthor(s): Shuting Cai, Mingqing Xiao In this note we study the boundary observability for onedimensional wave equation associated with nonlinear boundary condition that can generate complex dynamics. We discuss the exact observability and approximate observability, respectively, in terms of three different types of common boundary observations by studying the wave interactions on the boundary directly.
 Abstract: Publication date: Available online 21 September 2018Source: AutomaticaAuthor(s): Shuting Cai, Mingqing Xiao In this note we study the boundary observability for onedimensional wave equation associated with nonlinear boundary condition that can generate complex dynamics. We discuss the exact observability and approximate observability, respectively, in terms of three different types of common boundary observations by studying the wave interactions on the boundary directly.
 Analysis of averages over distributions of Markov processes
 Abstract: Publication date: Available online 21 September 2018Source: AutomaticaAuthor(s): Patricio E. Valenzuela, Cristian R. Rojas, Håkan Hjalmarsson In problems of optimal control of Markov decision processes and optimal design of experiments, the occupation measure of a Markov process is designed in order to maximize a specific reward function. When the memory of such a process is too long, or the process is nonMarkovian but mixing, it makes sense to approximate it by that of a shorter memory Markov process. This note provides a specific bound for the approximation error introduced in these schemes. The derived bound is then applied to the proposed solution of a recently introduced approach to optimal input design for nonlinear systems.
 Abstract: Publication date: Available online 21 September 2018Source: AutomaticaAuthor(s): Patricio E. Valenzuela, Cristian R. Rojas, Håkan Hjalmarsson In problems of optimal control of Markov decision processes and optimal design of experiments, the occupation measure of a Markov process is designed in order to maximize a specific reward function. When the memory of such a process is too long, or the process is nonMarkovian but mixing, it makes sense to approximate it by that of a shorter memory Markov process. This note provides a specific bound for the approximation error introduced in these schemes. The derived bound is then applied to the proposed solution of a recently introduced approach to optimal input design for nonlinear systems.
 Linear programming based time lag identification in event sequences
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Marco F. Huber, MarcAndré Zöller, Marcus Baum Many technical systems like manufacturing plants or software applications generate large event sequences. Knowing the temporal relationship between events is important for gaining insights into the status and behavior of the system. This paper proposes a novel approach for identifying the time lag between different event types. This identification task is formulated as a binary integer optimization problem that can be solved efficiently and close to optimality by means of a linear programming approximation. The performance of the proposed approach is demonstrated on synthetic and realworld event sequences.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Marco F. Huber, MarcAndré Zöller, Marcus Baum Many technical systems like manufacturing plants or software applications generate large event sequences. Knowing the temporal relationship between events is important for gaining insights into the status and behavior of the system. This paper proposes a novel approach for identifying the time lag between different event types. This identification task is formulated as a binary integer optimization problem that can be solved efficiently and close to optimality by means of a linear programming approximation. The performance of the proposed approach is demonstrated on synthetic and realworld event sequences.
 Particle Gaussian mixture filtersI
 Abstract: Publication date: Available online 14 September 2018Source: AutomaticaAuthor(s): Dilshad Raihan, Suman Chakravorty In this paper, we propose a particle based Gaussian mixture filtering approach for nonlinear estimation that is free of the particle depletion problem inherent to most particle filters. We employ an ensemble of possible state realizations for the propagation of state probability density. A Gaussian mixture model (GMM) of the propagated uncertainty is then recovered by clustering the ensemble. The posterior density is obtained subsequently through a Kalman measurement update of the mixture modes. We prove the convergence in probability of the resultant density to the true filter density assuming exponential forgetting of initial conditions. The performance of the proposed filtering approach is demonstrated through several test cases and is extensively compared to other nonlinear filters.
 Abstract: Publication date: Available online 14 September 2018Source: AutomaticaAuthor(s): Dilshad Raihan, Suman Chakravorty In this paper, we propose a particle based Gaussian mixture filtering approach for nonlinear estimation that is free of the particle depletion problem inherent to most particle filters. We employ an ensemble of possible state realizations for the propagation of state probability density. A Gaussian mixture model (GMM) of the propagated uncertainty is then recovered by clustering the ensemble. The posterior density is obtained subsequently through a Kalman measurement update of the mixture modes. We prove the convergence in probability of the resultant density to the true filter density assuming exponential forgetting of initial conditions. The performance of the proposed filtering approach is demonstrated through several test cases and is extensively compared to other nonlinear filters.
 A distributed approach to robust control of multirobot systems
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Yuan Zhou, Hesuan Hu, Yang Liu, ShangWei Lin, Zuohua Ding Motion planning of multirobot systems has been extensively investigated. Many proposed approaches assume that all robots are reliable. However, robots with priori known levels of reliability may be used in applications to account for: (1) the cost in terms of unit price per robot type, and (2) the cost in terms of robot wear in long term deployment. In the former case, higher reliability comes at a higher price, while in the latter replacement may cost more than periodic repairs, e.g., buses, trams, and subways. In this study, we investigate robust control of multirobot systems, such that the number of robots affected by the failed ones is minimized. It should mandate that the failure of a robot can only affect the motion of robots that collide directly with the failed one. We assume that the robots in a system are divided into reliable and unreliable ones, and each robot has a predetermined and closed path to execute persistent tasks. By modeling each robot’s motion as a labeled transition system, we propose two distributed robust control algorithms: one for reliable robots and the other for unreliable ones. The algorithms guarantee that wherever an unreliable robot fails, only the robots whose state spaces contain the failed state are blocked. Theoretical analysis shows that the proposed algorithms are practically operative. Simulations with seven robots are carried out and the results show the effectiveness of our algorithms.
 Abstract: Publication date: December 2018Source: Automatica, Volume 98Author(s): Yuan Zhou, Hesuan Hu, Yang Liu, ShangWei Lin, Zuohua Ding Motion planning of multirobot systems has been extensively investigated. Many proposed approaches assume that all robots are reliable. However, robots with priori known levels of reliability may be used in applications to account for: (1) the cost in terms of unit price per robot type, and (2) the cost in terms of robot wear in long term deployment. In the former case, higher reliability comes at a higher price, while in the latter replacement may cost more than periodic repairs, e.g., buses, trams, and subways. In this study, we investigate robust control of multirobot systems, such that the number of robots affected by the failed ones is minimized. It should mandate that the failure of a robot can only affect the motion of robots that collide directly with the failed one. We assume that the robots in a system are divided into reliable and unreliable ones, and each robot has a predetermined and closed path to execute persistent tasks. By modeling each robot’s motion as a labeled transition system, we propose two distributed robust control algorithms: one for reliable robots and the other for unreliable ones. The algorithms guarantee that wherever an unreliable robot fails, only the robots whose state spaces contain the failed state are blocked. Theoretical analysis shows that the proposed algorithms are practically operative. Simulations with seven robots are carried out and the results show the effectiveness of our algorithms.
 Particle Gaussian mixture filtersII
 Abstract: Publication date: Available online 13 September 2018Source: AutomaticaAuthor(s): Dilshad Raihan, Suman Chakravorty In our previous work, we proposed a particle Gaussian mixture (PGMI) filter for nonlinear estimation. The PGMI filter uses the transition kernel of the state Markov chain to sample from the propagated prior. It constructs a Gaussian mixture representation of the propagated prior density by clustering the samples. The measurement data are incorporated by updating individual mixture modes using the Kalman measurement update. However, the Kalman measurement update is inexact when the measurement function is nonlinear and leads to the restrictive assumption that the number of modes remains fixed during the measurement update. In this paper, we introduce an alternate PGMII filter that employs parallelized Markov Chain Monte Carlo (MCMC) sampling to perform the measurement update. The PGMII filter update is asymptotically exact and does not enforce any assumptions on the number of Gaussian modes. The PGMII filter is employed in the estimation of two test case systems. The results indicate that the PGMII filter is suitable for handling nonlinear/nonGaussian measurement update.
 Abstract: Publication date: Available online 13 September 2018Source: AutomaticaAuthor(s): Dilshad Raihan, Suman Chakravorty In our previous work, we proposed a particle Gaussian mixture (PGMI) filter for nonlinear estimation. The PGMI filter uses the transition kernel of the state Markov chain to sample from the propagated prior. It constructs a Gaussian mixture representation of the propagated prior density by clustering the samples. The measurement data are incorporated by updating individual mixture modes using the Kalman measurement update. However, the Kalman measurement update is inexact when the measurement function is nonlinear and leads to the restrictive assumption that the number of modes remains fixed during the measurement update. In this paper, we introduce an alternate PGMII filter that employs parallelized Markov Chain Monte Carlo (MCMC) sampling to perform the measurement update. The PGMII filter update is asymptotically exact and does not enforce any assumptions on the number of Gaussian modes. The PGMII filter is employed in the estimation of two test case systems. The results indicate that the PGMII filter is suitable for handling nonlinear/nonGaussian measurement update.