Automatica最新文献

{"title":"A rolling horizon game considering network effect in cluster forming for dynamic resilient multiagent systems","authors":"Yurid E. Nugraha ,&nbsp;Ahmet Cetinkaya ,&nbsp;Tomohisa Hayakawa ,&nbsp;Hideaki Ishii ,&nbsp;Quanyan Zhu","doi":"10.1016/j.automatica.2025.112137","DOIUrl":"10.1016/j.automatica.2025.112137","url":null,"abstract":"<div><div>A two-player game-theoretic problem on resilient graphs in a multiagent consensus setting is formulated. An attacker is capable to disable some of the edges of the network with the objective to divide the agents into clusters by emitting jamming signals while, in response, the defender recovers some of the edges by increasing the transmission power for the communication signals. Specifically, we consider repeated games between the attacker and the defender where the optimal strategies for the two players are derived in a rolling horizon fashion based on utility functions that take both the agents’ states and the sizes of clusters (known as network effect) into account. The players’ actions at each discrete-time step are constrained by their energy for transmissions of the signals, with a less strict constraint for the attacker. Necessary conditions and sufficient conditions of agent consensus are derived, and the number of clusters of agents at infinite time in the face of attacks and recoveries is also characterized. Simulation results are provided to demonstrate the effects of players’ actions on the cluster forming and to illustrate the players’ performance for different horizon parameters.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"175 ","pages":"Article 112137"},"PeriodicalIF":4.8,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143446121","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DualBi: A dual bisection algorithm for non-convex problems with a scalar complicating constraint","authors":"Lucrezia Manieri,&nbsp;Alessandro Falsone,&nbsp;Maria Prandini","doi":"10.1016/j.automatica.2025.112198","DOIUrl":"10.1016/j.automatica.2025.112198","url":null,"abstract":"<div><div>This paper addresses non-convex constrained optimization problems that are characterized by a scalar complicating constraint. We propose an iterative bisection method for the dual problem (DualBi Algorithm) that recovers a feasible primal solution, with a performance that progressively improves throughout iterations. Application to multi-agent problems with a scalar coupling constraint results in a decentralized resolution scheme where a central unit is in charge of updating the (scalar) dual variable while agents compute their local primal variables. In the case of multi-agent MILPs, simulations showcase the performance of the proposed method compared with state-of-the-art duality-based approaches.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"175 ","pages":"Article 112198"},"PeriodicalIF":4.8,"publicationDate":"2025-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143420849","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Adaptive non-cooperative differential games with a regulator","authors":"Nian Liu ,&nbsp;Shaolin Tan ,&nbsp;Ye Tao ,&nbsp;Jinhu Lü","doi":"10.1016/j.automatica.2025.112201","DOIUrl":"10.1016/j.automatica.2025.112201","url":null,"abstract":"<div><div>This paper considers linear–quadratic non-cooperative non-zero-sum stochastic differential games with a regulator and analyzes the adaptive problem when the systems matrices are unknown to both the regulator and the players. It is a typical problem of game-based control systems(GBCS) introduced and studied recently, which have a hierarchical decision-making structure. The main purpose of the paper is to study how the adaptive strategies can be designed to make the GBCS globally stable and at the same time to ensure a Nash equilibrium reached by both the regulator and the players. Under some suitable conditions on the system matrices, it is shown that the closed-loop adaptive GBCS will be globally stable, and at the same time reach a Nash equilibrium by both the regulator and the players, where the adaptive strategies are constructed based on the least squares estimator, the switching method and the diminishing excitation.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"175 ","pages":"Article 112201"},"PeriodicalIF":4.8,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143394779","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Power and bit scheduling of Markov jump systems with convergence rate as an optimization index","authors":"Jingjing Yan ,&nbsp;Yuanqing Xia ,&nbsp;Xinjing Wang ,&nbsp;Li Ma","doi":"10.1016/j.automatica.2025.112199","DOIUrl":"10.1016/j.automatica.2025.112199","url":null,"abstract":"<div><div>Existing power and bit scheduling algorithms mostly focus on open-loop system performance, i.e., improving estimation accuracy. This paper focuses on the scheduling methods for the closed-loop Markov jump systems in the unreliable transmission environments to improve the system stability and save energy. First, a control unit including feedback controller and predictive controller is proposed which improves the system performance while reducing the complexity of predictive controller design. Second, we design a novel optimization indicator based on time-varying convergence rate and sensor energy consumption. Third, by analyzing the relationship between Lyapunov function and the system state, an explicit expression of the time-varying convergence rate is gained. Next, a constant <span><math><mi>χ</mi></math></span> is introduced to obtain the effective power set, in which the convergence rate is always less than 1, thereby ensuring the system stability. Based on this, the optimal power and bit scheduling algorithm is obtained, which improves the system convergence speed while reducing energy consumption. Last, a two-tanks system is used to verify the effectiveness and superiority of the main algorithms.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"175 ","pages":"Article 112199"},"PeriodicalIF":4.8,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143394780","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Physics-informed neural network Lyapunov functions: PDE characterization, learning, and verification","authors":"Jun Liu,&nbsp;Yiming Meng,&nbsp;Maxwell Fitzsimmons,&nbsp;Ruikun Zhou","doi":"10.1016/j.automatica.2025.112193","DOIUrl":"10.1016/j.automatica.2025.112193","url":null,"abstract":"<div><div>We provide a systematic investigation of using physics-informed neural networks to compute Lyapunov functions. We encode Lyapunov conditions as a partial differential equation (PDE) and use this for training neural network Lyapunov functions. We analyze the analytical properties of the solutions to the Lyapunov and Zubov PDEs. In particular, we show that employing the Zubov equation in training neural Lyapunov functions can lead to verifiable approximate regions of attraction close to the true domain of attraction. We also examine approximation errors and the convergence of neural approximations to the unique solution of Zubov’s equation. We then provide sufficient conditions for the learned neural Lyapunov functions that can be readily verified by satisfiability modulo theories (SMT) solvers, enabling formal verification of both local stability analysis and region-of-attraction estimates in the large. Through a number of nonlinear examples, ranging from low to high dimensions, we demonstrate that the proposed framework can outperform traditional sum-of-squares (SOS) Lyapunov functions obtained using semidefinite programming (SDP).</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"175 ","pages":"Article 112193"},"PeriodicalIF":4.8,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143394778","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Scheduling-based stabilization for networked stochastic systems with control-dependent noise","authors":"Fengzhong Li,&nbsp;Yungang Liu","doi":"10.1016/j.automatica.2025.112192","DOIUrl":"10.1016/j.automatica.2025.112192","url":null,"abstract":"<div><div>Communication constraint is prominent under the networked architecture, for which the scheduling problem needs to be considered. This paper aims at validating scheduling-based stabilization for networked stochastic systems (NSSs) characterized by the presence of control-dependent noise. Remarkably, control-dependent noise has never been taken into account in the context of scheduling-based control. Such noise has intricate impact on system stability, and should not be simply treated as an unfavorable factor. As such, sophisticated analysis is entailed for scheduling-based control in the stochastic setting. However, no available theory on stochastic stability could support the application of dynamic scheduling protocols confronted with control-dependent noise, while the case via static scheduling protocols can resort to the existing results on stochastic differential equations with bounded-delay dependent diffusion coefficients. As the main contribution, a framework of scheduling-based stabilization is established for the NSSs, covering the architecture where a certain dynamic protocol rules the scheduling of sensor-to-controller information transmission. Crucially, a distinct pattern of stability analysis is proposed based on delicate comparison between the solutions under scheduling-based (discrete-time) and network-free (continuous-time) controls. As a consequence, an intrinsic relation between the system stability rendered by two types of controls is revealed, with the network-induced error on measured output suitably exploited. Based on the relation, we further propose several directly-verifiable conditions of achieving almost sure exponential stability for the NSSs with scheduling-based control via try-once-discard protocol, the most typical one among dynamic scheduling protocols. Particularly, incorporating the underlying positive effect of the stochastic noise, one of the conditions allows the Lyapunov function candidates to have non-negative infinitesimal under corresponding continuous-time control, while no counterpart exists in the non-stochastic setting.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"175 ","pages":"Article 112192"},"PeriodicalIF":4.8,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143394776","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"On the detection of Markov decision processes","authors":"Xiaoming Duan ,&nbsp;Yagiz Savas ,&nbsp;Rui Yan ,&nbsp;Zhe Xu ,&nbsp;Ufuk Topcu","doi":"10.1016/j.automatica.2025.112196","DOIUrl":"10.1016/j.automatica.2025.112196","url":null,"abstract":"<div><div>We study the detection problem for a finite set of Markov decision processes (MDPs) where the MDPs have the same state and action spaces but possibly different probabilistic transition functions. Any one of these MDPs could be the model for some underlying controlled stochastic process, but it is unknown a priori which MDP is the ground truth. We investigate whether it is possible to asymptotically detect the ground truth MDP model perfectly based on a single observed history (state–action sequence). Since the generation of histories depends on the policy adopted to control the MDPs, we discuss the existence and synthesis of policies that allow for perfect detection. We start with the case of two MDPs and establish a necessary and sufficient condition for the existence of policies that lead to perfect detection. Based on this condition, we then develop an algorithm that efficiently (in time polynomial in the size of the MDPs) determines the existence of policies and synthesizes one when they exist. We further extend the results to the more general case where there are more than two MDPs in the candidate set, and we develop a policy synthesis algorithm based on the breadth-first search and recursion. We demonstrate the effectiveness of our algorithms through numerical examples.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"175 ","pages":"Article 112196"},"PeriodicalIF":4.8,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143387820","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Simultaneous distributed localization and formation tracking control via matrix-weighted position constraints","authors":"Xu Fang ,&nbsp;Lihua Xie ,&nbsp;Dimos V. Dimarogonas","doi":"10.1016/j.automatica.2025.112188","DOIUrl":"10.1016/j.automatica.2025.112188","url":null,"abstract":"<div><div>This paper studies the problem of 3-D relative-measurement-based leader–follower simultaneous distributed localization and formation tracking control. The position information is only available to the leaders, and the followers have inter-agent relative measurements and communication with their neighbors. The key contribution is the development of a weight-matrix-based position constraint, which can make use of relative measurements such as bearing, ratio-of-distance, angle, distance, relative position and their mixture to describe the position relationship among each follower and its neighbors in 3-D space. A bearing-based distributed protocol is proposed for each follower to estimate its position and track its target position, which can drive the followers from their unlocalizable positions to localizable positions. The proposed algorithm is then extended to the case that both bearing and ratio-of-distance measurements are available, where the followers are localizable at all times if the followers and their neighbors are not collocated. In addition, the proposed method is also applicable to homogeneous or heterogeneous angle, distance, and relative position measurements as the ratio-of-distances or bearings can be obtained indirectly by these relative measurements. A remarkable advantage is that the proposed method can be implemented without persistently exciting motions. Some illustrative simulations are presented to verify the theoretical results.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"175 ","pages":"Article 112188"},"PeriodicalIF":4.8,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143387819","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Direct data-driven discounted infinite horizon linear quadratic regulator with robustness guarantees","authors":"Ramin Esmzad,&nbsp;Hamidreza Modares","doi":"10.1016/j.automatica.2025.112197","DOIUrl":"10.1016/j.automatica.2025.112197","url":null,"abstract":"<div><div>This paper presents a one-shot learning approach with performance and robustness guarantees for the linear quadratic regulator (LQR) control of stochastic linear systems. Even though data-based LQR control has been widely considered, existing results suffer either from data hungriness due to the inherently iterative nature of the optimization formulation (e.g., value learning or policy gradient reinforcement learning algorithms) or from a lack of robustness guarantees in one-shot non-iterative algorithms. To avoid data hungriness while ensuing robustness guarantees, an adaptive dynamic programming formalization of the LQR is presented that relies on solving a Bellman inequality. The control gain and the value function are directly learned by using a control-oriented approach that characterizes the closed-loop system using data and a decision variable from which the control is obtained. This closed-loop characterization is noise-dependent. The effect of the closed-loop system noise on the Bellman inequality is considered to ensure both robust stability and suboptimal performance despite ignoring the measurement noise. To ensure robust stability, it is shown that this system characterization leads to a closed-loop system with multiplicative and additive noise, enabling the application of distributional robust control techniques. The analysis of the suboptimality gap reveals that robustness can be achieved by construction without the need for regularization or parameter tuning. The simulation results on the active car suspension problem demonstrate the superiority of the proposed method in terms of robustness and performance gap compared to existing methods.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"175 ","pages":"Article 112197"},"PeriodicalIF":4.8,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143394777","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"General multi-step value iteration for optimal learning control","authors":"Ding Wang ,&nbsp;Jiangyu Wang ,&nbsp;Derong Liu ,&nbsp;Junfei Qiao","doi":"10.1016/j.automatica.2025.112168","DOIUrl":"10.1016/j.automatica.2025.112168","url":null,"abstract":"<div><div>Learning control methods have been widely enhanced by reinforcement learning, but it is challenging to analyze the effects of incorporating extra system information. This paper presents a novel multi-step framework that utilizes extra multi-step system information to solve optimal control problems. Within this framework, we establish and classify general multi-step value iteration (MsVI) algorithms based on the uniformity between policy evaluation and improvement stages. According to this uniformity concept, the convergence condition and the acceleration conclusion are analyzed for different kinds of MsVI algorithms. Besides, we introduce a swarm policy optimizer to relieve limitations of the traditional gradient optimizer. Specifically, we implement general MsVI using an actor–critic scheme, where the swarm optimizer and neural networks are employed for policy improvement and evaluation, respectively. Furthermore, the approximation error caused by the approximator is also considered to verify the advantage of using multi-step system information. Finally, we apply the proposed method to a nonlinear benchmark system, demonstrating superior learning ability and control performance compared to traditional methods.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"175 ","pages":"Article 112168"},"PeriodicalIF":4.8,"publicationDate":"2025-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143387817","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}