Automatica最新文献

{"title":"Deep-learning based KKL chain observer for discrete-time nonlinear systems with time-varying output delay","authors":"Yasmine Marani ,&nbsp;Ibrahima N’Doye ,&nbsp;Taous Meriem Laleg-Kirati","doi":"10.1016/j.automatica.2024.111955","DOIUrl":"10.1016/j.automatica.2024.111955","url":null,"abstract":"<div><div>This paper proposes a Kazantzis–Kravaris–Luenberger (KKL) observer design for discrete-time nonlinear systems whose output is affected by a time-varying measurement delay. Relying on an injective state transformation, a chain of observers is designed in the latent coordinates with exponential stability guarantees through the inverse map in the original coordinates. Moreover, the relationship between the number of sub-predictors and the lower and upper bounds of the delay is derived. The transformations involved in the design of the KKL observer are identified using an unsupervised learning-based approach that relies on neural networks. A disturbance rejection and robustness analysis against measurement noise and neural network approximation error are presented, respectively. Finally, we illustrate the performance and robustness of the proposed learning-based design KKL chain observer through numerical simulations.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"171 ","pages":"Article 111955"},"PeriodicalIF":4.8,"publicationDate":"2024-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142356667","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":"Variance-reduced reshuffling gradient descent for nonconvex optimization: Centralized and distributed algorithms","authors":"Xia Jiang ,&nbsp;Xianlin Zeng ,&nbsp;Lihua Xie ,&nbsp;Jian Sun ,&nbsp;Jie Chen","doi":"10.1016/j.automatica.2024.111954","DOIUrl":"10.1016/j.automatica.2024.111954","url":null,"abstract":"<div><div>Nonconvex finite-sum optimization plays a crucial role in signal processing and machine learning, fueling the development of numerous centralized and distributed stochastic algorithms. However, existing stochastic optimization algorithms often suffer from high stochastic gradient variance due to the use of random sampling with replacement. To address this issue, this paper introduces an explicit variance-reduction step and proposes variance-reduced reshuffling gradient algorithms with a sampling-without-replacement scheme. Specifically, this paper proves that the proposed centralized variance-reduced reshuffling gradient algorithm (VR-RG) with constant step sizes converges to a stationary point for nonconvex optimization under the Kurdyka–Łojasiewicz condition. Moreover, for nonconvex optimization over connected multi-agent networks, the proposed distributed variance-reduced reshuffling gradient algorithm (DVR-RG) converges to a neighborhood of stationary points, where the neighborhood can be made arbitrarily small under mild conditions. Notably, the proposed DVR-RG requires only one communication round at each epoch. Finally, numerical simulations demonstrate the efficiency of the proposed algorithms.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"171 ","pages":"Article 111954"},"PeriodicalIF":4.8,"publicationDate":"2024-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142356820","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":"Two-agent connectivity maintenance and collision avoidance under two relationships","authors":"Bofan Wu ,&nbsp;Zhaoxia Peng ,&nbsp;Guoguang Wen ,&nbsp;Shichun Yang ,&nbsp;Xiaoqin Zhai ,&nbsp;Tingwen Huang","doi":"10.1016/j.automatica.2024.111959","DOIUrl":"10.1016/j.automatica.2024.111959","url":null,"abstract":"<div><div>In this paper, we study a two-agent tracking problem with the connectivity maintenance and collision avoidance requirements (CM–CARs) under two relationships. One is the leader–follower relationship, where the leader just focuses on its tracking goal, but the follower still needs to ensure the CM–CARs for the leader. Since the feasible space of the follower is a nonconvex ring, it is inconvenient to search for a feasible control directly. Thus, we split the original nonconvex region and search for a convex one by a homeomorphism mapping. Under the leader–follower relationship, a control scheme is proposed for the follower in the original feasible region. It is transformed from a control protocol that ensures the follower satisfies the CM–CARs in the mapping convex region. The other is the colleague relationship, where two agents both need to satisfy the CM–CARs and their control strategies are interactive. Thus, we propose an optimization-based strategy to achieve the tracking mission for the two-agent system, which satisfies the CM–CARs. Finally, we provide some simulation results to verify the proposed strategies.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"171 ","pages":"Article 111959"},"PeriodicalIF":4.8,"publicationDate":"2024-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142356664","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":"Sufficient conditions for robust safety in differential inclusions using barrier functions","authors":"Mohamed Adlene Maghenem ,&nbsp;Masoumeh Ghanbarpour Mamaghani ,&nbsp;Adnane Saoud","doi":"10.1016/j.automatica.2024.111938","DOIUrl":"10.1016/j.automatica.2024.111938","url":null,"abstract":"<div><div>This paper considers <em>robust-safety</em> notions for differential inclusions. A general framework is proposed to certify the considered notions in terms of barrier functions. While most existing literature has focused solely on what we designate as <em>uniform robust safety</em>, in this paper, we make a clear distinction between uniform and non-uniform robust safety. For both cases, we establish sufficient conditions on the nominal (unperturbed) system. Our conditions involve only the barrier function and the system’s right-hand side. Our results allow for unbounded safety regions as well as non-smooth barrier functions. The paper concludes with applications to safety-critical self-triggered control and fast-slow dynamics as well as assume–guarantee contracts.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"171 ","pages":"Article 111938"},"PeriodicalIF":4.8,"publicationDate":"2024-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142356711","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":"Non-singular cooperative guiding vector field under a homotopy equivalence transformation","authors":"Zirui Chen ,&nbsp;Zongyu Zuo","doi":"10.1016/j.automatica.2024.111962","DOIUrl":"10.1016/j.automatica.2024.111962","url":null,"abstract":"<div><div>The present article introduces the concept of a non-singular cooperative guiding vector field under a homotopy equivalence transformation. Initially, an innovative non-singular guiding vector field is derived, which is capable of guiding the controlled robot to follow a path in a moving frame. Furthermore, the properties of homotopy equivalence transformation are explored to establish the existence of the aforementioned vector field. Subsequently, a cooperative vector field, based on an extensive parameter consensus protocol, is introduced, including an in-depth analysis of the impact of vector field parameters. Then, the practical implementation of this innovative vector field is demonstrated through its application to cooperative path following in both 2-D and 3-D scenarios, showcasing its effectiveness. Finally, for a non-holonomic robot at constant speed, a moving path following control law based on the guiding vector field is presented, along with a discussion on speed constraint.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"171 ","pages":"Article 111962"},"PeriodicalIF":4.8,"publicationDate":"2024-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142356911","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":"The stabilization problem for a class of discrete-time semi-Markov jump singular systems","authors":"Bao Wang ,&nbsp;Quanxin Zhu","doi":"10.1016/j.automatica.2024.111960","DOIUrl":"10.1016/j.automatica.2024.111960","url":null,"abstract":"<div><div>This paper studies the stabilization problem for a class of discrete-time semi-Markov jump singular systems with mode-dependent singular matrix. Since our model is mode-dependent singular matrix, the previous methods are unable to achieve the stability. To overcome this difficulty, the novel and less conservative stability conditions with the original system coefficient matrices are constructed by developing new theories and methods. Based on the above conditions, the linear matrix inequality (LMI) based stabilizing controller design method is proposed.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"171 ","pages":"Article 111960"},"PeriodicalIF":4.8,"publicationDate":"2024-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142356821","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":"Unknown input filtering under full accessibility attacks","authors":"Jianqi Chen","doi":"10.1016/j.automatica.2024.111966","DOIUrl":"10.1016/j.automatica.2024.111966","url":null,"abstract":"<div><div>This study initially addresses the state estimation problem for discrete-time linear time-invariant systems under the influence of both exogenous attacks and random noises. As the filtering side, we make no prior assumptions or have any prior knowledge about the nature of attacks. We employ a specific unknown input filtering approach, which has been examined in prior research, to simultaneously estimate both the system states and attacks. Differing from existing works, our emphasis on full accessibility attacks, a particular class of unknown inputs, reveals that the dynamic estimation gains of the adopted filtering reduce directly to static ones. The second contribution of this study is to mitigate the influence of attacks by utilizing the estimated attack signals. Under this strategy, the deviation between nominal states and attacked states is characterized by evaluating the upper bound of the covariance matrix of the errors.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"171 ","pages":"Article 111966"},"PeriodicalIF":4.8,"publicationDate":"2024-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142356665","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":"Robust output regulation for multi-dimensional heat equation under boundary control","authors":"Ren-Xi Zhao ,&nbsp;Bao-Zhu Guo ,&nbsp;Lassi Paunonen","doi":"10.1016/j.automatica.2024.111956","DOIUrl":"10.1016/j.automatica.2024.111956","url":null,"abstract":"<div><div>Over the past few years, the internal model principle has been extended to some systems described by one-dimensional partial differential equations (PDEs) from the PDE perspective. However, robustness has remained limited to specific cases, primarily due to the challenges in formulating it within the PDE framework. In this paper, we explore output regulation for a multi-dimensional heat equation under boundary control, where the output space is infinite-dimensional. We not only derive an analytic tracking error feedback control but also demonstrate robustness. This is achieved by leveraging abstract results and PDE design techniques.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"171 ","pages":"Article 111956"},"PeriodicalIF":4.8,"publicationDate":"2024-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142356668","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":"Robust global optimization on smooth compact manifolds via hybrid gradient-free dynamics","authors":"Daniel E. Ochoa,&nbsp;Jorge I. Poveda","doi":"10.1016/j.automatica.2024.111916","DOIUrl":"10.1016/j.automatica.2024.111916","url":null,"abstract":"<div><div>It is well known that smooth autonomous dynamical systems modeled by ordinary differential equations (ODEs) cannot robustly and globally stabilize a point on compact, boundaryless manifolds. This obstruction, which is topological in nature, has significant implications for optimization problems, rendering traditional continuous-time algorithms incapable of robustly solving <em>global</em> optimization problems in such spaces. In turn, <em>gradient-free</em> optimization algorithms, which usually inherit their stability and convergence properties from their gradient-based counterparts, can also suffer from similar topological obstructions. For instance, this is the case in zeroth-order methods and perturbation-based techniques, where gradients and Hessian matrices are usually estimated in real-time via measurements or evaluations of the cost function. To address this problem, in this paper we introduce a novel class of <em>hybrid gradient-free optimization dynamics</em> that combine continuous-time and discrete-time feedback to overcome the obstructions that emerge in traditional ODE-based optimization algorithms evolving on smooth compact manifolds. The proposed hybrid dynamics switch between different gradient-free feedback-laws obtained by applying suitable exploratory <em>geodesic dithers</em> to a family of synergistic diffeomorphisms adapted to the cost function that defines the optimization problem. The use of geodesic dithers enables a suitable exploration of the manifold while simultaneously preserving its forward invariance, a property that is fundamental for many practical applications with physics-based constraints. The hybrid dynamics exploit the information obtained from the geodesic dithers to achieve robust global practical stability of the set of minimizers of the cost function. This stabilization is achieved without having direct access to the gradients of the cost functions, but rather using only real-time and continuous evaluations of the cost. Examples and numerical results are presented to illustrate the main ideas and advantages of the method.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"171 ","pages":"Article 111916"},"PeriodicalIF":4.8,"publicationDate":"2024-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142356710","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":"Accelerated optimization landscape of linear–quadratic regulator","authors":"Lechen Feng,&nbsp;Yuan-Hua Ni","doi":"10.1016/j.automatica.2024.111927","DOIUrl":"10.1016/j.automatica.2024.111927","url":null,"abstract":"<div><div>This paper introduces an accelerated optimization framework of handling the linear–quadratic regulator (LQR) problem. Firstly, a Lipschitz Hessian property of LQR cost is presented, which turns out to be a crucial property for the application of modern optimization techniques. Secondly, a Nesterov-type method with a restarting rule is proposed for state-feedback LQR problem, which can converge exponentially to the optimal feedback gain with Nesterov-optimal order <span><math><mrow><mn>1</mn><mo>−</mo><mn>1</mn><mo>/</mo><msqrt><mrow><mi>κ</mi></mrow></msqrt></mrow></math></span>. Thirdly, a Hessian-free two-procedure accelerated framework is proposed for output-feedback LQR problem, which can find an <span><math><mi>ϵ</mi></math></span>-stationary point with second-order guarantee.</div></div>","PeriodicalId":55413,"journal":{"name":"Automatica","volume":"171 ","pages":"Article 111927"},"PeriodicalIF":4.8,"publicationDate":"2024-09-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142356822","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}