Pub Date : 2025-10-08DOI: 10.1109/TSMC.2025.3614262
Kun Li;Zhipeng L端;Junwen Ding;Zhouxing Su;Qingyun Zhang
This article studies the two-dimensional (2-D) rectangle packing area minimization problem (RPAMP), a key subproblem in floor planning for very large-scale integration (VLSI) chip design. The goal of RPAMP is to orthogonally pack a set of rectangles into a variable-sized rectangular container without overlap, while minimizing the area of the container. By transforming the original problem into a series of 2-D strip packing problems (2DSPs), we propose a two-stage adaptive search algorithm (TS-ASA) to tackle the RPAMP. TS-ASA incorporates several distinctive features: First, a new candidate width pruning strategy is introduced, which limits the number of rectangles used for width combinations, thus reducing the search space. Second, the packing process is divided into two stages, with distinct scoring rules for each stage to optimize space utilization. Additionally, a multirestart strategy is employed to identify the appropriate switching point for the scoring rules. Tested on 39 public benchmark instances and compared with existing state-of-the-art algorithms, TS-ASA improves the best-known solutions for 27 instances and matches the best results for two instances. The experimental results demonstrate the effectiveness and efficiency of the proposed TS-ASA.
{"title":"A Two-Stage Adaptive Search Algorithm for the 2-D Rectangle Packing Area Minimization Problem","authors":"Kun Li;Zhipeng L端;Junwen Ding;Zhouxing Su;Qingyun Zhang","doi":"10.1109/TSMC.2025.3614262","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3614262","url":null,"abstract":"This article studies the two-dimensional (2-D) rectangle packing area minimization problem (RPAMP), a key subproblem in floor planning for very large-scale integration (VLSI) chip design. The goal of RPAMP is to orthogonally pack a set of rectangles into a variable-sized rectangular container without overlap, while minimizing the area of the container. By transforming the original problem into a series of 2-D strip packing problems (2DSPs), we propose a two-stage adaptive search algorithm (TS-ASA) to tackle the RPAMP. TS-ASA incorporates several distinctive features: First, a new candidate width pruning strategy is introduced, which limits the number of rectangles used for width combinations, thus reducing the search space. Second, the packing process is divided into two stages, with distinct scoring rules for each stage to optimize space utilization. Additionally, a multirestart strategy is employed to identify the appropriate switching point for the scoring rules. Tested on 39 public benchmark instances and compared with existing state-of-the-art algorithms, TS-ASA improves the best-known solutions for 27 instances and matches the best results for two instances. The experimental results demonstrate the effectiveness and efficiency of the proposed TS-ASA.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 12","pages":"9120-9132"},"PeriodicalIF":8.7,"publicationDate":"2025-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145546982","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Personalized federated learning (PFL) has emerged as an efficient way to tailor global models to individual data features. However, in real-world deployments, the dynamic nature of client participation—where clients may join or quit arbitrarily—poses significant challenges for PFL systems. We propose elastic PFL (EPFL), an EPFL framework that accommodates dynamic client participation while preserving model performance. Our framework addresses two fundamental challenges: 1) seamlessly generating personalized models for newly joining unlabeled clients and 2) achieving efficient and precise multigrain unlearning for quitting clients. To address these challenges, we utilize hypernetworks for client-specific model generation and introduce a novel client embedding regularization (CER) to enhance generalization capabilities for new clients. Furthermore, we design an adversarial purification mechanism that enables efficient multigrained unlearning at the client, sample, and class levels. We provide theoretical analysis establishing both generalization bounds for new clients and convergence guarantees for the unlearning process. Extensive experimental results demonstrate that EPFL outperforms the existing PFL methods in terms of both accuracy and generalization capabilities. Through backdoor trigger experiments, we show that EPFL achieves unlearning efficacy and fidelity comparable to complete retraining while requiring only 3%–10% of the computational time of these existing approaches.
{"title":"EPFL: Toward Elastic Personalized Federated Learning With Seamless Client Joining and Quitting","authors":"Yuange Liu;Daobin Luo;Weishan Zhang;Wei Jiang;Chaoqun Zheng;Xiaohui Sun;Yuru Liu;Qiao Qiao;Tao Chen;Hongwei Zhao;Su Yang;Fei-Yue Wang","doi":"10.1109/TSMC.2025.3613624","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3613624","url":null,"abstract":"Personalized federated learning (PFL) has emerged as an efficient way to tailor global models to individual data features. However, in real-world deployments, the dynamic nature of client participation—where clients may join or quit arbitrarily—poses significant challenges for PFL systems. We propose elastic PFL (EPFL), an EPFL framework that accommodates dynamic client participation while preserving model performance. Our framework addresses two fundamental challenges: 1) seamlessly generating personalized models for newly joining unlabeled clients and 2) achieving efficient and precise multigrain unlearning for quitting clients. To address these challenges, we utilize hypernetworks for client-specific model generation and introduce a novel client embedding regularization (CER) to enhance generalization capabilities for new clients. Furthermore, we design an adversarial purification mechanism that enables efficient multigrained unlearning at the client, sample, and class levels. We provide theoretical analysis establishing both generalization bounds for new clients and convergence guarantees for the unlearning process. Extensive experimental results demonstrate that EPFL outperforms the existing PFL methods in terms of both accuracy and generalization capabilities. Through backdoor trigger experiments, we show that EPFL achieves unlearning efficacy and fidelity comparable to complete retraining while requiring only 3%–10% of the computational time of these existing approaches.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 12","pages":"9324-9338"},"PeriodicalIF":8.7,"publicationDate":"2025-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145546964","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The all-wheel steering technology enhances vehicle maneuverability, while a modular, plug-and-play chassis electrical architecture facilitates the integration of more electronic control units. Hence, this article aims to improve vehicle handling stability performance by integrating the active front-wheel steering system (AFS) and the active rear-wheel steering system (ARS) within a multiagent game theory framework. First, a vehicle dynamics model integrating AFS and ARS control, along with system uncertainties, is developed. Numerical simulations are performed to validate the effectiveness of integrated ARS in enhancing vehicle handling stability performance. Then, a multiagent system (MAS) framework is constructed to achieve coordinated control between AFS and ARS based on distributed model predictive control (DMPC), with interactive behaviors among agents defined. Furthermore, cooperative game theory is introduced to find optimal solutions. To guarantee the system’s stability and antidisturbance capabilities, robust compensation control and terminal constraints are designed. The MAS framework can also address the development requirements for modularization in chassis system design. Finally, we validated the effectiveness of the proposed method through experiments conducted under several typical test conditions. The comparative tests demonstrate that the MAS framework can improve the vehicle’s handling stability.
{"title":"Robust Game-Theory Control for All-Wheel Steering to Enhance Vehicle Handling Stability Performance","authors":"Jinhao Liang;Cheng Shen;Xin Xia;Dawei Pi;Guodong Yin","doi":"10.1109/TSMC.2025.3615943","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3615943","url":null,"abstract":"The all-wheel steering technology enhances vehicle maneuverability, while a modular, plug-and-play chassis electrical architecture facilitates the integration of more electronic control units. Hence, this article aims to improve vehicle handling stability performance by integrating the active front-wheel steering system (AFS) and the active rear-wheel steering system (ARS) within a multiagent game theory framework. First, a vehicle dynamics model integrating AFS and ARS control, along with system uncertainties, is developed. Numerical simulations are performed to validate the effectiveness of integrated ARS in enhancing vehicle handling stability performance. Then, a multiagent system (MAS) framework is constructed to achieve coordinated control between AFS and ARS based on distributed model predictive control (DMPC), with interactive behaviors among agents defined. Furthermore, cooperative game theory is introduced to find optimal solutions. To guarantee the system’s stability and antidisturbance capabilities, robust compensation control and terminal constraints are designed. The MAS framework can also address the development requirements for modularization in chassis system design. Finally, we validated the effectiveness of the proposed method through experiments conducted under several typical test conditions. The comparative tests demonstrate that the MAS framework can improve the vehicle’s handling stability.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 12","pages":"9230-9241"},"PeriodicalIF":8.7,"publicationDate":"2025-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145546989","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-07DOI: 10.1109/TSMC.2025.3614484
Alexander Diedrich;Mattias Krysander;Rene Heesch;Oliver Niggemann
Existing algorithms for consistency-based fault diagnosis are sound and complete according to some correct logical model. But obtaining a good model is the crucial and difficult part. Originally, the classical diagnosis algorithms were developed to analyze Boolean circuits, such that simple propositional or predicate logic models were sufficient to express the structure and behavior. Cyber–physical systems, however, exhibit hybrid behavior, meaning they generate continuous and discrete values that need to be interpreted. This adds significant complexity. Furthermore, modern cyber–physical systems may change their structure over their lifetime and thus require model adaptations. This article presents a novel formalism to model cyber–physical systems for consistency-based fault diagnosis. Drawing from the research fields of artificial intelligence and control theory, the approach models system structure and behavior through the use of satisfiability modulo nonlinear arithmetic. The approach proves advantageous compared to previous modeling techniques through its integration of nonlinear behavior models, implicit computation of residual values to compute fault symptoms, its representation of different operating modes of the system, and its integration with existing sound and complete fault diagnosis algorithms. The approach was validated empirically using two benchmarks from the process industry, a simulation of battery packs, and Boolean standard circuits. Throughout all experiments, an accuracy of 97% was achieved.
{"title":"Modeling Cyber–Physical Systems for Fault Diagnosis","authors":"Alexander Diedrich;Mattias Krysander;Rene Heesch;Oliver Niggemann","doi":"10.1109/TSMC.2025.3614484","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3614484","url":null,"abstract":"Existing algorithms for consistency-based fault diagnosis are sound and complete according to some correct logical model. But obtaining a good model is the crucial and difficult part. Originally, the classical diagnosis algorithms were developed to analyze Boolean circuits, such that simple propositional or predicate logic models were sufficient to express the structure and behavior. Cyber–physical systems, however, exhibit hybrid behavior, meaning they generate continuous and discrete values that need to be interpreted. This adds significant complexity. Furthermore, modern cyber–physical systems may change their structure over their lifetime and thus require model adaptations. This article presents a novel formalism to model cyber–physical systems for consistency-based fault diagnosis. Drawing from the research fields of artificial intelligence and control theory, the approach models system structure and behavior through the use of satisfiability modulo nonlinear arithmetic. The approach proves advantageous compared to previous modeling techniques through its integration of nonlinear behavior models, implicit computation of residual values to compute fault symptoms, its representation of different operating modes of the system, and its integration with existing sound and complete fault diagnosis algorithms. The approach was validated empirically using two benchmarks from the process industry, a simulation of battery packs, and Boolean standard circuits. Throughout all experiments, an accuracy of 97% was achieved.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 12","pages":"9266-9279"},"PeriodicalIF":8.7,"publicationDate":"2025-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11194731","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145546984","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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}
Pub Date : 2025-10-07DOI: 10.1109/TSMC.2025.3615639
Fang Liu;Deyuan Meng;Qiang Song
This article is aimed at solving the distributed control problem for a leaderless heterogeneous network with signed digraph topology and locally Lipschitz nonlinear dynamics, where the agents may have nonidentical nodal functions and different state dimensions. To realize the control objectives of the heterogeneous nonlinear signed network, a distributed iterative learning control (ILC) algorithm is developed based on the relative outputs of local neighbors from the previous iteration. Despite the difficulties induced by nodal heterogeneities and nonlinearities, it is shown that the outputs of leaderless heterogeneous signed network under the developed ILC algorithm can exhibit the typical behaviors of conventional signed networks over a finite time period. To be more specific, the heterogeneous nonlinear signed network achieves either bipartite or interval bipartite output consensus (respectively, global output stability) along the iteration axis when the subgraph formed by root nodes is structurally balanced (respectively, unbalanced). Numerical studies are conducted to illustrate the proposed ILC scheme.
{"title":"Distributed Iterative Learning Control of Leaderless Heterogeneous Nonlinear Signed Networks","authors":"Fang Liu;Deyuan Meng;Qiang Song","doi":"10.1109/TSMC.2025.3615639","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3615639","url":null,"abstract":"This article is aimed at solving the distributed control problem for a leaderless heterogeneous network with signed digraph topology and locally Lipschitz nonlinear dynamics, where the agents may have nonidentical nodal functions and different state dimensions. To realize the control objectives of the heterogeneous nonlinear signed network, a distributed iterative learning control (ILC) algorithm is developed based on the relative outputs of local neighbors from the previous iteration. Despite the difficulties induced by nodal heterogeneities and nonlinearities, it is shown that the outputs of leaderless heterogeneous signed network under the developed ILC algorithm can exhibit the typical behaviors of conventional signed networks over a finite time period. To be more specific, the heterogeneous nonlinear signed network achieves either bipartite or interval bipartite output consensus (respectively, global output stability) along the iteration axis when the subgraph formed by root nodes is structurally balanced (respectively, unbalanced). Numerical studies are conducted to illustrate the proposed ILC scheme.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 12","pages":"9242-9254"},"PeriodicalIF":8.7,"publicationDate":"2025-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145546992","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-06DOI: 10.1109/TSMC.2025.3612577
Wang Yang;Jiuxiang Dong;Fanwei Meng;Ju H. Park
This article investigates the leader–follower consensus tracking problem for Lipschitz nonlinear multiagent systems (MASs) with unmeasurable states. To address limitations in communication and sensing resources, a hybrid triggering consensus framework is proposed to reduce redundant transmissions and measurements. First, a local observer is designed to estimate the system state, where output sampling and transmission to the observer are determined by a time-triggered mechanism (TTM). Then, based on the observer and two nonlinear estimators, a fully distributed controller is developed. Communication among neighboring agents is governed by a hybrid triggering mechanism (HTM) that integrates both time-triggered and event-triggered conditions. Notably, the proposed strategy ensures a positive minimum triggering interval (PMTI). The effectiveness of the proposed method is validated on a single-link flexible joint manipulator.
{"title":"A Hybrid Triggering Consensus Framework to Nonlinear Multiagent Systems With Local Observer","authors":"Wang Yang;Jiuxiang Dong;Fanwei Meng;Ju H. Park","doi":"10.1109/TSMC.2025.3612577","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3612577","url":null,"abstract":"This article investigates the leader–follower consensus tracking problem for Lipschitz nonlinear multiagent systems (MASs) with unmeasurable states. To address limitations in communication and sensing resources, a hybrid triggering consensus framework is proposed to reduce redundant transmissions and measurements. First, a local observer is designed to estimate the system state, where output sampling and transmission to the observer are determined by a time-triggered mechanism (TTM). Then, based on the observer and two nonlinear estimators, a fully distributed controller is developed. Communication among neighboring agents is governed by a hybrid triggering mechanism (HTM) that integrates both time-triggered and event-triggered conditions. Notably, the proposed strategy ensures a positive minimum triggering interval (PMTI). The effectiveness of the proposed method is validated on a single-link flexible joint manipulator.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 12","pages":"9084-9096"},"PeriodicalIF":8.7,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145546981","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-06DOI: 10.1109/TSMC.2025.3613350
Man Gao;Qinghua Zhang;Fan Zhao;Qin Xie;Guoyin Wang;Weiping Ding
The shadowed set, as a three-way approximation model for a fuzzy set, has been extensively studied in model construction, theoretical analysis, data analysis, and applications. However, the current research on the construction of a shadowed set is all based on a single attribute to complete the approximate partition of a simple target concept, i.e., single-granularity-layer space. Multiple attributes are not considered comprehensively to achieve an approximate partition of the complex target concept, i.e., multigranularity-layer space. In addition, the current evaluation criteria of the shadowed set are all reasonable explanations for model construction and threshold determination, lacking an effective evaluation of approximate partition results. Therefore, the multigranularity-layer shadowed set (MGLSS) is proposed in this article, which aims to extend the construction of the shadowed set from single-granularity-layer space to multigranularity-layer space. First, MGLSS is analyzed based on information systems and divided into two models: optimistic-MGLSS (OPT-MGLSS) and pessimistic-MGLSS (PES-MGLSS). Second, the expression form, threshold selection, semantic interpretation, partition rules, basic mathematical theorems, and the definition of fusion operators of MGLSS are analyzed and discussed. Third, two evaluation criteria of coverage and accuracy of approximate partition results are proposed. Finally, six cases are analyzed to illustrate the application scenarios of MGLSS, and one instance and algorithm analysis are given to demonstrate the construction steps, and through the real datasets experiment and statistical hypothesis testing analysis, to provide an objective and quantitative scientific basis for the research conclusion. The experimental results demonstrate the validity and rationality of the MGLSS construction mechanism.
{"title":"Multigranularity-Layer Shadowed Set: A Three-Way Approximation Framework for Fuzzy Information","authors":"Man Gao;Qinghua Zhang;Fan Zhao;Qin Xie;Guoyin Wang;Weiping Ding","doi":"10.1109/TSMC.2025.3613350","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3613350","url":null,"abstract":"The shadowed set, as a three-way approximation model for a fuzzy set, has been extensively studied in model construction, theoretical analysis, data analysis, and applications. However, the current research on the construction of a shadowed set is all based on a single attribute to complete the approximate partition of a simple target concept, i.e., single-granularity-layer space. Multiple attributes are not considered comprehensively to achieve an approximate partition of the complex target concept, i.e., multigranularity-layer space. In addition, the current evaluation criteria of the shadowed set are all reasonable explanations for model construction and threshold determination, lacking an effective evaluation of approximate partition results. Therefore, the multigranularity-layer shadowed set (MGLSS) is proposed in this article, which aims to extend the construction of the shadowed set from single-granularity-layer space to multigranularity-layer space. First, MGLSS is analyzed based on information systems and divided into two models: optimistic-MGLSS (OPT-MGLSS) and pessimistic-MGLSS (PES-MGLSS). Second, the expression form, threshold selection, semantic interpretation, partition rules, basic mathematical theorems, and the definition of fusion operators of MGLSS are analyzed and discussed. Third, two evaluation criteria of coverage and accuracy of approximate partition results are proposed. Finally, six cases are analyzed to illustrate the application scenarios of MGLSS, and one instance and algorithm analysis are given to demonstrate the construction steps, and through the real datasets experiment and statistical hypothesis testing analysis, to provide an objective and quantitative scientific basis for the research conclusion. The experimental results demonstrate the validity and rationality of the MGLSS construction mechanism.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 12","pages":"9201-9215"},"PeriodicalIF":8.7,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145546970","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-06DOI: 10.1109/TSMC.2025.3614997
Xinwei Cao;Yiguo Yang;Shuai Li;Vasilios N. Katsikis
The $k$ -winner-takes-all ($k$ -WTA) problem involves selecting the top $k$ agents with the highest inputs from a set of $n$ candidates. This problem plays a fundamental role in modeling competitive behaviors in social systems and economic environments. In this article, we propose a structurally simplified dynamic neural network to solve the $k$ -WTA problem efficiently. The original $k$ -WTA task is first reformulated as a constrained quadratic programming (QP) problem. A smooth sigmoid function is then introduced to encode inequality constraints implicitly, simplifying the representation. Based on this formulation, we develop a continuous-time neural dynamic model capable of solving the problem in real time. The proposed model is theoretically proven to achieve global convergence and optimality with respect to the $k$ -WTA solution. Extensive numerical experiments, including tests on real-world data, validate the effectiveness of the proposed approach, demonstrating fast convergence, robustness, and practical applicability.
{"title":"k-Winner-Take-All Competition Based on Novel Dynamic Neural Networks","authors":"Xinwei Cao;Yiguo Yang;Shuai Li;Vasilios N. Katsikis","doi":"10.1109/TSMC.2025.3614997","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3614997","url":null,"abstract":"The <inline-formula> <tex-math>$k$ </tex-math></inline-formula>-winner-takes-all (<inline-formula> <tex-math>$k$ </tex-math></inline-formula>-WTA) problem involves selecting the top <inline-formula> <tex-math>$k$ </tex-math></inline-formula> agents with the highest inputs from a set of <inline-formula> <tex-math>$n$ </tex-math></inline-formula> candidates. This problem plays a fundamental role in modeling competitive behaviors in social systems and economic environments. In this article, we propose a structurally simplified dynamic neural network to solve the <inline-formula> <tex-math>$k$ </tex-math></inline-formula>-WTA problem efficiently. The original <inline-formula> <tex-math>$k$ </tex-math></inline-formula>-WTA task is first reformulated as a constrained quadratic programming (QP) problem. A smooth sigmoid function is then introduced to encode inequality constraints implicitly, simplifying the representation. Based on this formulation, we develop a continuous-time neural dynamic model capable of solving the problem in real time. The proposed model is theoretically proven to achieve global convergence and optimality with respect to the <inline-formula> <tex-math>$k$ </tex-math></inline-formula>-WTA solution. Extensive numerical experiments, including tests on real-world data, validate the effectiveness of the proposed approach, demonstrating fast convergence, robustness, and practical applicability.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 12","pages":"9255-9265"},"PeriodicalIF":8.7,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145546980","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-06DOI: 10.1109/TSMC.2025.3614342
Yifan Luo;Xiaomei Wang;Kai Zhao;Ben Niu
In this article, we propose an event-triggered unified fault-tolerant control strategy for continuum robots with prescribed performance, considering multiple levels of actuator failures. First, by integrating a Kelvin–Voigt model, this article introduces an improved constant curvature model for continuum robots accounting for dissipative effects. Second, a comprehensive fault-handling framework is established, which combines extreme fault detection, dynamic actuator redundancy, and asymmetric performance recovery within the context of a unified prescribed performance. Third, by taking advantage of the dynamic characteristic of auxiliary variable and designing suitable event-triggering conditions, a dual-channel event-triggering mechanism is introduced to effectively reduce the number of triggers and optimize the utilization of communication resources. The proposed unified fault-tolerant control strategy ensures the boundedness of all signals in the closed-loop system and multiple kinds of prescribed performance behaviors under actuator failures. Simulation results validate the effectiveness of the proposed control strategy.
{"title":"Event-Triggered Unified Prescribed Performance Control for Continuum Robots With Actuator Faults","authors":"Yifan Luo;Xiaomei Wang;Kai Zhao;Ben Niu","doi":"10.1109/TSMC.2025.3614342","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3614342","url":null,"abstract":"In this article, we propose an event-triggered unified fault-tolerant control strategy for continuum robots with prescribed performance, considering multiple levels of actuator failures. First, by integrating a Kelvin–Voigt model, this article introduces an improved constant curvature model for continuum robots accounting for dissipative effects. Second, a comprehensive fault-handling framework is established, which combines extreme fault detection, dynamic actuator redundancy, and asymmetric performance recovery within the context of a unified prescribed performance. Third, by taking advantage of the dynamic characteristic of auxiliary variable and designing suitable event-triggering conditions, a dual-channel event-triggering mechanism is introduced to effectively reduce the number of triggers and optimize the utilization of communication resources. The proposed unified fault-tolerant control strategy ensures the boundedness of all signals in the closed-loop system and multiple kinds of prescribed performance behaviors under actuator failures. Simulation results validate the effectiveness of the proposed control strategy.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 12","pages":"9175-9185"},"PeriodicalIF":8.7,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145546987","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-06DOI: 10.1109/TSMC.2025.3614905
Chen Ding;Li Ma;Shihong Ding;Xinghuo Yu;Keqi Mei
In this article, a novel adaptive second-order sliding mode (ASOSM) control law is constructed for a general category of sliding mode control (SMC) systems with mismatched uncertainties, including a nonvanishing external disturbance. This innovative control design proposal is accomplished through three key mechanisms. First, the new sliding mode dynamics subject to mismatched uncertainties is derived by selecting the appropriate sliding variables, which can significantly increase the uncertainties existing in the control input channel and relax the strict requirement on the relative degree assumption of the sliding variable. Second, a novel ASOSM controller, which contains some adaptive parameters generated via a three-layer nested adaptive mechanism, is constructed by utilizing the modified adding power integrator (API) approach and the adaptive control technique. Third, the practical finite-time stability of the closed-loop sliding mode system is confirmed by means of the systematic Lyapunov stability theory. The technical advancement of the developed adaptive control scheme lies in its ability to effectively deal with a more general sliding mode dynamics containing multiple uncertainties and guarantee that the practical second-order sliding mode (SOSM) is established in a finite time. Finally, simulation results, incorporating a practical application case, are provided to illustrate the effectiveness of the designed adaptive control scheme.
{"title":"Adaptive Second-Order Sliding Mode Controller Design Subject to Mismatched Uncertainties","authors":"Chen Ding;Li Ma;Shihong Ding;Xinghuo Yu;Keqi Mei","doi":"10.1109/TSMC.2025.3614905","DOIUrl":"https://doi.org/10.1109/TSMC.2025.3614905","url":null,"abstract":"In this article, a novel adaptive second-order sliding mode (ASOSM) control law is constructed for a general category of sliding mode control (SMC) systems with mismatched uncertainties, including a nonvanishing external disturbance. This innovative control design proposal is accomplished through three key mechanisms. First, the new sliding mode dynamics subject to mismatched uncertainties is derived by selecting the appropriate sliding variables, which can significantly increase the uncertainties existing in the control input channel and relax the strict requirement on the relative degree assumption of the sliding variable. Second, a novel ASOSM controller, which contains some adaptive parameters generated via a three-layer nested adaptive mechanism, is constructed by utilizing the modified adding power integrator (API) approach and the adaptive control technique. Third, the practical finite-time stability of the closed-loop sliding mode system is confirmed by means of the systematic Lyapunov stability theory. The technical advancement of the developed adaptive control scheme lies in its ability to effectively deal with a more general sliding mode dynamics containing multiple uncertainties and guarantee that the practical second-order sliding mode (SOSM) is established in a finite time. Finally, simulation results, incorporating a practical application case, are provided to illustrate the effectiveness of the designed adaptive control scheme.","PeriodicalId":48915,"journal":{"name":"IEEE Transactions on Systems Man Cybernetics-Systems","volume":"55 12","pages":"9151-9164"},"PeriodicalIF":8.7,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145546962","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：