Pub Date : 2026-01-01DOI: 10.1016/j.isatra.2025.11.034
Yong Hao , Bo Cheng , Kuo Hu , Cheng Huang
Unmanned surface vehicle swarms (USVS), comprising multiple unmanned surface vehicles (USVs), serve as critical platforms for a wide range of demanding marine operations. However, coordinating USVS poses challenges relating to flock cohesion and communication, as well as the adverse effects of uncertainty. Within this context, this paper introduces an event-based adaptive flocking control law for USVS to address these issues. Firstly, an integrated dual-mode potential function, combining a novel artificial potential function (APF) and a directionally consistent potential function (DCPF), is designed to enhance flocking cohesion while guaranteeing connectivity. Then, a boundary-locked event-triggered (BLET) mechanism effectively reduces the communication load with computable minimum inter-event time (MIET) and maximum triggering interval time (MTIT). Finally, a reinforcement learning-based echo state network (RLESN) is incorporated to compensate for unmodeled dynamics and external disturbances, thereby significantly improving the system’s robustness. Simulation results and comparative analyses validate the effectiveness and advantages of the proposed methodology.
{"title":"Boundary-locked event-triggered mechanism-based adaptive flocking control for multi-USV systems","authors":"Yong Hao , Bo Cheng , Kuo Hu , Cheng Huang","doi":"10.1016/j.isatra.2025.11.034","DOIUrl":"10.1016/j.isatra.2025.11.034","url":null,"abstract":"<div><div>Unmanned surface vehicle swarms (USVS), comprising multiple unmanned surface vehicles (USVs), serve as critical platforms for a wide range of demanding marine operations. However, coordinating USVS poses challenges relating to flock cohesion and communication, as well as the adverse effects of uncertainty. Within this context, this paper introduces an event-based adaptive flocking control law for USVS to address these issues. Firstly, an integrated dual-mode potential function, combining a novel artificial potential function (APF) and a directionally consistent potential function (DCPF), is designed to enhance flocking cohesion while guaranteeing connectivity. Then, a boundary-locked event-triggered (BLET) mechanism effectively reduces the communication load with computable minimum inter-event time (MIET) and maximum triggering interval time (MTIT). Finally, a reinforcement learning-based echo state network (RLESN) is incorporated to compensate for unmodeled dynamics and external disturbances, thereby significantly improving the system’s robustness. Simulation results and comparative analyses validate the effectiveness and advantages of the proposed methodology.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"168 ","pages":"Pages 29-40"},"PeriodicalIF":6.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145784098","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}
Pub Date : 2026-01-01DOI: 10.1016/j.isatra.2025.09.040
Xiaodong Fu, Zhaokui Wang
This study analyses the space manipulation system's contact collision dynamics of capturing a target spacecraft, considering the influence of multiple vibrations in the base, joints, and links. The dynamic models of a flexible-base-joint-link space manipulation system and the target spacecraft are derived. The impact generated by the capture collision is examined. A dynamic model for the combined spacecraft system is established. Using singular perturbation theory, the model is decomposed into controllable subsystems. A motion-vibration hybrid controller is proposed, which incorporates an adaptive constrained neural network control based on the barrier Lyapunov function, hybrid trajectory method, and linear quadratic optimal control. The controller satisfies the torque-limited requirements, and its parameters are adaptively updated. Numerical examples are presented to substantiate the validity of the main results.
{"title":"Motion-vibration hybrid control of flexible-base-joint-link space manipulation system capturing target spacecraft using barrier Lyapunov function","authors":"Xiaodong Fu, Zhaokui Wang","doi":"10.1016/j.isatra.2025.09.040","DOIUrl":"10.1016/j.isatra.2025.09.040","url":null,"abstract":"<div><div>This study analyses the space manipulation system's contact collision dynamics of capturing a target spacecraft, considering the influence of multiple vibrations in the base, joints, and links. The dynamic models of a flexible-base-joint-link space manipulation system and the target spacecraft are derived. The impact generated by the capture collision is examined. A dynamic model for the combined spacecraft system is established. Using singular perturbation theory, the model is decomposed into controllable subsystems. A motion-vibration hybrid controller is proposed, which incorporates an adaptive constrained neural network control based on the barrier Lyapunov function, hybrid trajectory method, and linear quadratic optimal control. The controller satisfies the torque-limited requirements, and its parameters are adaptively updated. Numerical examples are presented to substantiate the validity of the main results.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"168 ","pages":"Pages 390-400"},"PeriodicalIF":6.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145524819","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}
Pub Date : 2026-01-01DOI: 10.1016/j.isatra.2025.11.019
Xin Li , Bingyin Wang , Jin Lv , Lixin Wei , Kai Ma
In industrial environments, substantial background noise often obscures the crucial impact and periodic features within bearing vibration signals, hindering the efficacy of conventional diagnostic methods. To address this challenge, this paper proposes a novel fault diagnosis framework for rolling bearings, initiated by an advanced signal enhancement stage using a time–frequency extension of adaptive filtering (TFEAF) method. This method effectively enhances the signal’s time–frequency representation, achieving robust denoising and revealing the underlying fault characteristics previously masked by noise. Subsequently, we introduce a novel Huge Kernel Convolutional Bidirectional Long Short-Term Memory Network with a Cross-Attention mechanism (HCBM-CANet). This network is specifically designed to extract deep spatial and temporal features from the denoised signal. The integrated cross-attention mechanism synergistically fuses these features, achieving significant boosts in diagnostic precision under complex operating conditions. Experimental results on the CWRU and Paderborn University datasets demonstrate the superiority of our approach. Under high-noise scenarios, the proposed TFEAF-HCBM-CANet framework achieves an average diagnostic accuracy exceeding 99 %, outperforming seven state-of-the-art network models. These findings validate the exceptional robustness and effectiveness of our method for processing signals heavily corrupted by noise.
{"title":"Adaptive time-frequency signal enhancement and deep spatial-temporal fusion for noise-resistant bearing fault diagnosis","authors":"Xin Li , Bingyin Wang , Jin Lv , Lixin Wei , Kai Ma","doi":"10.1016/j.isatra.2025.11.019","DOIUrl":"10.1016/j.isatra.2025.11.019","url":null,"abstract":"<div><div>In industrial environments, substantial background noise often obscures the crucial impact and periodic features within bearing vibration signals, hindering the efficacy of conventional diagnostic methods. To address this challenge, this paper proposes a novel fault diagnosis framework for rolling bearings, initiated by an advanced signal enhancement stage using a time–frequency extension of adaptive filtering (TFEAF) method. This method effectively enhances the signal’s time–frequency representation, achieving robust denoising and revealing the underlying fault characteristics previously masked by noise. Subsequently, we introduce a novel Huge Kernel Convolutional Bidirectional Long Short-Term Memory Network with a Cross-Attention mechanism (HCBM-CANet). This network is specifically designed to extract deep spatial and temporal features from the denoised signal. The integrated cross-attention mechanism synergistically fuses these features, achieving significant boosts in diagnostic precision under complex operating conditions. Experimental results on the CWRU and Paderborn University datasets demonstrate the superiority of our approach. Under high-noise scenarios, the proposed TFEAF-HCBM-CANet framework achieves an average diagnostic accuracy exceeding 99 %, outperforming seven state-of-the-art network models. These findings validate the exceptional robustness and effectiveness of our method for processing signals heavily corrupted by noise.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"168 ","pages":"Pages 284-304"},"PeriodicalIF":6.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145566654","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}
The hydraulically-driven lower limb exoskeleton robot (HDLLER) can enhance the motion capabilities of human lower limbs. However, due to inherent uncertainties and external human–robot coupling disturbances, it poses significant challenges to precise trajectory tracking control. Therefore, a novel funnel control (FC) strategy based on an unknown system dynamics estimator (USDE) is proposed to compensate for unknown internal and external dynamic influences, thus further improving robot performance. Firstly, the HDLLER high-order dynamic model containing human–robot coupling disturbances is transformed into a Brunovsky canonical form, effectively circumventing the ‘explosion of complexity’ problem in high order strict feedback systems. Subsequently, a set of high gain observers (HGO) is introduced to reconstruct the states of the transformed system. A modified funnel function is embedded in feedback controller to constrain the tracking error within prescribed boundaries, ensuring satisfactory transient and steady-state performances. Additionally, a simple yet effective USDE is incorporated to compensate for the lumped unknown dynamics. Different comparative simulations and gait experiments have verified the effectiveness of the proposed control method.
{"title":"Funnel control based on unknown system dynamics estimator for hydraulically-driven lower limb exoskeleton robot","authors":"Jinsong Zhao , Huidong Hou , Yunpeng Zhang , Quan Miao","doi":"10.1016/j.isatra.2025.11.012","DOIUrl":"10.1016/j.isatra.2025.11.012","url":null,"abstract":"<div><div>The hydraulically-driven lower limb exoskeleton robot (HDLLER) can enhance the motion capabilities of human lower limbs. However, due to inherent uncertainties and external human–robot coupling disturbances, it poses significant challenges to precise trajectory tracking control. Therefore, a novel funnel control (FC) strategy based on an unknown system dynamics estimator (USDE) is proposed to compensate for unknown internal and external dynamic influences, thus further improving robot performance. Firstly, the HDLLER high-order dynamic model containing human–robot coupling disturbances is transformed into a Brunovsky canonical form, effectively circumventing the ‘explosion of complexity’ problem in high order strict feedback systems. Subsequently, a set of high gain observers (HGO) is introduced to reconstruct the states of the transformed system. A modified funnel function is embedded in feedback controller to constrain the tracking error within prescribed boundaries, ensuring satisfactory transient and steady-state performances. Additionally, a simple yet effective USDE is incorporated to compensate for the lumped unknown dynamics. Different comparative simulations and gait experiments have verified the effectiveness of the proposed control method.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"168 ","pages":"Pages 455-464"},"PeriodicalIF":6.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145558498","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}
Pub Date : 2026-01-01DOI: 10.1016/j.isatra.2025.11.008
Minxuan Zha, Haoping Wang, Yang Tian
In this paper, a compensation function observer based model-free adaptive boundary layer sliding mode control (CFO-DIBLSMC) is proposed for hydraulic manipulators with dead-zone compensation. The proposed CFO-DIBLSMC adopts a dual-loop structure, comprising a force sub-control loop for precise force tracking using an adaptive admittance controller and a position sub-control loop where the complex hydraulic manipulator dynamics are approximated by an ultra-local model (ULM) to establish a model-free control framework. Then, considering the unknown dead-zone in the directional valve, a smooth dead-zone inverse method with online parameter adaptation is employed to compensate for dead-zone uncertainties. Besides, a compensation function observer is designed based on ULM to estimate lumped uncertainties of the system, achieving zero error estimation and robustness against high-frequency disturbances. Moreover, a nonsingular fast terminal sliding mode (NFTSM) sub-control law is constructed to accelerate error convergence, incorporating an adaptive switching sliding gain law within a boundary layer framework. The effectiveness of CFO-DIBLSMC is verified through comparative co-simulations against conventional control methods under dead-zone and joint friction disturbances.
{"title":"Compensation function observer based model-free adaptive boundary layer sliding mode control for hydraulic manipulators with dead-zone compensation","authors":"Minxuan Zha, Haoping Wang, Yang Tian","doi":"10.1016/j.isatra.2025.11.008","DOIUrl":"10.1016/j.isatra.2025.11.008","url":null,"abstract":"<div><div>In this paper, a compensation function observer based model-free adaptive boundary layer sliding mode control (CFO-DIBLSMC) is proposed for hydraulic manipulators with dead-zone compensation. The proposed CFO-DIBLSMC adopts a dual-loop structure, comprising a force sub-control loop for precise force tracking using an adaptive admittance controller and a position sub-control loop where the complex hydraulic manipulator dynamics are approximated by an ultra-local model (ULM) to establish a model-free control framework. Then, considering the unknown dead-zone in the directional valve, a smooth dead-zone inverse method with online parameter adaptation is employed to compensate for dead-zone uncertainties. Besides, a compensation function observer is designed based on ULM to estimate lumped uncertainties of the system, achieving zero error estimation and robustness against high-frequency disturbances. Moreover, a nonsingular fast terminal sliding mode (NFTSM) sub-control law is constructed to accelerate error convergence, incorporating an adaptive switching sliding gain law within a boundary layer framework. The effectiveness of CFO-DIBLSMC is verified through comparative co-simulations against conventional control methods under dead-zone and joint friction disturbances.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"168 ","pages":"Pages 211-224"},"PeriodicalIF":6.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145643920","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}
Sliding mode control (SMC) is valued for its robustness and capacity to handle uncertainties in robot-manipulator applications that require precise tracking. However, it is limited by chattering, and starting far from the sliding surface can lead to extended reaching phases, compromising the global control efficacy and robustness. While Time-Varying Sliding Mode Surfaces (TVSMS) have been proposed to eliminate the reaching phase, they often suffer from sensitivity to initial conditions and parameter selection, limiting precise finite-time error convergence. Ensuring robustness during the reaching and sliding phases while achieving finite-time convergence from any initial position is a challenging task. This study presents a novel approach by integrating a Finite-Time Prescribed Performance Function (FTPPF) into a TVSMS design. The proposed TVSMS, based on FTPPF, ensures error convergence within a predetermined time frame, eliminates the reaching phase, and reduces sensitivity to initial conditions. Furthermore, the designed TVSMS addresses the weakness of robustness during the reaching phase of the Power Rate Reaching Law (PRRL) employed in the control law design, thereby mitigating the chattering problem of the SMC. Three FTPPFs with minimal parameter tuning are introduced, offering flexible transient response shaping, robustness, and improved error convergence compared to traditional TVSMS. The proposed Time-Varying Sliding-Mode Controller (TVSMC) not only simplifies control implementation but also significantly enhances robustness and resilience to external disturbances, making it a promising solution for high-precision robotic applications. Finite-time stability analysis is validated using the Lyapunov theorem, and experimental validation on the MICO 4-DOF robot demonstrates superior performance across various case studies compared to conventional methods.
{"title":"Time-varying sliding mode control based finite-time prescribed performance function for robotic manipulators","authors":"Sana Stihi , Raouf Fareh , Sofiane Khadraoui , Maamar Bettayeb , Mohamed Tadjine","doi":"10.1016/j.isatra.2025.11.024","DOIUrl":"10.1016/j.isatra.2025.11.024","url":null,"abstract":"<div><div>Sliding mode control (SMC) is valued for its robustness and capacity to handle uncertainties in robot-manipulator applications that require precise tracking. However, it is limited by chattering, and starting far from the sliding surface can lead to extended reaching phases, compromising the global control efficacy and robustness. While Time-Varying Sliding Mode Surfaces (TVSMS) have been proposed to eliminate the reaching phase, they often suffer from sensitivity to initial conditions and parameter selection, limiting precise finite-time error convergence. Ensuring robustness during the reaching and sliding phases while achieving finite-time convergence from any initial position is a challenging task. This study presents a novel approach by integrating a Finite-Time Prescribed Performance Function (FTPPF) into a TVSMS design. The proposed TVSMS, based on FTPPF, ensures error convergence within a predetermined time frame, eliminates the reaching phase, and reduces sensitivity to initial conditions. Furthermore, the designed TVSMS addresses the weakness of robustness during the reaching phase of the Power Rate Reaching Law (PRRL) employed in the control law design, thereby mitigating the chattering problem of the SMC. Three FTPPFs with minimal parameter tuning are introduced, offering flexible transient response shaping, robustness, and improved error convergence compared to traditional TVSMS. The proposed Time-Varying Sliding-Mode Controller (TVSMC) not only simplifies control implementation but also significantly enhances robustness and resilience to external disturbances, making it a promising solution for high-precision robotic applications. Finite-time stability analysis is validated using the Lyapunov theorem, and experimental validation on the MICO 4-DOF robot demonstrates superior performance across various case studies compared to conventional methods.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"168 ","pages":"Pages 543-560"},"PeriodicalIF":6.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145650652","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}
Pub Date : 2026-01-01DOI: 10.1016/j.isatra.2025.11.047
Hugan Zhang , Xianku Zhang , Yongjin Liu , Shihang Gao , Daocheng Ma
This study addresses key challenges in the trajectory tracking control of unmanned surface vehicles, including external environmental disturbances, excessive sensor communication burdens, and complex controller design. An enhanced adaptive optimal backstepping control method based on an event-triggered mechanism is proposed. The approach employs an actor-critic reinforcement learning framework, in which the critic network performs online evaluation of system performance and the actor network generates optimal control decisions. Neural networks are designed to estimate the error gradient of the cost function, thereby simplifying the complex matrix differentiation process required in conventional methods. Additionally, an event-triggered mechanism is introduced to significantly reduce sensor communication frequency, while a disturbance observer is developed to estimate and compensate for environmental disturbances in real time, thus enhancing system robustness. Theoretical analysis establishes the stability and effectiveness of the proposed algorithm, and simulation results verify its superior performance.
{"title":"Event-triggered adaptive tracking control for USV based on enhanced optimized backstepping technique","authors":"Hugan Zhang , Xianku Zhang , Yongjin Liu , Shihang Gao , Daocheng Ma","doi":"10.1016/j.isatra.2025.11.047","DOIUrl":"10.1016/j.isatra.2025.11.047","url":null,"abstract":"<div><div>This study addresses key challenges in the trajectory tracking control of unmanned surface vehicles, including external environmental disturbances, excessive sensor communication burdens, and complex controller design. An enhanced adaptive optimal backstepping control method based on an event-triggered mechanism is proposed. The approach employs an actor-critic reinforcement learning framework, in which the critic network performs online evaluation of system performance and the actor network generates optimal control decisions. Neural networks are designed to estimate the error gradient of the cost function, thereby simplifying the complex matrix differentiation process required in conventional methods. Additionally, an event-triggered mechanism is introduced to significantly reduce sensor communication frequency, while a disturbance observer is developed to estimate and compensate for environmental disturbances in real time, thus enhancing system robustness. Theoretical analysis establishes the stability and effectiveness of the proposed algorithm, and simulation results verify its superior performance.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"168 ","pages":"Pages 67-80"},"PeriodicalIF":6.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145702398","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}
Pub Date : 2026-01-01DOI: 10.1016/j.isatra.2025.11.021
Rui Liu , Jimeng Li , Xilei Guan , Chenbo Jia , Jinfeng Zhang
Advanced fault diagnosis techniques typically rely on large amounts of labeled data from known domains, yet collected datasets often suffer from significant class imbalance. In addition, unseen domain data, which arises from variations in operating conditions or equipment heterogeneity, usually lacks sufficient prior knowledge, leading to a marked decline in model performance. To address these issues, this study proposes a contrast-enhanced adversarial domain generalization framework that integrates data augmentation and Bayesian inference for imbalanced fault diagnosis of rolling bearings under diverse scenarios. Specifically, a correlation-guided adaptive mixup method is developed to alleviate class imbalance in source domains by adaptively adjusting sample weights according to their similarity. A feature extractor based on multiscale pinwheel-shaped convolutions and spatial-channel collaborative attention, together with a parallel multi-classifier training architecture, is then designed to enhance feature learning from multiple source domains. To further strengthen generalization, an inter-domain contrastive loss is incorporated into adversarial training, encouraging the model to capture more robust domain-invariant representations. Finally, a Bayesian fusion mechanism with dynamic weighting is introduced to integrate the complementary strengths of different classifiers for accurate recognition of unseen domain data. Two rolling bearing datasets are employed to construct cross-condition and cross-machine experimental tasks. Comparative results demonstrate that the proposed approach achieves superior diagnostic accuracy and strong generalization capability, thus providing a reliable solution for industrial equipment health monitoring.
{"title":"Contrast-enhanced adversarial domain generalization network with data augmentation and Bayesian inference for imbalanced bearing fault diagnosis","authors":"Rui Liu , Jimeng Li , Xilei Guan , Chenbo Jia , Jinfeng Zhang","doi":"10.1016/j.isatra.2025.11.021","DOIUrl":"10.1016/j.isatra.2025.11.021","url":null,"abstract":"<div><div>Advanced fault diagnosis techniques typically rely on large amounts of labeled data from known domains, yet collected datasets often suffer from significant class imbalance. In addition, unseen domain data, which arises from variations in operating conditions or equipment heterogeneity, usually lacks sufficient prior knowledge, leading to a marked decline in model performance. To address these issues, this study proposes a contrast-enhanced adversarial domain generalization framework that integrates data augmentation and Bayesian inference for imbalanced fault diagnosis of rolling bearings under diverse scenarios. Specifically, a correlation-guided adaptive mixup method is developed to alleviate class imbalance in source domains by adaptively adjusting sample weights according to their similarity. A feature extractor based on multiscale pinwheel-shaped convolutions and spatial-channel collaborative attention, together with a parallel multi-classifier training architecture, is then designed to enhance feature learning from multiple source domains. To further strengthen generalization, an inter-domain contrastive loss is incorporated into adversarial training, encouraging the model to capture more robust domain-invariant representations. Finally, a Bayesian fusion mechanism with dynamic weighting is introduced to integrate the complementary strengths of different classifiers for accurate recognition of unseen domain data. Two rolling bearing datasets are employed to construct cross-condition and cross-machine experimental tasks. Comparative results demonstrate that the proposed approach achieves superior diagnostic accuracy and strong generalization capability, thus providing a reliable solution for industrial equipment health monitoring.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"168 ","pages":"Pages 668-685"},"PeriodicalIF":6.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145552518","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}
Pub Date : 2026-01-01DOI: 10.1016/j.isatra.2025.10.053
Xinhao Ji , Yuexing Wang , Haibo Xie , Chengzhen Wang
High-precision control technology for the pump controlled single-rod actuator (PCSA) system is in high demand in industrial applications. In the PCSA system, there are both matched and mismatched uncertainties. However, it is still a challenging task for existing observer-based controllers to simultaneously observe and compensate for the uncertainty of mismatched and matched. To solve this issue, in this article, a modeling-uncertainty-observer based nonlinear robust control strategy was designed. The working mechanism of the proposed method is to observe the uncertainties from measurable state variables and then, based on the uncertainties observation, take feedforward control to compensate for the uncertainties. The developed observer can simultaneously observe and compensate for the uncertainties of mismatched and matched. Meanwhile, to improve transient response, the robust control technology was employed to deal with the compensation error. Comparative experiments show that the proposed method can effectively estimate and compensate for the model uncertainty, achieving the prescribed control accuracy. Compared with the robust controller and PID controller, the proposed method reduces the maximum control error by more than 20 % and 61 %, respectively.
{"title":"Nonlinear robust control of pump controlled single-rod actuator with observing and compensating modeling uncertainties","authors":"Xinhao Ji , Yuexing Wang , Haibo Xie , Chengzhen Wang","doi":"10.1016/j.isatra.2025.10.053","DOIUrl":"10.1016/j.isatra.2025.10.053","url":null,"abstract":"<div><div>High-precision control technology for the pump controlled single-rod actuator (PCSA) system is in high demand in industrial applications. In the PCSA system, there are both matched and mismatched uncertainties. However, it is still a challenging task for existing observer-based controllers to simultaneously observe and compensate for the uncertainty of mismatched and matched. To solve this issue, in this article, a modeling-uncertainty-observer based nonlinear robust control strategy was designed. The working mechanism of the proposed method is to observe the uncertainties from measurable state variables and then, based on the uncertainties observation, take feedforward control to compensate for the uncertainties. The developed observer can simultaneously observe and compensate for the uncertainties of mismatched and matched. Meanwhile, to improve transient response, the robust control technology was employed to deal with the compensation error. Comparative experiments show that the proposed method can effectively estimate and compensate for the model uncertainty, achieving the prescribed control accuracy. Compared with the robust controller and PID controller, the proposed method reduces the maximum control error by more than 20 % and 61 %, respectively.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"168 ","pages":"Pages 160-173"},"PeriodicalIF":6.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145515149","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}
Pub Date : 2026-01-01DOI: 10.1016/j.isatra.2025.10.045
Peng Cheng , Wenjun Luo , Jason J.R. Liu , Zhiguang Feng
This study proposes a fixed-time disturbance observer (DO)-based dynamic event-triggered fault-tolerant control (DETFTC) framework with prescribed performance for the attitude tracking of rigid spacecraft hampered by actuator faults, parameter uncertainties, and environmental disturbances. In contrast to conventional DO approaches that rely on initial conditions or upper disturbance bounds, this work introduces a fixed-time DO to accurately reconstruct lumped perturbations arising from exogenous disturbances, inertia uncertainties, and actuator faults. A DETFTC strategy is further developed to minimize unnecessary control signal updates while ensuring rapid and precise attitude tracking. Integrating fixed-time performance functions with barrier Lyapunov functions guarantees that attitude-tracking errors converge to predefined ranges within a bounded time. The proposed method ensures that all signals in the closed-loop system are practically fixed-time stable, with attitude tracking errors constrained within predetermined boundaries and Zeno behavior effectively excluded. Lastly, case studies are conducted to demonstrate the effectiveness and practical applicability of the developed methodology.
{"title":"Dynamic event-triggered fault-tolerant control of rigid spacecraft with prescribed performance","authors":"Peng Cheng , Wenjun Luo , Jason J.R. Liu , Zhiguang Feng","doi":"10.1016/j.isatra.2025.10.045","DOIUrl":"10.1016/j.isatra.2025.10.045","url":null,"abstract":"<div><div>This study proposes a fixed-time disturbance observer (DO)-based dynamic event-triggered fault-tolerant control (DETFTC) framework with prescribed performance for the attitude tracking of rigid spacecraft hampered by actuator faults, parameter uncertainties, and environmental disturbances. In contrast to conventional DO approaches that rely on initial conditions or upper disturbance bounds, this work introduces a fixed-time DO to accurately reconstruct lumped perturbations arising from exogenous disturbances, inertia uncertainties, and actuator faults. A DETFTC strategy is further developed to minimize unnecessary control signal updates while ensuring rapid and precise attitude tracking. Integrating fixed-time performance functions with barrier Lyapunov functions guarantees that attitude-tracking errors converge to predefined ranges within a bounded time. The proposed method ensures that all signals in the closed-loop system are practically fixed-time stable, with attitude tracking errors constrained within predetermined boundaries and Zeno behavior effectively excluded. Lastly, case studies are conducted to demonstrate the effectiveness and practical applicability of the developed methodology.</div></div>","PeriodicalId":14660,"journal":{"name":"ISA transactions","volume":"168 ","pages":"Pages 52-66"},"PeriodicalIF":6.5,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145515046","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号