State-of-the-art generalist manipulation policies have enabled the deployment of robotic manipulators in unstructured human environments. However, these frameworks struggle in cluttered environments primarily because they utilize auxiliary modules for low-level motion planning and control. Motion planning remains challenging due to the high dimensionality of the robot’s configuration space and the presence of workspace obstacles. Neural motion planners have enhanced motion planning efficiency by offering fast inference and effectively handling the inherent multi-modality of the motion planning problem. Despite such benefits, current neural motion planners often struggle to generalize to unseen, out-of-distribution planning settings. This paper reviews and analyzes the state-of-the-art neural motion planners, highlighting both their benefits and limitations. It also outlines a path toward establishing generalist neural motion planners capable of handling domain-specific challenges. For a list of the reviewed papers, please refer to https://davoodsz.github.io/planning-manip-survey.github.io/ Note to Practitioners—The paper reviews and summarizes rapidly evolving studies that leverage deep learning for motion planning of robotic manipulators. As robotic manipulators continue to transition from controlled laboratory environments to real-world settings, the demand for efficient, robust, and adaptable motion planning algorithms grows significantly. Thanks to characteristics such as fast inference time and inherent inductive bias, deep learning has been leveraged to facilitate this transition. This paper extensively reviews state-of-the-art deep learning methods used for motion planning of robotic manipulators, and outlines promising avenues and challenges for future research. It specifically evaluates and summarizes the performance of the most commonly used deep learning methods on various key components of motion planning, such as informed sampling, warm-starting trajectory optimization, and collision checking. This paper can serve as a resource for both experts and newcomers in high-DoF robotic motion planning using deep learning.
{"title":"Toward Generalist Neural Motion Planners for Robotic Manipulators: Challenges and Opportunities","authors":"Davood Soleymanzadeh;Ivan Lopez-Sanchez;Hao Su;Yunzhu Li;Xiao Liang;Minghui Zheng","doi":"10.1109/TASE.2026.3660830","DOIUrl":"10.1109/TASE.2026.3660830","url":null,"abstract":"State-of-the-art generalist manipulation policies have enabled the deployment of robotic manipulators in unstructured human environments. However, these frameworks struggle in cluttered environments primarily because they utilize auxiliary modules for low-level motion planning and control. Motion planning remains challenging due to the high dimensionality of the robot’s configuration space and the presence of workspace obstacles. Neural motion planners have enhanced motion planning efficiency by offering fast inference and effectively handling the inherent multi-modality of the motion planning problem. Despite such benefits, current neural motion planners often struggle to generalize to unseen, out-of-distribution planning settings. This paper reviews and analyzes the state-of-the-art neural motion planners, highlighting both their benefits and limitations. It also outlines a path toward establishing generalist neural motion planners capable of handling domain-specific challenges. For a list of the reviewed papers, please refer to <uri>https://davoodsz.github.io/planning-manip-survey.github.io/</uri> Note to Practitioners—The paper reviews and summarizes rapidly evolving studies that leverage deep learning for motion planning of robotic manipulators. As robotic manipulators continue to transition from controlled laboratory environments to real-world settings, the demand for efficient, robust, and adaptable motion planning algorithms grows significantly. Thanks to characteristics such as fast inference time and inherent inductive bias, deep learning has been leveraged to facilitate this transition. This paper extensively reviews state-of-the-art deep learning methods used for motion planning of robotic manipulators, and outlines promising avenues and challenges for future research. It specifically evaluates and summarizes the performance of the most commonly used deep learning methods on various key components of motion planning, such as informed sampling, warm-starting trajectory optimization, and collision checking. This paper can serve as a resource for both experts and newcomers in high-DoF robotic motion planning using deep learning.","PeriodicalId":51060,"journal":{"name":"IEEE Transactions on Automation Science and Engineering","volume":"23 ","pages":"4488-4531"},"PeriodicalIF":6.4,"publicationDate":"2026-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146109884","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 : 2026-02-03DOI: 10.1109/tase.2026.3660357
Ya Liu, Yueer Wu, Fan Zhang, Panfeng Huang, Yingbo Lu, Haitao Chang
{"title":"Collision-free Trajectory Generation and Robust Nonlinear Distributed Model Predictive Control for Tethered Multi-rotor Unmanned Aerial Vehicles","authors":"Ya Liu, Yueer Wu, Fan Zhang, Panfeng Huang, Yingbo Lu, Haitao Chang","doi":"10.1109/tase.2026.3660357","DOIUrl":"https://doi.org/10.1109/tase.2026.3660357","url":null,"abstract":"","PeriodicalId":51060,"journal":{"name":"IEEE Transactions on Automation Science and Engineering","volume":"34 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146109880","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 : 2026-02-03DOI: 10.1109/tase.2026.3660614
Jun Zhang, Sheik M. Mohiuddin, Junjian Qi
{"title":"Distributed Optimal Control for Grid-Forming and Grid-Feeding Converters in DC Microgrid","authors":"Jun Zhang, Sheik M. Mohiuddin, Junjian Qi","doi":"10.1109/tase.2026.3660614","DOIUrl":"https://doi.org/10.1109/tase.2026.3660614","url":null,"abstract":"","PeriodicalId":51060,"journal":{"name":"IEEE Transactions on Automation Science and Engineering","volume":"117 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146109881","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 : 2026-02-03DOI: 10.1109/TASE.2026.3660685
Jinsen Zhang;Xiaobing Nie;Jinde Cao;Liang Hua
This paper studies the finite-time multistability of impulsive Hopfield neural networks with a general class of activation functions. First, the existence of $prod _{i=1}^{n}(2M_{i}+1)$ equilibrium points and $prod _{i=1}^{n}(M_{i}+1)$ invariant sets in such $n$ -neuron neural networks can be guaranteed by applying the Brouwer’s fixed-point theorem as well as upper and lower functions method. Furthermore, it is demonstrated that these equilibrium points and invariant sets remain valid for the same neural networks when subjected to an appropriate controller. Then, on the basis of Lyapunov function method and impulsive control theory, two finite-time multistability theorems are established for Hopfield neural networks under distinct impulse scenarios: stabilizing impulses and destabilizing impulses. The settling time estimations for determining the local finite-time stability of $prod _{i=1}^{n}(M_{i}+1)$ equilibrium points are developed by designing general impulsive sequences, which reveal that the settling time is dependent on initial state, impulsive effects and control parameters. From the perspective of impulsive effects, the introduced stabilizing impulses in neural networks not only accelerate the convergence rate but also yield tighter upper bound of settling time estimation relative to impulse-free systems. In stark contrast, destabilizing impulses significantly degrade the convergence performance while resulting in more conservative upper bound of settling time estimation. Finally, theoretical results are shown to be effective by two illustrative examples and two associative memory applications of grayscale image. Note to Practitioners—Multistability analysis, which investigates the coexistence and and local stability of multiple equilibrium points in dynamical systems, is directly applicable to some critical areas such as associative memory, pattern recognition, and combinatorial optimization, where each stable equilibrium point can represent a stored pattern or a feasible solution. However, prevailing multistability results primarily focus on asymptotic or exponential stability of these equilibria, which often fails to meet practical engineering requirements due to slow convergence speed and limited steady-state accuracy. To overcome these limitations, we design a novel controller coupled with a suitably constructed impulsive sequence that drives the system trajectory to multiple stable equilibrium points within the settling time. This approach guarantees significantly accelerated convergence and enhanced precision, offering a practical and effective solution for implementing high-speed and high-accuracy intelligent systems.
{"title":"Finite-Time Multistability of Impulsive Hopfield Neural Networks Under New Impulsive Sequence Designs","authors":"Jinsen Zhang;Xiaobing Nie;Jinde Cao;Liang Hua","doi":"10.1109/TASE.2026.3660685","DOIUrl":"10.1109/TASE.2026.3660685","url":null,"abstract":"This paper studies the finite-time multistability of impulsive Hopfield neural networks with a general class of activation functions. First, the existence of <inline-formula> <tex-math>$prod _{i=1}^{n}(2M_{i}+1)$ </tex-math></inline-formula> equilibrium points and <inline-formula> <tex-math>$prod _{i=1}^{n}(M_{i}+1)$ </tex-math></inline-formula> invariant sets in such <inline-formula> <tex-math>$n$ </tex-math></inline-formula>-neuron neural networks can be guaranteed by applying the Brouwer’s fixed-point theorem as well as upper and lower functions method. Furthermore, it is demonstrated that these equilibrium points and invariant sets remain valid for the same neural networks when subjected to an appropriate controller. Then, on the basis of Lyapunov function method and impulsive control theory, two finite-time multistability theorems are established for Hopfield neural networks under distinct impulse scenarios: stabilizing impulses and destabilizing impulses. The settling time estimations for determining the local finite-time stability of <inline-formula> <tex-math>$prod _{i=1}^{n}(M_{i}+1)$ </tex-math></inline-formula> equilibrium points are developed by designing general impulsive sequences, which reveal that the settling time is dependent on initial state, impulsive effects and control parameters. From the perspective of impulsive effects, the introduced stabilizing impulses in neural networks not only accelerate the convergence rate but also yield tighter upper bound of settling time estimation relative to impulse-free systems. In stark contrast, destabilizing impulses significantly degrade the convergence performance while resulting in more conservative upper bound of settling time estimation. Finally, theoretical results are shown to be effective by two illustrative examples and two associative memory applications of grayscale image. Note to Practitioners—Multistability analysis, which investigates the coexistence and and local stability of multiple equilibrium points in dynamical systems, is directly applicable to some critical areas such as associative memory, pattern recognition, and combinatorial optimization, where each stable equilibrium point can represent a stored pattern or a feasible solution. However, prevailing multistability results primarily focus on asymptotic or exponential stability of these equilibria, which often fails to meet practical engineering requirements due to slow convergence speed and limited steady-state accuracy. To overcome these limitations, we design a novel controller coupled with a suitably constructed impulsive sequence that drives the system trajectory to multiple stable equilibrium points within the settling time. This approach guarantees significantly accelerated convergence and enhanced precision, offering a practical and effective solution for implementing high-speed and high-accuracy intelligent systems.","PeriodicalId":51060,"journal":{"name":"IEEE Transactions on Automation Science and Engineering","volume":"23 ","pages":"4627-4638"},"PeriodicalIF":6.4,"publicationDate":"2026-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146109882","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 : 2026-02-02DOI: 10.1109/TASE.2026.3660080
Guangrong Chen;Qizhe Yang;Jiehao Li;C. L. Philip Chen;Chenguang Yang
This paper presents a cooperative control framework for dual-arm robots that integrates vision-language models (VLMs) with online reinforcement learning (RL) to enhance autonomy and adaptability in complex manipulation tasks. The proposed framework adopts a hierarchical architecture: at the top level, the VLM interprets natural language instructions and visual image to generate task plans; at the middle level, an online RL module refines manipulation policies and ensures adaptive decision-making under environmental uncertainty; and at the bottom level, compliant control based on trajectory planning and impedance regulation enables safe and robust execution. In the feedback, YOLOv5 is used to detect the object, GraspNet is used to obtain the optimal grasp pose, and CLIP (Contrastive Language-Image Pre-Training) is used to judge whether task is completed. Simulations and real-world experiments validate the effectiveness of the proposed method. The dual-arm robot successfully performed various cooperative tasks such as grasping, bottle-cap unscrewing, water pouring, and box carrying, achieving an increase in the task success rate from 43% to 100% with online adaptive learning and training. These results demonstrate that the proposed framework effectively bridges high-level reasoning with low-level control, providing a scalable solution for future applications in service robotics, industrial automation, and human-robot collaboration. Note to Practitioners—This work is motivated by the practical challenge of enabling dual-arm robots to execute complex tasks in unstructured environments such as warehouses, factories, and service settings. Traditional robots often struggle with coordinating both arms, adapting to novel objects, and making real-time decisions. Our framework integrates a VLM for high-level task planning with RL for adaptive execution. The VLM interprets human instructions and environmental cues, while RL enables the robot to refine its performance through trial and error. Practically, this allows the robot to decide when to use the left arm, the right arm, or both arms cooperatively, improving efficiency and flexibility across tasks-such as single-arm grasping of lightweight objects or dual-arm handling of heavier and elongated items. The framework enhances productivity while reducing manual programming effort. Current limitations include reliance on a fixed-depth camera, which can cause occlusions, and the computational cost of online model updates. These are partly mitigated through impedance control and torque feedback, but further improvements in perception and real-time learning are needed. Overall, the approach offers practitioners a pathway toward more versatile and adaptive dual-arm robotic systems for real-world deployment.
{"title":"Cooperative Control Framework for Dual-Arm Robot Enhanced by Vision Language Model and Reinforcement Learning","authors":"Guangrong Chen;Qizhe Yang;Jiehao Li;C. L. Philip Chen;Chenguang Yang","doi":"10.1109/TASE.2026.3660080","DOIUrl":"10.1109/TASE.2026.3660080","url":null,"abstract":"This paper presents a cooperative control framework for dual-arm robots that integrates vision-language models (VLMs) with online reinforcement learning (RL) to enhance autonomy and adaptability in complex manipulation tasks. The proposed framework adopts a hierarchical architecture: at the top level, the VLM interprets natural language instructions and visual image to generate task plans; at the middle level, an online RL module refines manipulation policies and ensures adaptive decision-making under environmental uncertainty; and at the bottom level, compliant control based on trajectory planning and impedance regulation enables safe and robust execution. In the feedback, YOLOv5 is used to detect the object, GraspNet is used to obtain the optimal grasp pose, and CLIP (Contrastive Language-Image Pre-Training) is used to judge whether task is completed. Simulations and real-world experiments validate the effectiveness of the proposed method. The dual-arm robot successfully performed various cooperative tasks such as grasping, bottle-cap unscrewing, water pouring, and box carrying, achieving an increase in the task success rate from 43% to 100% with online adaptive learning and training. These results demonstrate that the proposed framework effectively bridges high-level reasoning with low-level control, providing a scalable solution for future applications in service robotics, industrial automation, and human-robot collaboration. Note to Practitioners—This work is motivated by the practical challenge of enabling dual-arm robots to execute complex tasks in unstructured environments such as warehouses, factories, and service settings. Traditional robots often struggle with coordinating both arms, adapting to novel objects, and making real-time decisions. Our framework integrates a VLM for high-level task planning with RL for adaptive execution. The VLM interprets human instructions and environmental cues, while RL enables the robot to refine its performance through trial and error. Practically, this allows the robot to decide when to use the left arm, the right arm, or both arms cooperatively, improving efficiency and flexibility across tasks-such as single-arm grasping of lightweight objects or dual-arm handling of heavier and elongated items. The framework enhances productivity while reducing manual programming effort. Current limitations include reliance on a fixed-depth camera, which can cause occlusions, and the computational cost of online model updates. These are partly mitigated through impedance control and torque feedback, but further improvements in perception and real-time learning are needed. Overall, the approach offers practitioners a pathway toward more versatile and adaptive dual-arm robotic systems for real-world deployment.","PeriodicalId":51060,"journal":{"name":"IEEE Transactions on Automation Science and Engineering","volume":"23 ","pages":"4314-4328"},"PeriodicalIF":6.4,"publicationDate":"2026-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146101558","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 : 2026-02-02DOI: 10.1109/tase.2026.3660049
Wenxue Zhang, Yujie Zhang, Zhao Wan, Xiaohui Jia, Dušan M. Stipanović, Jinyue Liu
{"title":"Event-Triggered Collision-Free Formation Control of Symmetric-Steering Vehicles using Generalized p-norm and Smooth Risk Metrics","authors":"Wenxue Zhang, Yujie Zhang, Zhao Wan, Xiaohui Jia, Dušan M. Stipanović, Jinyue Liu","doi":"10.1109/tase.2026.3660049","DOIUrl":"https://doi.org/10.1109/tase.2026.3660049","url":null,"abstract":"","PeriodicalId":51060,"journal":{"name":"IEEE Transactions on Automation Science and Engineering","volume":"44 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146101560","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 : 2026-02-02DOI: 10.1109/tase.2026.3660098
Yu Ding, Carlos Costa, Bing Si
{"title":"Federated Function-on-function Regression with an Efficient Gradient Boosting Algorithm for Privacy-Preserving Telemedicine","authors":"Yu Ding, Carlos Costa, Bing Si","doi":"10.1109/tase.2026.3660098","DOIUrl":"https://doi.org/10.1109/tase.2026.3660098","url":null,"abstract":"","PeriodicalId":51060,"journal":{"name":"IEEE Transactions on Automation Science and Engineering","volume":"232 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146101559","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 : 2026-01-30DOI: 10.1109/tase.2026.3659481
Yingying Ren, Qian Wang, Da-Wei Ding
{"title":"Output Feedback Control Synthesis for Positive Fuzzy Systems Using a Sequential Linear Programming Approach","authors":"Yingying Ren, Qian Wang, Da-Wei Ding","doi":"10.1109/tase.2026.3659481","DOIUrl":"https://doi.org/10.1109/tase.2026.3659481","url":null,"abstract":"","PeriodicalId":51060,"journal":{"name":"IEEE Transactions on Automation Science and Engineering","volume":"136 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146089873","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 Tube-Based Model Predictive Control for Docking Process Self-Balancing Control of a Reconfigurable Unmanned Ground Vehicle","authors":"Congnan Yang, Xiaojun Xu, Yijie Zhao, Jianwen Liu, Wenhao Wang, Yaoyao Chen, Yongxiang Lei","doi":"10.1109/tase.2026.3654162","DOIUrl":"https://doi.org/10.1109/tase.2026.3654162","url":null,"abstract":"","PeriodicalId":51060,"journal":{"name":"IEEE Transactions on Automation Science and Engineering","volume":"33 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146089875","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}
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：