Recent advancements in 3D Gaussian Splatting have achieved high-quality and real-time novel view synthesis for 3D scenes. However, this method primarily focuses on appearance and geometric modeling, lacks the ability to comprehend scenes with fine granularity at the object level. Some techniques enhance Gaussian Splatting, empowering it with the capability to perform unified 3D reconstruction and segmentation on real-world 3D scenes. Despite its improvements, these methods still face challenges in segmentation accuracy and reconstruction quality using 3D Gaussian representation. Additionally, the use of 16-bit identity codes to distinguish different Gaussian groups significantly increases memory overhead. To address these issues, we propose Identity-Encoding Half-Gaussian (ID-HGS) kernels. Our approach introduces a plane to split each Gaussian into two parts with distinct opacity values, enabling precise reconstruction of details and object boundaries. We replace the adaptive density control (ADC) used in Gaussian Grouping with localized half—gaussian point management (LHPM), which performs finer densification in under-reconstructed regions. LHPM resets pathological Gaussians and optimizes Gaussian density, reducing their impact on segmentation accuracy. Furthermore, we assign a global contribution score to each Gaussian and prune low-contribution Gaussians during training, saving memory and accelerating training. Compared to Gaussian Grouping, our method improves both reconstruction quality and segmentation accuracy while effectively controlling memory usage. Extensive experiments demonstrate that HGS-3D outperforms prior Gaussian Grouping on both reconstruction and segmentation: it achieves higher mask accuracy on the LERF-Localization benchmark and reduces peak memory usage while improving render quality on the mipnerf360 dataset. Note to Practitioners—3D Gaussian Splatting enables real-time and high-quality 3D scene rendering/reconstruction, but struggles with object-level understanding and efficient memory usage. This paper presents HGS-3DSeg that enhances reconstruction and segmentation performance. By introducing Half-Gaussian kernels with identity encoding, our method better preserves object boundaries and fine details. We further propose Localized Half-Gaussian Point Management (LHPM) for targeted densification and Gaussian contribution scoring for memory-efficient pruning. Experiments show that our approach improves segmentation accuracy, rendering quality, and memory efficiency, making it suitable for AR/VR, robotics, and 3D scene understanding.
{"title":"HGS-3DSeg: Identity-Encoding Half-Gaussian Splatting for Memory-Efficient 3D Reconstruction and Segmentation","authors":"Weiqing Yan;Jiahao Li;Liang Liao;Kaile Su;Chang Tang","doi":"10.1109/TASE.2026.3659445","DOIUrl":"10.1109/TASE.2026.3659445","url":null,"abstract":"Recent advancements in 3D Gaussian Splatting have achieved high-quality and real-time novel view synthesis for 3D scenes. However, this method primarily focuses on appearance and geometric modeling, lacks the ability to comprehend scenes with fine granularity at the object level. Some techniques enhance Gaussian Splatting, empowering it with the capability to perform unified 3D reconstruction and segmentation on real-world 3D scenes. Despite its improvements, these methods still face challenges in segmentation accuracy and reconstruction quality using 3D Gaussian representation. Additionally, the use of 16-bit identity codes to distinguish different Gaussian groups significantly increases memory overhead. To address these issues, we propose Identity-Encoding Half-Gaussian (ID-HGS) kernels. Our approach introduces a plane to split each Gaussian into two parts with distinct opacity values, enabling precise reconstruction of details and object boundaries. We replace the adaptive density control (ADC) used in Gaussian Grouping with localized half—gaussian point management (LHPM), which performs finer densification in under-reconstructed regions. LHPM resets pathological Gaussians and optimizes Gaussian density, reducing their impact on segmentation accuracy. Furthermore, we assign a global contribution score to each Gaussian and prune low-contribution Gaussians during training, saving memory and accelerating training. Compared to Gaussian Grouping, our method improves both reconstruction quality and segmentation accuracy while effectively controlling memory usage. Extensive experiments demonstrate that HGS-3D outperforms prior Gaussian Grouping on both reconstruction and segmentation: it achieves higher mask accuracy on the LERF-Localization benchmark and reduces peak memory usage while improving render quality on the mipnerf360 dataset. Note to Practitioners—3D Gaussian Splatting enables real-time and high-quality 3D scene rendering/reconstruction, but struggles with object-level understanding and efficient memory usage. This paper presents HGS-3DSeg that enhances reconstruction and segmentation performance. By introducing Half-Gaussian kernels with identity encoding, our method better preserves object boundaries and fine details. We further propose Localized Half-Gaussian Point Management (LHPM) for targeted densification and Gaussian contribution scoring for memory-efficient pruning. Experiments show that our approach improves segmentation accuracy, rendering quality, and memory efficiency, making it suitable for AR/VR, robotics, and 3D scene understanding.","PeriodicalId":51060,"journal":{"name":"IEEE Transactions on Automation Science and Engineering","volume":"23 ","pages":"4590-4601"},"PeriodicalIF":6.4,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146089825","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 : 2026-01-29DOI: 10.1109/TASE.2026.3658557
Jiarong Hu;Wangxie Gu;Jiaying Hu;Jie Xiao;Jianzhong Fu;Songyu Hu
Contact sensing along the soft catheter robots (SCRs) is crucial for enhancing surgical performance and intraoperative safety during minimally invasive procedures. Nevertheless, current approaches relying on medical imaging devices or optical fiber sensors are noticeably costly and pose challenges for clinical applications. This study develops a more implementable and cost-effective contact sensing method based on magnetic signals for SCR-based surgery. Specifically, several miniature Hall sensors are integrated into the SCR, and a heterogeneous magnetic field is applied. When the SCR comes into contact with the external environment, its deformation leads to changes in the magnetic signal detected by the Hall sensors. A contact sensing deep learning (ContactSenseDL) model is then developed to map the magnetic signal variation to contact position and force along the SCR. The proposed approach is validated on a catheter robot system prototype and achieves remarkable performance. In contact position prediction, the average error of axial arc length is as low as 1.97 mm (2.81% of the maximum insert length of the SCR), and the prediction of radial position exhibits high consistency with actual values. In contact force prediction, the average errors of friction and pressure are 2.43 (2.59% of the maximum friction) and 1.61 mN (2.45% of the maximum pressure), respectively. Additionally, contact sensing experiments are conducted on a knee model to demonstrate the potential application of this method. Overall, the proposed contact sensing strategy can effectively sense the contact position and force along SCRs in 3D space, holding promise for enhancing safety in SCR-based surgery. Note to Practitioners—This research is motivated by the growing demand for more feasible and cost-effective contact sensing methods in SCR-based minimally invasive surgery to ensure surgical efficacy and safety. In this work, a magnetic signal-based contact sensing method is developed. The proposed method dispersedly integrates several Hall sensors into the SCR and utilizes a deep learning model to map the measured magnetic signal to contact information. After the training of the deep learning model, the presented contact sensing method only requires Hall sensors and electromagnets in hardware to achieve effective contact position and force estimation along the SCRs in 3D space. Therefore, this method features low operational and maintenance costs, offering potential for widespread adoption in resource-constrained healthcare settings.
{"title":"Contact Sensing Along Soft Catheter Robots in Minimally Invasive Surgery Based on Magnetic Signals","authors":"Jiarong Hu;Wangxie Gu;Jiaying Hu;Jie Xiao;Jianzhong Fu;Songyu Hu","doi":"10.1109/TASE.2026.3658557","DOIUrl":"10.1109/TASE.2026.3658557","url":null,"abstract":"Contact sensing along the soft catheter robots (SCRs) is crucial for enhancing surgical performance and intraoperative safety during minimally invasive procedures. Nevertheless, current approaches relying on medical imaging devices or optical fiber sensors are noticeably costly and pose challenges for clinical applications. This study develops a more implementable and cost-effective contact sensing method based on magnetic signals for SCR-based surgery. Specifically, several miniature Hall sensors are integrated into the SCR, and a heterogeneous magnetic field is applied. When the SCR comes into contact with the external environment, its deformation leads to changes in the magnetic signal detected by the Hall sensors. A contact sensing deep learning (ContactSenseDL) model is then developed to map the magnetic signal variation to contact position and force along the SCR. The proposed approach is validated on a catheter robot system prototype and achieves remarkable performance. In contact position prediction, the average error of axial arc length is as low as 1.97 mm (2.81% of the maximum insert length of the SCR), and the prediction of radial position exhibits high consistency with actual values. In contact force prediction, the average errors of friction and pressure are 2.43 (2.59% of the maximum friction) and 1.61 mN (2.45% of the maximum pressure), respectively. Additionally, contact sensing experiments are conducted on a knee model to demonstrate the potential application of this method. Overall, the proposed contact sensing strategy can effectively sense the contact position and force along SCRs in 3D space, holding promise for enhancing safety in SCR-based surgery. Note to Practitioners—This research is motivated by the growing demand for more feasible and cost-effective contact sensing methods in SCR-based minimally invasive surgery to ensure surgical efficacy and safety. In this work, a magnetic signal-based contact sensing method is developed. The proposed method dispersedly integrates several Hall sensors into the SCR and utilizes a deep learning model to map the measured magnetic signal to contact information. After the training of the deep learning model, the presented contact sensing method only requires Hall sensors and electromagnets in hardware to achieve effective contact position and force estimation along the SCRs in 3D space. Therefore, this method features low operational and maintenance costs, offering potential for widespread adoption in resource-constrained healthcare settings.","PeriodicalId":51060,"journal":{"name":"IEEE Transactions on Automation Science and Engineering","volume":"23 ","pages":"4567-4578"},"PeriodicalIF":6.4,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146089879","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 : 2026-01-29DOI: 10.1109/TASE.2026.3659225
Xinyi Yuan;Weiwei Wan;Kensuke Harada
This paper revisits the numerical inverse kinematics (IK) problem, leveraging modern computational resources and refining the seed selection process to develop a solver that is competitive with analytical-based methods. The proposed method discretizes the robot configuration space via Centroidal Voronoi Tessellation (CVT) indexed in a KDTree, ranks candidate joint configurations by minimal joint-space adjustment, and iteratively re-attempt with the next-nearest seeds in pose space according to CVT Voronoi distance. The joint space adjustment-based seed selection increases the likelihood of rapid convergence, while the re-attempt strategy effectively helps circumvent local minima and joint limit constraints. Comparison results with both traditional numerical solvers and learning-based methods demonstrate the strengths of the proposed approach in terms of success rate, time efficiency, and accuracy. Additionally, we conduct detailed ablation studies to analyze the effects of various parameters and solver settings, providing practical insights for customization and optimization. The proposed method consistently exhibits high success rates and computational efficiency. It is suitable for time-sensitive applications. Note to Practitioners—The proposed IK solver serves as the default IK solver of our WRS Robot Planning and Control System (https://github.com/wanweiwei07/wrs). The core implementation of the method is located in the $mathtt {ik_sel.py}$ file within the $mathtt {wrs.robot_sim._kinematics}$ package. For rapid prototyping, we provide a manipulator interface class in $mathtt {wrs.robot_sim.manipulators.manipulator_interface}$ . Practitioners can define specialized manipulators by implementing the interface and then invoking the inherited $mathtt {ik()}$ member function to solve the manipulator’s IK using the proposed method. Users may adjust the maximum allowed number of re-attempts by modifying line 172 in the $mathtt {ik_sel.py}$ file to achieve an optimal balance between success rate and computation efficiency based on their requirements. For integration with other simulation systems, practitioners can refer directly to the structure and logic provided in the standalone implementation of the $mathtt {ik_sel.py}$ file. This file is fully aligned with the methods presented in this paper and offers a clear, modular reference for reproducing the proposed selection strategy.
{"title":"IKSel: Selecting Good Seed Joint Values for Fast Numerical Inverse Kinematics Iterations","authors":"Xinyi Yuan;Weiwei Wan;Kensuke Harada","doi":"10.1109/TASE.2026.3659225","DOIUrl":"10.1109/TASE.2026.3659225","url":null,"abstract":"This paper revisits the numerical inverse kinematics (IK) problem, leveraging modern computational resources and refining the seed selection process to develop a solver that is competitive with analytical-based methods. The proposed method discretizes the robot configuration space via Centroidal Voronoi Tessellation (CVT) indexed in a KDTree, ranks candidate joint configurations by minimal joint-space adjustment, and iteratively re-attempt with the next-nearest seeds in pose space according to CVT Voronoi distance. The joint space adjustment-based seed selection increases the likelihood of rapid convergence, while the re-attempt strategy effectively helps circumvent local minima and joint limit constraints. Comparison results with both traditional numerical solvers and learning-based methods demonstrate the strengths of the proposed approach in terms of success rate, time efficiency, and accuracy. Additionally, we conduct detailed ablation studies to analyze the effects of various parameters and solver settings, providing practical insights for customization and optimization. The proposed method consistently exhibits high success rates and computational efficiency. It is suitable for time-sensitive applications. Note to Practitioners—The proposed IK solver serves as the default IK solver of our WRS Robot Planning and Control System (<uri>https://github.com/wanweiwei07/wrs</uri>). The core implementation of the method is located in the <inline-formula> <tex-math>$mathtt {ik_sel.py}$ </tex-math></inline-formula> file within the <inline-formula> <tex-math>$mathtt {wrs.robot_sim._kinematics}$ </tex-math></inline-formula> package. For rapid prototyping, we provide a manipulator interface class in <inline-formula> <tex-math>$mathtt {wrs.robot_sim.manipulators.manipulator_interface}$ </tex-math></inline-formula>. Practitioners can define specialized manipulators by implementing the interface and then invoking the inherited <inline-formula> <tex-math>$mathtt {ik()}$ </tex-math></inline-formula> member function to solve the manipulator’s IK using the proposed method. Users may adjust the maximum allowed number of re-attempts by modifying line 172 in the <inline-formula> <tex-math>$mathtt {ik_sel.py}$ </tex-math></inline-formula> file to achieve an optimal balance between success rate and computation efficiency based on their requirements. For integration with other simulation systems, practitioners can refer directly to the structure and logic provided in the standalone implementation of the <inline-formula> <tex-math>$mathtt {ik_sel.py}$ </tex-math></inline-formula> file. This file is fully aligned with the methods presented in this paper and offers a clear, modular reference for reproducing the proposed selection strategy.","PeriodicalId":51060,"journal":{"name":"IEEE Transactions on Automation Science and Engineering","volume":"23 ","pages":"4410-4427"},"PeriodicalIF":6.4,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146089878","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 : 2026-01-28DOI: 10.1109/tase.2026.3659055
Lin Li, Ziyang Chen, Zhen Kan
{"title":"Environment-Driven and LLM-Guided Multi-Robot Task Inference and Allocation under Temporal Logic Specifications","authors":"Lin Li, Ziyang Chen, Zhen Kan","doi":"10.1109/tase.2026.3659055","DOIUrl":"https://doi.org/10.1109/tase.2026.3659055","url":null,"abstract":"","PeriodicalId":51060,"journal":{"name":"IEEE Transactions on Automation Science and Engineering","volume":"297 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146070246","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}
This paper presents a control framework designed to enhance the stability and robustness of legged robots in the presence of uncertainties, including model uncertainties, external disturbances, and faults. The framework enables the full-state feedback estimator to estimate and compensate for uncertainties in the whole-body dynamics of the legged robots. First, we propose a novel moving horizon extended state observer (MH-ESO) to estimate uncertainties and mitigate noise in legged systems, which can be integrated into the framework for disturbance compensation. Second, we introduce a three-level whole-body disturbance rejection control framework (T-WB-DRC). Unlike the previous two-level approach, this three-level framework considers both the plan based on whole-body dynamics without uncertainties and the plan based on dynamics with uncertainties, significantly improving payload transportation, external disturbance rejection, and fault tolerance. Third, simulations of both humanoid and quadruped robots in the Gazebo simulator demonstrate the effectiveness and versatility of T-WB-DRC. Finally, extensive experimental trials on a quadruped robot validate the robustness and stability of the system when using T-WB-DRC under various disturbance conditions. Note to Practitioners—This paper presents a practical control framework to significantly improve the robustness of legged robots against real-world uncertainties like unknown payloads, external pushes, and actuator faults. Its core is a novel three-level whole-body controller (T-WB-DRC) that uses a moving horizon estimator (MH-ESO) to accurately identify and compensate for disturbances in real-time. This dual-planning approach, which considers both ideal and disturbance-injected dynamics, outperforms previous methods. The framework’s effectiveness in enhancing stability under disturbances has been successfully validated through extensive simulations and physical experiments on a quadruped robot.
{"title":"A Three-Level Whole-Body Disturbance Rejection Control Framework for Dynamic Motions in Legged Robots","authors":"Bolin Li;Gewei Zuo;Zhixiang Wang;Xiaotian Ke;Lijun Zhu;Han Ding","doi":"10.1109/TASE.2026.3657333","DOIUrl":"10.1109/TASE.2026.3657333","url":null,"abstract":"This paper presents a control framework designed to enhance the stability and robustness of legged robots in the presence of uncertainties, including model uncertainties, external disturbances, and faults. The framework enables the full-state feedback estimator to estimate and compensate for uncertainties in the whole-body dynamics of the legged robots. First, we propose a novel moving horizon extended state observer (MH-ESO) to estimate uncertainties and mitigate noise in legged systems, which can be integrated into the framework for disturbance compensation. Second, we introduce a three-level whole-body disturbance rejection control framework (T-WB-DRC). Unlike the previous two-level approach, this three-level framework considers both the plan based on whole-body dynamics without uncertainties and the plan based on dynamics with uncertainties, significantly improving payload transportation, external disturbance rejection, and fault tolerance. Third, simulations of both humanoid and quadruped robots in the Gazebo simulator demonstrate the effectiveness and versatility of T-WB-DRC. Finally, extensive experimental trials on a quadruped robot validate the robustness and stability of the system when using T-WB-DRC under various disturbance conditions. Note to Practitioners—This paper presents a practical control framework to significantly improve the robustness of legged robots against real-world uncertainties like unknown payloads, external pushes, and actuator faults. Its core is a novel three-level whole-body controller (T-WB-DRC) that uses a moving horizon estimator (MH-ESO) to accurately identify and compensate for disturbances in real-time. This dual-planning approach, which considers both ideal and disturbance-injected dynamics, outperforms previous methods. The framework’s effectiveness in enhancing stability under disturbances has been successfully validated through extensive simulations and physical experiments on a quadruped robot.","PeriodicalId":51060,"journal":{"name":"IEEE Transactions on Automation Science and Engineering","volume":"23 ","pages":"4233-4246"},"PeriodicalIF":6.4,"publicationDate":"2026-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146070243","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}
In large-scale shipbuilding, welding tasks represent a significant portion of all tasks, requiring automated robot operation. However, current welding robots are not automated for high and narrow spaces because they cannot pull heavy welding cables, causing deviations from the intended path and reducing the welding accuracy. This paper proposes a cable-towing stabilization method considering factors such as the self-weight of multiple vehicles, magnetic adhesion force, and force necessary to hold the cables. The proposed approach integrates welding robots, which perform welding tasks, with towing robots, which alleviate the load imposed by welding cables and wire feeders. Cable-towing on walls requires reducing excessive distances between vehicles, as well as their excessive acceleration, in addition to maintaining the mechanical stability of each vehicle. Therefore, the optimal positions and postures of the towing vehicles are sequentially calculated using an optimization problem. The proposed method was evaluated through simulations and real-world experiments, confirming stable cable-towing on a wall surface of approximately <inline-formula> <tex-math>$3times 1.5$ </tex-math></inline-formula> m. The findings of this research enhance the safety and efficiency of managing deformable linear objects with robots, focusing on mechanical safety while expanding the operational range from single-vehicle wall-mounted operations to cooperative multi-vehicle wall-mounted tasks, thereby increasing applicability to various wall-towing scenarios. Note to Practitioners—This study addressed a key challenge in automating welding tasks on vertical or confined surfaces: managing heavy and flexible welding cables. During manual operations, workers naturally compensate for cable weight and slack. However, in automated systems, these cables often cause robots to deviate from their intended paths, reducing accuracy and stability. To address this issue, we propose a system that coordinates welding and towing robots designed to relieve cable tension. The towing robots are magnetically adhered to the surface and repositioned through optimization to ensure stable cable guidance while minimizing excessive motion or distance. This enables the welding robot to operate more precisely and reliably, even on steep wall surfaces. The proposed system was validated through simulations and real-world experiments on a <inline-formula> <tex-math>$3~{times }~1.5$ </tex-math></inline-formula> m vertical wall, confirming its feasibility for industrial settings. The proposed approach facilitates scalable, multi-robot collaboration in narrow or elevated environments, which are common in large-scale shipbuilding and plant maintenance. A significant avenue for future development lies in extending the proposed method to handle curved or irregular surfaces. Real-world infrastructure are rarely flat, and adapting the coordination strategy to such geometries would further improve the versatility
{"title":"Mobile Robot System for Optimal Towing Welding Cables on Walls","authors":"Kazuya Oguma;Yoshito Okada;Hirokazu Fujimoto;Kenichi Murano;Haruhiko Eto;Kazunori Ohno;Kenjiro Tadakuma;Satoshi Tadokoro","doi":"10.1109/TASE.2026.3656652","DOIUrl":"10.1109/TASE.2026.3656652","url":null,"abstract":"In large-scale shipbuilding, welding tasks represent a significant portion of all tasks, requiring automated robot operation. However, current welding robots are not automated for high and narrow spaces because they cannot pull heavy welding cables, causing deviations from the intended path and reducing the welding accuracy. This paper proposes a cable-towing stabilization method considering factors such as the self-weight of multiple vehicles, magnetic adhesion force, and force necessary to hold the cables. The proposed approach integrates welding robots, which perform welding tasks, with towing robots, which alleviate the load imposed by welding cables and wire feeders. Cable-towing on walls requires reducing excessive distances between vehicles, as well as their excessive acceleration, in addition to maintaining the mechanical stability of each vehicle. Therefore, the optimal positions and postures of the towing vehicles are sequentially calculated using an optimization problem. The proposed method was evaluated through simulations and real-world experiments, confirming stable cable-towing on a wall surface of approximately <inline-formula> <tex-math>$3times 1.5$ </tex-math></inline-formula> m. The findings of this research enhance the safety and efficiency of managing deformable linear objects with robots, focusing on mechanical safety while expanding the operational range from single-vehicle wall-mounted operations to cooperative multi-vehicle wall-mounted tasks, thereby increasing applicability to various wall-towing scenarios. Note to Practitioners—This study addressed a key challenge in automating welding tasks on vertical or confined surfaces: managing heavy and flexible welding cables. During manual operations, workers naturally compensate for cable weight and slack. However, in automated systems, these cables often cause robots to deviate from their intended paths, reducing accuracy and stability. To address this issue, we propose a system that coordinates welding and towing robots designed to relieve cable tension. The towing robots are magnetically adhered to the surface and repositioned through optimization to ensure stable cable guidance while minimizing excessive motion or distance. This enables the welding robot to operate more precisely and reliably, even on steep wall surfaces. The proposed system was validated through simulations and real-world experiments on a <inline-formula> <tex-math>$3~{times }~1.5$ </tex-math></inline-formula> m vertical wall, confirming its feasibility for industrial settings. The proposed approach facilitates scalable, multi-robot collaboration in narrow or elevated environments, which are common in large-scale shipbuilding and plant maintenance. A significant avenue for future development lies in extending the proposed method to handle curved or irregular surfaces. Real-world infrastructure are rarely flat, and adapting the coordination strategy to such geometries would further improve the versatility","PeriodicalId":51060,"journal":{"name":"IEEE Transactions on Automation Science and Engineering","volume":"23 ","pages":"4182-4197"},"PeriodicalIF":6.4,"publicationDate":"2026-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11367399","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146070242","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-28DOI: 10.1109/tase.2026.3658948
Weiming Liu, Guodong Wang, Xiangyu Wang
{"title":"Modular Formation Control for Multi-mobile Robots with Collision Avoidance and Communication Maintenance","authors":"Weiming Liu, Guodong Wang, Xiangyu Wang","doi":"10.1109/tase.2026.3658948","DOIUrl":"https://doi.org/10.1109/tase.2026.3658948","url":null,"abstract":"","PeriodicalId":51060,"journal":{"name":"IEEE Transactions on Automation Science and Engineering","volume":"179 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146070245","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：