IEEE Robotics and Automation Letters最新文献

{"title":"Riemannian Time Warping: Multiple Sequence Alignment in Curved Spaces","authors":"Julian Richter;Christopher A. Erdős;Christian Scheurer;Jochen J. Steil;Niels Dehio","doi":"10.1109/LRA.2025.3604734","DOIUrl":"https://doi.org/10.1109/LRA.2025.3604734","url":null,"abstract":"Temporal alignment of multiple signals through time warping is crucial in many fields, such as classification within speech recognition or robot motion learning. Almost all related works are limited to data in Euclidean space. Although an attempt was made in 2011 to adapt this concept to unit quaternions, a general extension to Riemannian manifolds remains absent. Given its importance for numerous applications in robotics and beyond, we introduce Riemannian Time Warping (RTW). This novel approach efficiently aligns multiple signals by considering the geometric structure of the Riemannian manifold in which the data is embedded. Extensive experiments on synthetic and real-world data, including tests with an LBR iiwa robot, demonstrate that RTW consistently outperforms state-of-the-art baselines in both averaging and classification tasks.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 10","pages":"10894-10901"},"PeriodicalIF":5.3,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145061911","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":"CTS-PIBT: An Efficient Method for Multi-Agent Collaborative Task Sequencing and Path Finding","authors":"Junkai Jiang;Yibin Yang;Ruochen Li;Yitao Xu;Shaobing Xu;Jianqiang Wang","doi":"10.1109/LRA.2025.3604726","DOIUrl":"https://doi.org/10.1109/LRA.2025.3604726","url":null,"abstract":"The Collaborative Task Sequencing and Multi-Agent Path Finding (CTS-MAPF) problem involves planning collision-free paths for multiple agents while determining the sequence of intermediate tasks. This problem is particularly challenging due to its combinatorial complexity, as it combines both task sequencing and pathfinding. This letter introduces CTS-PIBT, a novel and efficient algorithm designed to address the CTS-MAPF problem. CTS-PIBT adopts a hierarchical framework with three key components: task sequencing, solution finding following the sequence, and a low-level search using an extended version of PIBT. This framework effectively leverages the advantages of the configuration-based approach, enabling the rapid generation of feasible solutions within a short period. To further enhance performance, we incorporate an anytime refinement mechanism and a quick task sequencing technique (called greedy insertion with 2-opt) to improve solution quality and solving efficiency. Extensive simulations demonstrate that CTS-PIBT significantly outperforms existing methods in success rate and runtime, particularly in large-scale and complex scenarios. Furthermore, physical robot experiments validate its practical applicability in real-world environments.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 10","pages":"10626-10633"},"PeriodicalIF":5.3,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145049814","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":"IAE-BEV:Instance-Adaptive Enhancement for BEV-Based Multi-View 3D Object Detection","authors":"Bin Wang;Qiang Zhao;Chongben Tao;Yaoqi Sun;Chenggang Yan","doi":"10.1109/LRA.2025.3604760","DOIUrl":"https://doi.org/10.1109/LRA.2025.3604760","url":null,"abstract":"Camera-based Bird's-Eye-View (BEV) representation has become a viable solution for 3D object detection in cost-effective autonomous driving. Currently, the explicit paradigm based on the lift-splat-shoot (LSS) pipeline has become one of the mainstream methods due to its efficiency and ease of deployment. However, the process of flattening the spatial representation in this pipeline mixes target features with excessive background noise. In addition, the generated BEV features are sparse due to the inherent characteristics of camera imaging. To address these limitations, we propose IAE-BEV, a novel two-stage multi-view 3D object detector that adaptively integrates instance features into BEV features, ultimately constructing a BEV representation that highlights instance information and alleviates sparsity. We also introduce the Occupancy Mask Pool, which enables instance features to interact with the 2D image plane in a more targeted and efficient manner. To further distinguish instances along the same camera ray, we design the Angle-Adaptive Self-Attention, which learns appropriate feature weights under the guidance of queries. Extensive experiments on the nuScenes dataset demonstrate the effectiveness and generalizability of our proposed framework, achieving up to +2.96% mAP improvement over state-of-the-art LSS-based baseline.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 10","pages":"10610-10617"},"PeriodicalIF":5.3,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145050991","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":"Superelastic Tendon-Like Bowden Cables: Advancing Assistive Exoskeletons","authors":"Gregorio Pisaneschi;José M. Catalán;Andrea Blanco;Nicola Sancisi;Nicolas García;Andrea Zucchelli","doi":"10.1109/LRA.2025.3605094","DOIUrl":"https://doi.org/10.1109/LRA.2025.3605094","url":null,"abstract":"This study introduces a novel Bowden cable (BC) system for hand-assistive exoskeletons employing superelastic (SE) shape memory alloy wires to address key limitations such as efficiency and safety limitations. The unique properties of SE wires enable a single-wire transmission, offering enhanced performance, plus inherent self-sensing and self-limiting capabilities that provide tendon-like overload protection. Experimental results obtained with a setup simulating use conditions demonstrate the superior efficiency of SE wires, with 1/4 the friction of conventional steel cables. In addition, a validated force-sensing capability, achieved by monitoring electrical resistance, proves to accurately detect overloads within 1% force error. This, along with the inherent passive force self-limiting behaviour during simulated collisions, demonstrates the ability of the SE BC to effectively mimic the protective function of biological tendons. Therefore, this biomimetic innovation in soft robotic transmission significantly improves safety and efficiency, presenting a promising advancement for human-robot interaction in assistive and rehabilitative robotics.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 10","pages":"10761-10766"},"PeriodicalIF":5.3,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145050786","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":"Optimal Deployment of Multi-Robot Based Regional Positioning System Using Relative Bearing Measurement","authors":"Kangwen Lin;Liang Zhang;Keyue Wu;Zhaochen Liu","doi":"10.1109/LRA.2025.3604740","DOIUrl":"https://doi.org/10.1109/LRA.2025.3604740","url":null,"abstract":"Recently, the Multi-Robot based Regional Positioning System (MR-RPS) utilizing cooperative localization have emerged as a promising paradigm in Global Navigation Satellite Systems (GNSS)-constraint region: robots first localize themselves using extravagant sensors and then serve as anchors to propagate their noisy positions via cooperative localization and communication with users using accessible devices. However, research on the optimal deployment strategies for bearing-only sensor-based MR-RPS, such as using cameras, remains limited. The bearing-only approach provides a cost-effective and energy-efficient alternative to range-based solutions such as radio systems. This letter proposes a distributed optimal strategy for MR-RPS equipped with bearing-only sensors to provide effective positioning services to the GNSS-limited scenarios. We first employ the Fisher Information Matrix (FIM) and D-optimality as the localziation performance metric using the information received from the Multi-Robot System (MRS). Inspired by the coverage control problem from the Wireless Sensor Network, the optimal deployment of the MRS is then formulated into a locational optimization problem by integrating the value of D-optimality over the entire mission space. A distributed deployment strategy is thus developed, enabling each robot to iteratively compute optimal deployment strategies to enhance the overall performance for stochastically appearing users. Extensive simulations and physical experiments validate the resilience and effectiveness of the proposed method in delivering high-quality positioning services.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 10","pages":"10602-10609"},"PeriodicalIF":5.3,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145051057","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":"DSFormer: A Dual-Scale Cross-Learning Transformer for Visual Place Recognition","authors":"Haiyang Jiang;Songhao Piao;Chao Gao;Lei Yu;Liguo Chen","doi":"10.1109/LRA.2025.3604761","DOIUrl":"https://doi.org/10.1109/LRA.2025.3604761","url":null,"abstract":"Visual Place Recognition (VPR) is crucial for robust mobile robot localization, yet it faces significant challenges in maintaining reliable performance under varying environmental conditions and viewpoints. To address this, we propose a novel framework that integrates Dual-Scale-Former (DSFormer), a Transformer-based cross-learning module, with an innovative block clustering strategy. DSFormer enhances feature representation by enabling bidirectional information transfer between dual-scale features extracted from the final two CNN layers, capturing both semantic richness and spatial details through self-attention for long-range dependencies within each scale and shared cross-attention for cross-scale learning. Complementing this, our block clustering strategy repartitions the widely used San Francisco eXtra Large (SF-XL) training dataset from multiple distinct perspectives, optimizing data organization to further bolster robustness against viewpoint variations. Together, these innovations not only yield a robust global embedding adaptable to environmental changes but also reduce the required training data volume by approximately 30% compared to previous partitioning methods. Comprehensive experiments demonstrate that our approach achieves state-of-the-art performance across most benchmark datasets, surpassing advanced reranking methods like DELG, Patch-NetVLAD, TransVPR, and R2Former as a global retrieval solution using 512-dim global descriptors, while significantly improving computational efficiency.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 10","pages":"10799-10806"},"PeriodicalIF":5.3,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145036927","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":"DiffRP: Diffusion-Driven Promising Region Prediction for Sampling-Based Path Planning","authors":"Zongwu Xie;Yiming Ji;Yang Liu;Yiqian Xie;Zhengpu Wang;Boyu Ma;Baoshi Cao","doi":"10.1109/LRA.2025.3604747","DOIUrl":"https://doi.org/10.1109/LRA.2025.3604747","url":null,"abstract":"Utilizing neural networks to predict potential regions containing optimal paths in advance and subsequently biasing the sampling probability towards these promising regions has been proven to effectively enhance the path planning efficiency of sampling-based algorithms. Undoubtedly, the accuracy of the promising regions is of paramount importance. Currently, the generalizability of many CNN- or Transformer-based models remains limited, often performing poorly in unknown environments. To enhance generalization capability, we reformulate the promising region prediction problem as a conditional generation task and address it using a diffusion model, referred to as the DiffRP (Diffusion-based Region Prediction). We propose three paradigms for generating promising regions, among which we innovatively introduce a biased noise initialization method for the diffusion process. Specifically, we bias the mean of the noise distribution using obstacle maps and design a map-conditioned denoising model to progressively generate accurate promising regions from the biased noise. Experiments on public datasets demonstrate that our proposed DiffRP method outperforms existing state-of-the-art models by 35<inline-formula><tex-math>$sim$</tex-math></inline-formula>42% in promising region prediction accuracy. Moreover, the non-uniform sampling algorithm (DiffRP-RRT*) based on this region achieves a 3<inline-formula><tex-math>$sim$</tex-math></inline-formula>52% reduction in sample number compared with other neural-network-driven approaches.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 10","pages":"10753-10760"},"PeriodicalIF":5.3,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145050842","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":"Enhanced Robust Locomotion of Wheeled-Bipedal Robot via Hierarchical Optimization and Online Wheel Position Planning","authors":"Yu Wang;Teng Chen;Shao Xu;Xuewen Rong;Guoteng Zhang;Yaxian Xin;Yibin Li","doi":"10.1109/LRA.2025.3604748","DOIUrl":"https://doi.org/10.1109/LRA.2025.3604748","url":null,"abstract":"In this letter, we present a novel control framework for wheeled-bipedal robots to address the challenges posed by underactuation characteristics and complex dynamics coupling. An integrated dynamic model is constructed by combining wheel dynamics and centroidal dynamics of the bipedal body based on rolling constraints and interaction force transmission, facilitating dynamic coordination between wheels and the base. Considering the non-minimum phase behavior, the online dynamic planner captures the fundamental dynamics of wheeled-bipedal robots to generate wheel position constraints, ensuring adaptation to the current center of mass (CoM) height and dynamic balance requirements. A hierarchical optimization control framework integrating model predictive control (MPC) and weighted multi-task whole-body control (WM-WBC) is proposed, taking into account the full-body dynamics, nonholonomic constraints, optimal interaction forces, and multi-task coordination. Experimental results demonstrate that the proposed method achieves precise trajectory tracking, compliant adaptation to various terrains, and exhibits superior robustness against disturbances.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 10","pages":"10918-10925"},"PeriodicalIF":5.3,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145061950","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":"High-Precision Autonomous Control of Flexible Needles via Real-Time Finite Element Simulation and Cross-Entropy Optimization","authors":"Yanzhou Wang;Chang Chang;Junling Mei;Simon Leonard;Russell Taylor;Iulian Iordachita","doi":"10.1109/LRA.2025.3604744","DOIUrl":"https://doi.org/10.1109/LRA.2025.3604744","url":null,"abstract":"This letter presents a unified framework for autonomous flexible needle control in soft tissues using real-time finite element (FE) simulation and cross-entropy (CE) optimization. The method combines a sampling-based model predictive controller (MPC) for trajectory tracking with a kinematic-based bang-bang strategy to coordinate needle insertion, lateral adjustments, and bevel rotations. Sparse electromagnetic (EM) tracking feedback enables needle state reconstruction and compensates for model uncertainties. Experiments in plastisol and <italic>ex vivo</i> chicken breast phantoms show sub-millimeter targeting accuracy, with respective targeting errors <inline-formula><tex-math>$0.16 pm 0.29$</tex-math></inline-formula> mm and <inline-formula><tex-math>$0.22 pm 0.78$</tex-math></inline-formula> mm as reported by the tracker.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 10","pages":"10578-10585"},"PeriodicalIF":5.3,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144998202","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":"B$^{*}$: Efficient and Optimal Base Placement for Fixed-Base Manipulators","authors":"Zihang Zhao;Leiyao Cui;Sirui Xie;Saiyao Zhang;Zhi Han;Lecheng Ruan;Yixin Zhu","doi":"10.1109/LRA.2025.3604741","DOIUrl":"https://doi.org/10.1109/LRA.2025.3604741","url":null,"abstract":"Proper base placement is crucial for task execution feasibility and performance of fixed-base manipulators, the dominant solution in robotic automation. Current methods rely on pre-computed kinematics databases generated through sampling to search for solutions. However, they face an inherent trade-off between solution optimality and computational efficiency when determining sampling resolution—a challenge that intensifies when considering long-horizon trajectories, self-collision avoidance, and task-specific requirements. To address these limitations, we present B<inline-formula><tex-math>$^{*}$</tex-math></inline-formula>, a novel optimization framework for determining the optimal base placement that unifies these multiple objectives without relying on pre-computed databases. B<inline-formula><tex-math>$^{*}$</tex-math></inline-formula> addresses this inherently non-convex problem via a two-layer hierarchical approach: The outer layer systematically manages terminal constraints through progressively tightening them, particularly the base mobility constraint, enabling feasible initialization and broad solution space exploration. Concurrently, the inner layer addresses the non-convexities of each outer-layer subproblem by sequential local linearization, effectively transforming the original problem into a tractable sequential linear program (SLP). Comprehensive evaluations across multiple robot platforms and task complexities demonstrate the effectiveness of B<inline-formula><tex-math>$^{*}$</tex-math></inline-formula>: it achieves solution optimality five orders of magnitude better than sampling-based approaches while maintaining perfect success rates, all with reduced computational overhead. Operating directly in configuration space, B<inline-formula><tex-math>$^{*}$</tex-math></inline-formula> not only solves the base placement problem but also enables simultaneous path planning with customizable optimization criteria, making it a versatile framework for various robotic motion planning challenges. B<inline-formula><tex-math>$^{*}$</tex-math></inline-formula> serves as a crucial initialization tool for robotic applications, bridging the gap between theoretical motion planning and practical deployment where feasible trajectory existence is fundamental.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 10","pages":"10634-10641"},"PeriodicalIF":5.3,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145050824","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}