IEEE Robotics and Automation Letters最新文献_第6页

DGTFNet: Depth-Guided Tri-Axial Fusion Network for Efficient Generalizable Stereo Matching 深度引导的三轴融合网络，用于高效的广义立体匹配

IF 5.3 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2025-09-04 DOI: 10.1109/LRA.2025.3606382

Seunghun Moon;Haeuk Lee;Suk-Ju Kang

{"title":"DGTFNet: Depth-Guided Tri-Axial Fusion Network for Efficient Generalizable Stereo Matching","authors":"Seunghun Moon;Haeuk Lee;Suk-Ju Kang","doi":"10.1109/LRA.2025.3606382","DOIUrl":"https://doi.org/10.1109/LRA.2025.3606382","url":null,"abstract":"Stereo matching is a crucial task in computer vision that estimates pixel-level disparities from rectified image pairs to reconstruct three-dimensional depth information. It has diverse applications, ranging from augmented reality to autonomous driving. While deep learning-based methods have achieved remarkable progress through 3D CNNs and Transformer-based architectures, their reliance on domain-specific fine-tuning and localized feature extraction often hampers robustness and generalization in real-world scenarios. This letter introduces the Depth-Guided Tri-Axial Fusion Network (DGTFNet), which overcomes these limitations by integrating depth priors from a monocular depth foundation model via the Depth-Guided Cross-Modal Attention (DGCMA) module. Additionally, we propose a Tri-Axial Attention (TAA) module that employs directional strip convolutions to capture long-range dependencies across horizontal, vertical, and spatial dimensions. Extensive evaluations on public stereo benchmarks demonstrate that DGTFNet significantly outperforms state-of-the-art methods in zero-shot evaluations. Ablation studies further validate the contribution of each module in delivering robust and efficient stereo matching.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 10","pages":"10791-10798"},"PeriodicalIF":5.3,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145036938","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}

引用次数: 0

KAN Policy: Learning Efficient and Smooth Robotic Trajectories via Kolmogorov-Arnold Networks KAN策略：基于Kolmogorov-Arnold网络的高效平滑机器人轨迹学习

IF 5.3 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2025-09-04 DOI: 10.1109/LRA.2025.3606354

Zikang Chen;Fei Gao;Ziya Yu;Peng Li

{"title":"KAN Policy: Learning Efficient and Smooth Robotic Trajectories via Kolmogorov-Arnold Networks","authors":"Zikang Chen;Fei Gao;Ziya Yu;Peng Li","doi":"10.1109/LRA.2025.3606354","DOIUrl":"https://doi.org/10.1109/LRA.2025.3606354","url":null,"abstract":"Modernrobotic visuomotor policy learning has witnessed significant progress through Diffusion Policy (DP) frameworks built upon <italic>Convolutional Neural Networks (CNNs) and Transformers. Despite their empirical success, these architectures remain fundamentally constrained by their relatively discrete computational nature, inherently limiting their capacity to generate efficient and smooth motion trajectories. To address this challenge, we introduce <italic>Kolmogorov-Arnold Networks (KANs) into Diffusion Policy learning. The proposed <italic>KAN Policy (KP) leverages KANs' intrinsic continuity through learnable base-parameterized activation functions, thereby producing continuous trajectories with shorter execution time and fewer jerks. Specifically, we design a novel <italic>Embedding KAN (Emb-KAN) for CNN-based models, which preserves structural continuity in high-dimensional latent spaces through adaptive spline embeddings. Besides, we apply Group-KAN to Transformer-based models for learning continuous representations. Across main simulation experiments, KP achieves average improvements of 6.06%, 8.03%, and 26.4% in terms of success rate, execution time, and smoothness, respectively. Similarly, in real-world experiments, KP achieves average improvements of 53.8%, 7.89%, and 29.4% across the same metrics.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 11","pages":"11164-11171"},"PeriodicalIF":5.3,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145078696","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}

引用次数: 0

ApexNAV: An Adaptive Exploration Strategy for Zero-Shot Object Navigation With Target-Centric Semantic Fusion 基于目标中心语义融合的零射击目标导航自适应探索策略ApexNAV

IF 5.3 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2025-09-04 DOI: 10.1109/LRA.2025.3606388

Mingjie Zhang;Yuheng Du;Chengkai Wu;Jinni Zhou;Zhenchao Qi;Jun Ma;Boyu Zhou

{"title":"ApexNAV: An Adaptive Exploration Strategy for Zero-Shot Object Navigation With Target-Centric Semantic Fusion","authors":"Mingjie Zhang;Yuheng Du;Chengkai Wu;Jinni Zhou;Zhenchao Qi;Jun Ma;Boyu Zhou","doi":"10.1109/LRA.2025.3606388","DOIUrl":"https://doi.org/10.1109/LRA.2025.3606388","url":null,"abstract":"Navigating unknown environments to find a target object is a significant challenge. While semantic information is crucial for navigation, relying solely on it for decision-making may not always be efficient, especially in environments with weak semantic cues. Additionally, many methods are susceptible to misdetections, especially in environments with visually similar objects. To address these limitations, we propose ApexNav, a zero-shot object navigation framework that is both more efficient and reliable. For efficiency, ApexNav adaptively utilizes semantic information by analyzing its distribution in the environment, guiding exploration through semantic reasoning when cues are strong, and switching to geometry-based exploration when they are weak. For reliability, we propose a target-centric semantic fusion method that preserves long-term memory of the target and similar objects, enabling robust object identification even under noisy detections. We evaluate ApexNav on the HM3Dv1, HM3Dv2, and MP3D datasets, where it outperforms state-of-the-art methods in both SR and SPL metrics. Comprehensive ablation studies further demonstrate the effectiveness of each module. Furthermore, real-world experiments validate the practicality of ApexNav in physical environments.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 11","pages":"11530-11537"},"PeriodicalIF":5.3,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145210184","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}

引用次数: 0

Structure-Preserving Model Order Reduction of Slender Soft Robots via Autoencoder-Parameterized Strain 基于自编码器参数化应变的细长软机器人保结构模型降阶

IF 5.3 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2025-09-04 DOI: 10.1109/LRA.2025.3606389

Abdulaziz Y. Alkayas;Anup Teejo Mathew;Daniel Feliu-Talegon;Yahya Zweiri;Thomas George Thuruthel;Federico Renda

{"title":"Structure-Preserving Model Order Reduction of Slender Soft Robots via Autoencoder-Parameterized Strain","authors":"Abdulaziz Y. Alkayas;Anup Teejo Mathew;Daniel Feliu-Talegon;Yahya Zweiri;Thomas George Thuruthel;Federico Renda","doi":"10.1109/LRA.2025.3606389","DOIUrl":"https://doi.org/10.1109/LRA.2025.3606389","url":null,"abstract":"While soft robots offer advantages in adaptability and safe interaction, their modeling remains challenging. This letter presents a novel, data-driven approach for model order reduction of slender soft robots using autoencoder-parameterized strain within the Geometric Variable Strain (GVS) framework. We employ autoencoders (AEs) to learn low-dimensional strain parameterizations from data to construct reduced-order models (ROMs), preserving the Lagrangian structure of the system while significantly reducing the degrees of freedom. Our comparative analysis demonstrates that AE-based ROMs consistently outperform proper orthogonal decomposition (POD) approaches, achieving lower errors for equivalent degrees of freedom across multiple test cases. Additionally, we demonstrate that our proposed approach achieves computational speed-ups over the high-order models (HOMs) in all cases, and outperforms the POD-based ROM in scenarios where accuracy is matched. We highlight the intrinsic dimensionality discovery capabilities of autoencoders, revealing that HOM often operate in lower-dimensional nonlinear manifolds. Through both simulation and experimental validation on a cable-actuated soft manipulator, we demonstrate the effectiveness of our approach, achieving near-identical behavior with just a single degree of freedom. This structure-preserving method offers significant reductions in the system degrees of freedom and computational effort while maintaining physical model interpretability, offering a promising direction for soft robot modeling and control.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 10","pages":"11006-11013"},"PeriodicalIF":5.3,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11150703","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145061863","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}

引用次数: 0

Previous Knowledge Utilization in Online Anytime Belief Space Planning 在线随时信念空间规划中的先验知识利用

IF 5.3 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2025-09-04 DOI: 10.1109/LRA.2025.3606381

Michael Novitsky;Moran Barenboim;Vadim Indelman

引用次数: 0

A Blockchain Framework for Equitable and Secure Task Allocation in Robot Swarms 机器人群中公平安全任务分配的区块链框架

IF 5.3 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2025-09-04 DOI: 10.1109/LRA.2025.3606349

Hanqing Zhao;Alexandre Pacheco;Giovanni Beltrame;Xue Liu;Marco Dorigo;Gregory Dudek

{"title":"A Blockchain Framework for Equitable and Secure Task Allocation in Robot Swarms","authors":"Hanqing Zhao;Alexandre Pacheco;Giovanni Beltrame;Xue Liu;Marco Dorigo;Gregory Dudek","doi":"10.1109/LRA.2025.3606349","DOIUrl":"https://doi.org/10.1109/LRA.2025.3606349","url":null,"abstract":"Recent studies demonstrate the potential of blockchain to enable robots in a swarm to achieve secure consensus about the environment, particularly when robots are homogeneous and perform identical tasks. Typically, robots receive rewards for their contributions to consensus achievement, but no studies have yet targeted heterogeneous swarms, in which the robots have distinct physical capabilities suited to different tasks. We present a novel framework that leverages domain knowledge to decompose the swarm mission into a hierarchy of tasks within smart contracts. This allows the robots to reach a consensus about both the environment and the action plan, allocating tasks among robots with diverse capabilities to improve their performance while maintaining security against faults and malicious behaviors. We refer to this concept as <italic>equitable and secure task allocation. Validated in Simultaneous Localization and Mapping missions, our approach not only achieves equitable task allocation among robots with varying capabilities, improving mapping accuracy and efficiency, but also shows resilience against malicious attacks.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 10","pages":"10862-10869"},"PeriodicalIF":5.3,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145036746","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}

引用次数: 0

Receding Horizon Control for Signal Temporal Logic Using Robustness-Conserving Partial Formula Evaluation 基于保鲁棒部分公式计算的信号时间逻辑后退水平控制

IF 5.3 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2025-09-04 DOI: 10.1109/LRA.2025.3606350

Roland Ilyes;Lara Brudermüller;Nick Hawes;Bruno Lacerda

引用次数: 0

Spin Swimmer : A Fast, Efficient and Agile Fish-Like Robot 旋转游泳者：一种快速、高效、敏捷的鱼状机器人

IF 5.3 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2025-09-04 DOI: 10.1109/LRA.2025.3606357

Prashanth Chivkula;Phanindra Tallapragada

{"title":"Spin Swimmer : A Fast, Efficient and Agile Fish-Like Robot","authors":"Prashanth Chivkula;Phanindra Tallapragada","doi":"10.1109/LRA.2025.3606357","DOIUrl":"https://doi.org/10.1109/LRA.2025.3606357","url":null,"abstract":"Engineers and scientists designing underwater robots have sought to emulate the speed, efficiency, and agility of fish. Much of the engineering of fish-like robotics reduces to the design of soft or articulated multi-body tails that can oscillate or undulate at frequencies and amplitudes similar to those of the fish they seek to mimic in the hope of achieving their efficiency and speed. Such kinematic approaches do not account for the dynamic interaction between power efficient actuation, response of flexible appendages and hydrodynamic forces. This letter presents a fundamentally novel means of mechanical actuation: a fast spinning unbalanced rotor internal to the body of the robot, that transfers a periodic axial force to an otherwise passive flexible tail. The net result is that the tail acts as a parametric oscillator that undergoes a <inline-formula><tex-math>$2:1$</tex-math></inline-formula> subharmonic resonance. High tail-beat frequencies are achieved with minimal input power due to this parametric resonance. The resulting robot has the lowest cost of transport amongst free swimming robots while also being fast, extremely agile and gyroscopically roll and pitch stable. The results demonstrate the importance of exploiting parametric resonances in designing efficient fish-like robots.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 10","pages":"10942-10949"},"PeriodicalIF":5.3,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145061889","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}

引用次数: 0

Unveiling $SO(3)$ Parallel Robot Variants: Application of the Optimal Robot to a Humanoid Eye 揭示$SO(3)$并联机器人变体：最优机器人在仿人眼上的应用

IF 5.3 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2025-09-04 DOI: 10.1109/LRA.2025.3606380

Hassen Nigatu;Jihao Li;Gaokun Shi;Jianguo Wang;Guodong Lu;Howard Li;Huixu Dong

{"title":"Unveiling $SO(3)$ Parallel Robot Variants: Application of the Optimal Robot to a Humanoid Eye","authors":"Hassen Nigatu;Jihao Li;Gaokun Shi;Jianguo Wang;Guodong Lu;Howard Li;Huixu Dong","doi":"10.1109/LRA.2025.3606380","DOIUrl":"https://doi.org/10.1109/LRA.2025.3606380","url":null,"abstract":"This study presents a systematic motion analysis and classification of <inline-formula><tex-math>$SO(3)$</tex-math></inline-formula>-type parallel robot variants using an analytical Lie algebra approach. These robots are known for their ability to perform arbitrary rotations around a fixed point, making them suitable for various applications. Despite their architectural diversity, existing research has largely treated them on a case-by-case basis, limiting the exploration of all potential variants and the benefits derived from this diversity. By applying a generalized analytical approach through the reciprocal screw method, we systematically examine the kinematic conditions for limbs that generate <inline-formula><tex-math>$SO(3)$</tex-math></inline-formula> motion. As a result, we identify 73 distinct non-redundant limb types capable of producing the desired <inline-formula><tex-math>$SO(3)$</tex-math></inline-formula> motion. Our approach includes an in-depth algebraic motion-constraint analysis, uncovering common characteristics across different variants. This leads us to identify 73 symmetric and 5,256 asymmetric variants, for a total of 5,329, each with unique capabilities. Finally, we selected a computationally optimized, miniaturized robot from this set for use in a humanoid eye system.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 11","pages":"11227-11234"},"PeriodicalIF":5.3,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145090005","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}

引用次数: 0

Light Reflection-Guided RRT$^{*}$: Efficient Path Planning in Narrow Passages 光反射引导RRT$^{*}$：狭窄通道的有效路径规划

IF 5.3 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2025-09-04 DOI: 10.1109/LRA.2025.3606385

Xiaotong Xun;Runda Zhang;Senchun Chai;Runqi Chai;Yuanqing Xia

引用次数: 0