{"title":"FORGE: Force-Guided Exploration for Robust Contact-Rich Manipulation Under Uncertainty","authors":"Michael Noseworthy;Bingjie Tang;Bowen Wen;Ankur Handa;Chad Kessens;Nicholas Roy;Dieter Fox;Fabio Ramos;Yashraj Narang;Iretiayo Akinola","doi":"10.1109/LRA.2025.3551637","DOIUrl":"https://doi.org/10.1109/LRA.2025.3551637","url":null,"abstract":"We present FORGE, a method for sim-to-real transfer of force-aware manipulation policies in the presence of significant pose uncertainty. During simulation-based policy learning, FORGE combines a <italic>force threshold</i> mechanism with a <italic>dynamics randomization</i> scheme to enable robust transfer of the learned policies to the real robot. At deployment, FORGE policies, conditioned on a maximum allowable force, adaptively perform contact-rich tasks while avoiding aggressive and unsafe behaviour, regardless of the controller gains. Additionally, FORGE policies predict task success, enabling efficient termination and autonomous tuning of the force threshold. We show that FORGE can be used to learn a variety of robust contact-rich policies, including the forceful insertion of snap-fit connectors. We further demonstrate the multistage assembly of a planetary gear system, which requires success across three assembly tasks: nut threading, insertion, and gear meshing.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 5","pages":"4436-4443"},"PeriodicalIF":4.6,"publicationDate":"2025-03-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143706808","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}
Vincent S. Boon;Brendon Ortolano;Andrew J. Gunnell;Margaret Meagher;Rosemarie C. Murray;Lukas Gabert;Tommaso Lenzi
{"title":"The Only Way Is Up: Active Knee Exoskeleton Reduces Muscular Effort in Quadriceps During Weighted Stair Ascent","authors":"Vincent S. Boon;Brendon Ortolano;Andrew J. Gunnell;Margaret Meagher;Rosemarie C. Murray;Lukas Gabert;Tommaso Lenzi","doi":"10.1109/LRA.2025.3551543","DOIUrl":"https://doi.org/10.1109/LRA.2025.3551543","url":null,"abstract":"Firefighters consistently rank stair ascent with gear, which can weigh over 35 kg, as their most demanding activity. Weighted stair climbing requires dynamic motions and large knee torques, which can cause exhaustion in the short term, and overuse injuries in the long term. An active knee exoskeleton could potentially alleviate the burden on the wearer by injecting positive energy at key phases of the gait cycle. Similar devices have reduced the metabolic cost for various locomotion activities in previous studies. However, no information is available on the effect of active knee exoskeletons on muscular effort during prolonged weighted stair ascent. Here we show that our knee exoskeletons reduce the net muscular effort in the lower limbs when ascending several flights of stairs while wearing additional weight. In a task analogous to part of the physical fitness test for firefighters in the US, eight participants climbed stairs for three minutes at a constant pace while wearing a 9.1 kg vest. We compared lower limb muscle activation required to perform the task with and without two bilaterally worn Utah Knee Exoskeletons. We found that bilateral knee assistance reduced average peak quadriceps muscle activation measured through surface electromyography by 32% while reducing overall muscle activity at the quadriceps by 29%. These results suggest that an active knee exoskeleton can lower the overall muscular effort required to ascend stairs while weighted. In turn, this could aid firefighters by preserving energy for fighting fires and reducing overexertion injuries.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 5","pages":"4484-4491"},"PeriodicalIF":4.6,"publicationDate":"2025-03-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10925862","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143716455","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}
{"title":"Robust Nonprehensile Object Transportation With Uncertain Inertial Parameters","authors":"Adam Heins;Angela P. Schoellig","doi":"10.1109/LRA.2025.3551067","DOIUrl":"https://doi.org/10.1109/LRA.2025.3551067","url":null,"abstract":"We consider the nonprehensile object transportation task known as the <italic>waiter's problem</i>—in which a robot must move an object on a tray from one location to another—when the transported object has uncertain inertial parameters. In contrast to existing approaches that completely ignore uncertainty in the inertia matrix or which only consider small parameter errors, we are interested in pushing the limits of the amount of inertial parameter uncertainty that can be handled. We first show how constraints that are robust to inertial parameter uncertainty can be incorporated into an optimization-based motion planning framework to transport objects while moving quickly. Next, we develop necessary conditions for the inertial parameters to be realizable on a bounding shape based on moment relaxations, allowing us to verify whether a trajectory will violate the constraints for <italic>any</i> realizable inertial parameters. Finally, we demonstrate our approach on a mobile manipulator in simulations and real hardware experiments: our proposed robust constraints consistently successfully transport a 56 cm tall object with substantial inertial parameter uncertainty in the real world, while the baseline approaches drop the object while transporting it.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 5","pages":"4492-4499"},"PeriodicalIF":4.6,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10925467","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143716545","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}
{"title":"Language-Driven Policy Distillation for Cooperative Driving in Multi-Agent Reinforcement Learning","authors":"Jiaqi Liu;Chengkai Xu;Peng Hang;Jian Sun;Mingyu Ding;Wei Zhan;Masayoshi Tomizuka","doi":"10.1109/LRA.2025.3551098","DOIUrl":"https://doi.org/10.1109/LRA.2025.3551098","url":null,"abstract":"The cooperative driving technology of Connected and Autonomous Vehicles (CAVs) is crucial for improving the efficiency and safety of transportation systems. Learning-based methods, such as Multi-Agent Reinforcement Learning (MARL), have demonstrated strong capabilities in cooperative decision-making tasks. However, existing MARL approaches still face challenges in terms of learning efficiency and performance. In recent years, Large Language Models (LLMs) have rapidly advanced and shown remarkable abilities in various sequential decision-making tasks. To enhance the learning capabilities of cooperative agents while ensuring decision-making efficiency and cost-effectiveness, we propose LDPD, a language-driven policy distillation method for guiding MARL exploration. In this framework, a teacher agent based on LLM trains smaller student agents to achieve cooperative decision-making through its own decision-making demonstrations. The teacher agent enhances the observation information of CAVs and utilizes LLMs to perform complex cooperative decision-making reasoning, which also leverages carefully designed decision-making tools to achieve expert-level decisions, providing high-quality teaching experiences. The student agent then refines the teacher's prior knowledge into its own model through gradient policy updates. The experiments demonstrate that the students can rapidly improve their capabilities with minimal guidance from the teacher and eventually surpass the teacher's performance. Extensive experiments show that our approach demonstrates better performance and learning efficiency compared to baseline methods.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 5","pages":"4292-4299"},"PeriodicalIF":4.6,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143688097","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}
Yidi Zhang;Fulin Tang;Zewen Xu;Yihong Wu;Pengju Ma
{"title":"PGD-VIO: A Plane-Aided RGB-D Inertial Odometry With Graph-Based Drift Suppression","authors":"Yidi Zhang;Fulin Tang;Zewen Xu;Yihong Wu;Pengju Ma","doi":"10.1109/LRA.2025.3550835","DOIUrl":"https://doi.org/10.1109/LRA.2025.3550835","url":null,"abstract":"Generally, high-level features provide more geometrical information compared to point features, which can be exploited to further constrain motions. Planes are commonplace in man-made environments, offering an active means to reduce drift, due to their extensive spatial and temporal observability. To make full use of planar information, we propose a novel visual-inertial odometry using an RGB-D camera and an inertial measurement unit, effectively integrating point and plane features in an extended Kalman filter framework. Depth information of point features is leveraged to improve the accuracy of point triangulation, while plane features serve as direct observations added into the state vector. Notably, to benefit long-term navigation, a novel graph-based drift detection strategy is proposed to search overlapping and identical structures in the plane map so that the cumulative drift is suppressed subsequently. The experimental results on two public datasets demonstrate that our system outperforms state-of-the-art methods in localization accuracy and meanwhile generates a compact and consistent plane map.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 5","pages":"4276-4283"},"PeriodicalIF":4.6,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143688077","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}
Lennox Mart;Muhammad Akmal Bin Mohammed Zaffir;Sumitaka Honji;Takahiro Wada
{"title":"Development of a Combination Device of Vibration Tactile Device and Tightening Device to Realize Human-Robot Handover Operation","authors":"Lennox Mart;Muhammad Akmal Bin Mohammed Zaffir;Sumitaka Honji;Takahiro Wada","doi":"10.1109/LRA.2025.3550702","DOIUrl":"https://doi.org/10.1109/LRA.2025.3550702","url":null,"abstract":"With the ever-increasing spread of collaborative robotics, humans and robots working side by side in the workplace have become more common. When working on different subtasks of a bigger main task, the user and robot need only interact on a few occasions, like a handover, instead of being in constant contact. Interacting with the robot becomes a secondary focus for the human and should not distract from the main task. To free the human from having to keep their attention on the robot while ensuring an efficient handover, the use of physical stimuli is suggested. These signals allow the user to understand the robot's state while keeping their other senses free. In this research, we aim to combine two devices: one capable of informing the user about the handover position through vibrations and the other using tightening signals to inform about the state of the gripper. Through this combination, we expect an increase in handover efficiency and better human concentration on the main task. Experiments are performed to compare the completion of a handover task with and without the device.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 4","pages":"4109-4116"},"PeriodicalIF":4.6,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143676020","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":"TSO-BoW: Accurate Long-Term Loop Closure Detection With Constant Query Time via Online Bag of Words and Trajectory Segmentation","authors":"Shufang Zhang;Jiazheng Wu;Kaiyi Wang;Sanpeng Deng","doi":"10.1109/LRA.2025.3550799","DOIUrl":"https://doi.org/10.1109/LRA.2025.3550799","url":null,"abstract":"This letter presents TSO-BoW, a lightweight trajectory segmentation-based Bag-of-Words algorithm for loop closure detection, utilizing intermittent online training for collected segments. In the online training phase, segments of collected data form sub-trajectories that are used for online training based on their features, ultimately creating corresponding sub-databases for querying. In the querying phase, we use a multiple-level querying approach. Initially, candidate sub-databases are selected based on geometric distance using prior pose information. Subsequently, a lower bound criterion is applied to filter out some sub-databases, followed by PnP-RANSAC for geometric verification and precise relative pose estimation. Our algorithm mitigates the pose drift issue in prior pose selection-based loop detection algorithms by using a segmented Bag-of-Words and lower bound elimination. It maintains constant query time and memory cost without compromising query performance in long-term (Simultaneous localization and mapping) SLAM. Evaluations on large-scale public datasets demonstrate our algorithm's excellent computational and memory efficiency, query time efficiency, and superior query performance in long-term SLAM system.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 5","pages":"4388-4395"},"PeriodicalIF":4.6,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143706821","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":"Sparse Prototype Network for Explainable Pedestrian Behavior Prediction","authors":"Yan Feng;Alexander Carballo;Kazuya Takeda","doi":"10.1109/LRA.2025.3550728","DOIUrl":"https://doi.org/10.1109/LRA.2025.3550728","url":null,"abstract":"Predicting pedestrian behavior is challenging yet crucial for applications such as autonomous driving and smart cities. Recent deep learning models have achieved remarkable performance in making accurate predictions, but they fail to provide explanations of their inner workings. One reason for this problem is the multi-modal inputs. To bridge this gap, we present Sparse Prototype Network (SPN), an explainable method designed to simultaneously predict a pedestrian's future action, trajectory, and pose. SPN leverages an intermediate prototype bottleneck layer to provide sample-based explanations for its predictions. The prototypes are modality-independent, meaning that they can correspond to any modality from the input. Therefore, SPN can extend to arbitrary combinations of modalities. Regularized by mono-semanticity and clustering constraints, the prototypes learn consistent and human-understandable features and achieve state-of-the-art performance on action, trajectory and pose prediction on TITAN and PIE. Finally, we propose a metric named Top-K Mono-semanticity Scale to quantitatively evaluate the explainability. Qualitative results show a positive correlation between sparsity and explainability.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 5","pages":"4196-4203"},"PeriodicalIF":4.6,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143667416","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":"MAFF-Net: Enhancing 3D Object Detection With 4D Radar via Multi-Assist Feature Fusion","authors":"Xin Bi;Caien Weng;Panpan Tong;Baojie Fan;Arno Eichberge","doi":"10.1109/LRA.2025.3550707","DOIUrl":"https://doi.org/10.1109/LRA.2025.3550707","url":null,"abstract":"Perception systems are crucial for the safe operation of autonomous vehicles, particularly for 3D object detection. While LiDAR-based methods are limited by adverse weather conditions, 4D radars offer promising all-weather capabilities. However, 4D radars introduce challenges such as extreme sparsity, noise, and limited geometric information in point clouds. To address these issues, we propose MAFF-Net, a novel multi-assist feature fusion network specifically designed for 3D object detection using a single 4D radar. We introduce a sparsity pillar attention (SPA) module to mitigate the effects of sparsity while ensuring a sufficient receptive field. Additionally, we design the cluster query cross-attention (CQCA) module, which uses velocity-based clustered features as queries in the cross-attention fusion process. This helps the network enrich feature representations of potential objects while reducing measurement errors caused by angular resolution and multipath effects. Furthermore, we develop a cylindrical denoising assist (CDA) module to reduce noise interference, improving the accuracy of 3D bounding box predictions. Experiments on the VoD and TJ4DRadSet datasets demonstrate that MAFF-Net achieves state-of-the-art performance, outperforming 16-layer LiDAR systems and operating at over 17.9 FPS, making it suitable for real-time detection in autonomous vehicles.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 5","pages":"4284-4291"},"PeriodicalIF":4.6,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143688096","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":"Learning Decentralized Multi-Robot PointGoal Navigation","authors":"Takieddine Soualhi;Nathan Crombez;Yassine Ruichek;Alexandre Lombard;Stéphane Galland","doi":"10.1109/LRA.2025.3550798","DOIUrl":"https://doi.org/10.1109/LRA.2025.3550798","url":null,"abstract":"Integrating robots into real-world applications requires effective consideration of various agents, including other robots. Multi-agent reinforcement learning (MARL) is an established field that addresses multi-agent systems problems by leveraging reinforcement learning techniques. Despite its potential, the study of multi-robot systems, particularly in vision-based robotics, remains in its early stages. In this context, this article tackles the PointGoal navigation problem for multi-robot systems, extending the traditional single agent focus to a multi-agent context. To this end, we introduce a training environment designed to address vision-based multi-robot challenges. In addition, we propose a method based on the centralized training-decentralized execution paradigm within MARL to explore three PointGoal navigation scenarios: the SpecificGoal scenario, where each agent has a distinct target; the CommonGoal scenario, where all agents share the same target; and the Ad-hoCoop scenario, which requires agents to adapt to varying team sizes. Our results contribute to lay the groundwork for adopting MARL approaches to address vision-based tasks for multi-robot systems.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 4","pages":"4117-4124"},"PeriodicalIF":4.6,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143676024","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}