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

OpenObj: Open-Vocabulary Object-Level Neural Radiance Fields With Fine-Grained Understanding

IF 4.6 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2024-12-04 DOI: 10.1109/LRA.2024.3511401

Yinan Deng;Jiahui Wang;Jingyu Zhao;Jianyu Dou;Yi Yang;Yufeng Yue

{"title":"OpenObj: Open-Vocabulary Object-Level Neural Radiance Fields With Fine-Grained Understanding","authors":"Yinan Deng;Jiahui Wang;Jingyu Zhao;Jianyu Dou;Yi Yang;Yufeng Yue","doi":"10.1109/LRA.2024.3511401","DOIUrl":"https://doi.org/10.1109/LRA.2024.3511401","url":null,"abstract":"In recent years, there has been a surge of interest in open-vocabulary 3D scene reconstruction facilitated by visual language models (VLMs), which showcase remarkable capabilities in open-set retrieval tasks. Although the semantic ambiguity of existing point-wise feature maps is alleviated by open-vocabulary mask segmenters for object-level understanding, effectively retaining fine-grained features within objects simultaneously remains challenging. To address these challenges, we introduce OpenObj, an innovative approach to build open-vocabulary object-level Neural Radiance Fields (NeRF) with fine-grained understanding. In essence, OpenObj establishes a robust framework for efficient and watertight scene modeling and comprehension at the object level. Specifically, we obtain cross-frame consistent instance-level masks for supervision through our two-stage mask clustering module. Moreover, by incorporating part-level features into the object NeRF models, OpenObj not only captures object-level instances but also preserves an understanding of their internal granularity. The results on multiple datasets demonstrate that OpenObj achieves superior performance in zero-shot segmentation and retrieval tasks. Additionally, OpenObj supports real-world robotics tasks at several levels, including global movement and local manipulation.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 1","pages":"652-659"},"PeriodicalIF":4.6,"publicationDate":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142810587","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

Magnetorheological Fluid for Adaptive Soft Robotics

IF 4.6 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2024-12-04 DOI: 10.1109/LRA.2024.3511382

Lin Li;Yunquan Sun;Yi Gong;Qi Tong;Lizhe Qi

{"title":"Magnetorheological Fluid for Adaptive Soft Robotics","authors":"Lin Li;Yunquan Sun;Yi Gong;Qi Tong;Lizhe Qi","doi":"10.1109/LRA.2024.3511382","DOIUrl":"https://doi.org/10.1109/LRA.2024.3511382","url":null,"abstract":"Soft robotics, renowned for its robust adaptability to complex environments, holds great promise for a wide array of applications. Among the various actuation methods, magnetic fields stand out as particularly advantageous for soft robotics due to their ability to enable remote control and their biocompatibility. Current research in magnetic soft robots primarily centers on magnetic elastomers, which possess a certain load-bearing capacity and exhibit desired motions. However, these robots often lack the adaptability required to function effectively in diverse environments. Here we introduce an untethered Magnetorheological Fluid robot (MRF robot) that employs an innovative approach by encapsulating MRF within an elastic membrane. This design not only preserves the fluidity of the MRF, thereby ensuring environmental adaptability, but also provides the MRF robots with structural integrity when exposed to magnetic fields, thereby enhancing their load-bearing capabilities. The proposed MRF robot showcases its ability to navigate through narrow tunnels, manipulate and operate soft or complex objects. These results hold significant potential for inspiring versatile applications, such as the removal of foreign objects in biological systems via remote control, thereby potentially advancing medical procedures.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 1","pages":"620-627"},"PeriodicalIF":4.6,"publicationDate":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142810602","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

Hierarchical Deep Reinforcement Learning for Computation Offloading in Autonomous Multi-Robot Systems

IF 4.6 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2024-12-04 DOI: 10.1109/LRA.2024.3511408

Wen Gao;Zhiwen Yu;Liang Wang;Helei Cui;Bin Guo;Hui Xiong

{"title":"Hierarchical Deep Reinforcement Learning for Computation Offloading in Autonomous Multi-Robot Systems","authors":"Wen Gao;Zhiwen Yu;Liang Wang;Helei Cui;Bin Guo;Hui Xiong","doi":"10.1109/LRA.2024.3511408","DOIUrl":"https://doi.org/10.1109/LRA.2024.3511408","url":null,"abstract":"To ensure system responsiveness, some compute-intensive tasks are usually offloaded to cloud or edge computing devices. In environments where connection to external computing facilities is unavailable, computation offloading among members within an autonomous multi-robot system (AMRS) becomes a solution. The challenge lies in how to maximize the use of other members' idle resources without disrupting their local computation tasks. Therefore, this study proposes HRL-AMRS, a hierarchical deep reinforcement learning framework designed to distribute computational loads and reduce the processing time of computational tasks within an AMRS. In this framework, the high-level must consider the impact of data loading scales determined by low-level under varying computational device states on the actual processing times. In addition, the low-level employs Long Short-Term Memory (LSTM) networks to enhance the understanding of time-series states of computing devices. Experimental results show that, across various task sizes and numbers of robots, the framework reduces processing times by an average of 4.32% compared to baseline methods.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 1","pages":"540-547"},"PeriodicalIF":4.6,"publicationDate":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142821142","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

DotTip: Enhancing Dexterous Robotic Manipulation With a Tactile Fingertip Featuring Curved Perceptual Morphology DotTip：利用具有弧形感知形态的触觉指尖增强机器人的灵巧操作能力

IF 4.6 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2024-12-04 DOI: 10.1109/LRA.2024.3511431

Haoran Zheng;Xiaohang Shi;Ange Bao;Yongbin Jin;Pei Zhao

引用次数: 0

A LiDAR Odometry With Multi-Metric Feature Association and Contribution Constraint Selection

IF 4.6 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2024-12-04 DOI: 10.1109/LRA.2024.3511392

Nuo Li;Yiqing Yao;Xiaosu Xu;Zijian Wang

{"title":"A LiDAR Odometry With Multi-Metric Feature Association and Contribution Constraint Selection","authors":"Nuo Li;Yiqing Yao;Xiaosu Xu;Zijian Wang","doi":"10.1109/LRA.2024.3511392","DOIUrl":"https://doi.org/10.1109/LRA.2024.3511392","url":null,"abstract":"LiDAR-based simultaneous localization and mapping (SLAM) is crucial for achieving accurate pose estimation and map generation, thus serving as a foundational technology in the advancement of autonomous driving systems. In this letter, we introduce an accurate and robust feature-based LiDAR odometry method. Initially, we propose a feature extraction method centered on local extreme points, which capitalizes on the structural characteristics of local regions in LiDAR scans. Secondly, we purpose a multi-metric feature association approach for keyframe registration. This method leverages sparse and abstract geometric primitives to improve the accuracy and speed of keyframe matching. Additionally, Considering the varying impact of different metric features on pose constraints, an constraint contribution selection method is introduced to identify the most valuable features within the multi-metric feature set. Finally, the performance and efficiency of the proposed method are evaluated on the public KITTI, M2DGR, and The Newer College dataset, as well as our collected campus dataset. Experimental results demonstrate that the proposed method exhibits comparable performance compared to state-of-the-art LiDAR odometry methods across various scenarios.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 1","pages":"756-763"},"PeriodicalIF":4.6,"publicationDate":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142821144","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

DynaMoN: Motion-Aware Fast and Robust Camera Localization for Dynamic Neural Radiance Fields DynaMoN：面向动态神经辐射场的运动感知型快速稳健相机定位系统

IF 4.6 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2024-12-04 DOI: 10.1109/LRA.2024.3511399

Nicolas Schischka;Hannah Schieber;Mert Asim Karaoglu;Melih Gorgulu;Florian Grötzner;Alexander Ladikos;Nassir Navab;Daniel Roth;Benjamin Busam

{"title":"DynaMoN: Motion-Aware Fast and Robust Camera Localization for Dynamic Neural Radiance Fields","authors":"Nicolas Schischka;Hannah Schieber;Mert Asim Karaoglu;Melih Gorgulu;Florian Grötzner;Alexander Ladikos;Nassir Navab;Daniel Roth;Benjamin Busam","doi":"10.1109/LRA.2024.3511399","DOIUrl":"https://doi.org/10.1109/LRA.2024.3511399","url":null,"abstract":"The accurate reconstruction of dynamic scenes with neural radiance fields is significantly dependent on the estimation of camera poses. Widely used structure-from-motion pipelines encounter difficulties in accurately tracking the camera trajectory when faced with separate dynamics of the scene content and the camera movement. To address this challenge, we propose \u0000<underline>Dyna</u>\u0000mic \u0000<underline>Mo</u>\u0000tion-Aware Fast and Robust Camera Localization for Dynamic \u0000<underline>N</u>\u0000eural Radiance Fields (DynaMoN). DynaMoN utilizes semantic segmentation and generic motion masks to handle dynamic content for initial camera pose estimation and statics-focused ray sampling for fast and accurate novel-view synthesis. Our novel iterative learning scheme switches between training the NeRF and updating the pose parameters for an improved reconstruction and trajectory estimation quality. The proposed pipeline shows significant acceleration of the training process. We extensively evaluate our approach on two real-world dynamic datasets, the TUM RGB-D dataset and the BONN RGB-D Dynamic dataset. DynaMoN improves over the state-of-the-art both in terms of reconstruction quality and trajectory accuracy. We plan to make our code public to enhance research in this area.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 1","pages":"548-555"},"PeriodicalIF":4.6,"publicationDate":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10777295","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142821207","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

A Three-Finger Adaptive Gripper With Finger-Embedded Suction Cups for Enhanced Object Grasping Mechanism

IF 4.6 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2024-12-04 DOI: 10.1109/LRA.2024.3511428

Jimin Yoon;Heeyeon Jeong;Jae Hyeong Park;Young Jin Gong;Dongsu Shin;Hyeon-Woong Seo;Seung Jae Moon;Hyouk Ryeol Choi

引用次数: 0

Mechanism Characteristics Identification and Anti-Disturbance Control for Door-Opening Using Supernumerary Robotic Limbs

IF 4.6 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2024-12-04 DOI: 10.1109/LRA.2024.3511172

Kerui Sun;Tianjiao Zheng;Qinghua Zhang;Jian Qi;Hongwei Jing;Xianglong Li;Lele Li;Jie Zhao;Yanhe Zhu

{"title":"Mechanism Characteristics Identification and Anti-Disturbance Control for Door-Opening Using Supernumerary Robotic Limbs","authors":"Kerui Sun;Tianjiao Zheng;Qinghua Zhang;Jian Qi;Hongwei Jing;Xianglong Li;Lele Li;Jie Zhao;Yanhe Zhu","doi":"10.1109/LRA.2024.3511172","DOIUrl":"https://doi.org/10.1109/LRA.2024.3511172","url":null,"abstract":"Supernumerary robotic limbs (SRLs) are wearable robots that serve as extra limbs, assisting wearers in task completion. The strong coupling between SRLs and wearers results in wearer movements influencing SRLs' actions and the forces exerted on mechanisms during operations. The wearer's movements lead to uncertain and uncontrollable base movements, making operational control of the SRLs challenging. This study proposes a full-process assisted door-opening method using SRLs, integrating a task model and an anti-disturbance door-opening controller. Multimodal information from the SRLs system is utilized to facilitate task state transitions. The anti-disturbance door-opening controller consists of compliant controller, online identification of mechanism characteristics, end-effector trajectory generator, and notch filter. This method enables the SRLs to complete the door-opening task under motion disturbances caused by the wearer, without prior knowledge of the door's resistance, radius, or axis direction. Experimental results show that the SRLs can successfully open doors, with estimates converging to actual values under various radius and axis direction conditions (with radius estimation error less than 0.01 m and axial estimation error less than 3.8\u0000<inline-formula><tex-math>$^{circ }$</tex-math></inline-formula>\u0000). The effectiveness of the proposed control framework is validated in daily environments.","PeriodicalId":13241,"journal":{"name":"IEEE Robotics and Automation Letters","volume":"10 1","pages":"556-563"},"PeriodicalIF":4.6,"publicationDate":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142821143","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

From Instantaneous to Predictive Control: A More Intuitive and Tunable MPC Formulation for Robot Manipulators

IF 4.6 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2024-12-04 DOI: 10.1109/LRA.2024.3511439

Johan Ubbink;Ruan Viljoen;Erwin Aertbeliën;Wilm Decré;Joris De Schutter

引用次数: 0

Development and Evaluation of a Quasi-Passive Stiffness Display

IF 4.6 2区计算机科学

IEEE Robotics and Automation Letters Pub Date : 2024-12-04 DOI: 10.1109/LRA.2024.3511349

Ke Shi;Yingchen Gao;Yichen Xiang;Maozeng Zhang;Lifeng Zhu;Aiguo Song

引用次数: 0