2021 IEEE International Conference on Real-time Computing and Robotics (RCAR)最新文献_第6页

State Sensing of Spinal Surgical Robot Based on Fusion of Sound and Force Signals 基于声力融合的脊柱手术机器人状态感知

2021 IEEE International Conference on Real-time Computing and Robotics (RCAR) Pub Date : 2021-07-15 DOI: 10.1109/RCAR52367.2021.9517529

Meng Li, Xiaozhi Qi, Fengqing Guan, Haiyang Jin, Ying Hu, W. Tian

引用次数: 2

A Human-Computer Interaction Scheme of Lower-Limb Power-Assist Flexible Robot 一种下肢助力柔性机器人人机交互方案

2021 IEEE International Conference on Real-time Computing and Robotics (RCAR) Pub Date : 2021-07-15 DOI: 10.1109/RCAR52367.2021.9517694

Detian Zeng, Lei Sun, Xin Chen, Yunfei Li, Mu Zhu, Xinxiang Gong

引用次数: 1

Data-efficient Deep Reinforcement Learning Method Toward Scaling Continuous Robotic Task with Sparse Rewards 基于稀疏奖励扩展连续机器人任务的数据高效深度强化学习方法

2021 IEEE International Conference on Real-time Computing and Robotics (RCAR) Pub Date : 2021-07-15 DOI: 10.1109/RCAR52367.2021.9517647

Junkai Ren, Yichuan Zhang, Yujun Zeng, Yixing Lan

{"title":"Data-efficient Deep Reinforcement Learning Method Toward Scaling Continuous Robotic Task with Sparse Rewards","authors":"Junkai Ren, Yichuan Zhang, Yujun Zeng, Yixing Lan","doi":"10.1109/RCAR52367.2021.9517647","DOIUrl":"https://doi.org/10.1109/RCAR52367.2021.9517647","url":null,"abstract":"Dealing with the robotic continuous control problem with sparse rewards is a longstanding challenge in deep reinforcement learning (RL). While existing DRL algorithms have demonstrated great progress in learning policies from visual observations, learning effective policies still requires an impractical number of real-world data samples. Moreover, some robotic tasks are naturally specified with sparse rewards, which makes the precious data inefficient and slows down the learning process, making DRL infeasible. In addition, manually shaping reward functions is a complex work because it needs specific domain knowledge and human intervention. To alleviate the issue, this paper proposes a model-free, off-policy RL approach named TD3MHER, to learn the manipulating policy for continuous robotic tasks with sparse rewards. To be specific, TD3MHER utilizes Twin Delayed Deep Deterministic policy gradient algorithm (TD3) and Model-driven Hindsight Experience Replay (MHER) to achieve highly sample-efficient training property. Because while the agent is learning the policy, TD3MHER could also help it to learn the potation physical model of the robot which is helpful to solve the task, and it does not necessitate any novel robot-environment interactions. The performance of TD3MHER is assessed on a simulated robotic task using a 7-DOF manipulator to compare the proposed technique to a previous DRL algorithm and to verify the usefulness of our method. Results of the experiments on simulated robotic task show that the proposed approach is capable of successfully utilizing previously store samples with sparse rewards, and obtain a faster learning speed.","PeriodicalId":232892,"journal":{"name":"2021 IEEE International Conference on Real-time Computing and Robotics (RCAR)","volume":"58 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133784814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Dynamics modeling of a soft arm under the Cosserat theory 基于Cosserat理论的软臂动力学建模

2021 IEEE International Conference on Real-time Computing and Robotics (RCAR) Pub Date : 2021-07-15 DOI: 10.1109/RCAR52367.2021.9517660

Jie Ma, Zhiji Han, Linsen Yang, Gaochen Min, Zhijie Liu, W. He

引用次数: 1

Sliding-Window Informed RRT*: A Method for Speeding Up the Optimization and Path Smoothing 滑动窗口通知RRT*:一种加速优化和路径平滑的方法

2021 IEEE International Conference on Real-time Computing and Robotics (RCAR) Pub Date : 2021-07-15 DOI: 10.1109/RCAR52367.2021.9517672

Chenming Li, Chaoqun Wang, Jiankun Wang, Yantao Shen, M. Meng

引用次数: 5

Visualized Small-size Pipeline Model Building Using Multilink-articulated Wheeled In-pipe Inspection Robot 基于多连杆铰接轮式管内检测机器人的可视化小尺寸管道模型构建

2021 IEEE International Conference on Real-time Computing and Robotics (RCAR) Pub Date : 2021-07-15 DOI: 10.1109/RCAR52367.2021.9517622

Dianzhen Guo, Zhaohan Yuan, Sheng Bao, Jianjun Yuan, Shugen Ma, Liang Du

引用次数: 3

Research on Mechanical Properties of Bevel-Tip Needle Based on Image Guidance 基于图像引导的斜尖针力学性能研究

2021 IEEE International Conference on Real-time Computing and Robotics (RCAR) Pub Date : 2021-07-15 DOI: 10.1109/RCAR52367.2021.9517682

Linze Wang, Dedong Gao, Jiali Cui, Yan Zhao, Juntao Zhang

{"title":"Research on Mechanical Properties of Bevel-Tip Needle Based on Image Guidance","authors":"Linze Wang, Dedong Gao, Jiali Cui, Yan Zhao, Juntao Zhang","doi":"10.1109/RCAR52367.2021.9517682","DOIUrl":"https://doi.org/10.1109/RCAR52367.2021.9517682","url":null,"abstract":"The flexible needle is a minimally invasive medical device mainly used for human biopsy. The insertion operation has a small wound, which is conducive to the rehabilitation of patients and has a good application prospect. However, the force of the flexibility on the tissue is not certain, which seriously affects the doctor's judgment of the injury caused by the needle insert to the patient. In this paper, the mechanical properties of the flexible needle are studied by image analysis, the mechanical model between the needle holder and the force sensor is established, and the force of the needle holder is calculated by the force sensor data. Based on the cantilever beam model, the cutting force of the needle tip to the soft tissue and the friction force of the needle body are analyzed by using the force of the needle holder. The mechanical properties of flexible needles and the damage to tissues caused by needle insertion have been studied. Finally, the deflection of the needle body is calculated using the experimental force data, which are compared with the analysis by the image processing system. The results show that the cantilever beam model can predict the force of the needle body more accurately, and then predict the insertion trajectory.","PeriodicalId":232892,"journal":{"name":"2021 IEEE International Conference on Real-time Computing and Robotics (RCAR)","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115001334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Research on SLAM of Corridor Environment Based on Multi-Sensor 基于多传感器的廊道环境SLAM研究

2021 IEEE International Conference on Real-time Computing and Robotics (RCAR) Pub Date : 2021-07-15 DOI: 10.1109/RCAR52367.2021.9517540

Fei Wang, H. Shao, Q. Zhao, Zhiquan Feng

引用次数: 0

Autonomous mobile robot navigation in uncertain dynamic environments based on deep reinforcement learning 基于深度强化学习的不确定动态环境下自主移动机器人导航

2021 IEEE International Conference on Real-time Computing and Robotics (RCAR) Pub Date : 2021-07-15 DOI: 10.1109/RCAR52367.2021.9517635

Zhangfan Lu, Ran Huang

{"title":"Autonomous mobile robot navigation in uncertain dynamic environments based on deep reinforcement learning","authors":"Zhangfan Lu, Ran Huang","doi":"10.1109/RCAR52367.2021.9517635","DOIUrl":"https://doi.org/10.1109/RCAR52367.2021.9517635","url":null,"abstract":"In this paper, we study autonomous end-to-end navigation for wheeled robots based on deep reinforcement learning (DRL) in an unknown environment without a priori map. The DRL network is mainly based on deep deterministic policy gradient algorithm together with long short-term memory. The input for the network is the data from a 2D lidar as well as the relative position to the target point, while the outputs are the linear velocity and angular velocity that actuate the robot. A novel reward function is proposed to avoid the collision with dynamic obstacles and to generate a smooth trajectory for the robot. The network is trained without supervision in an unknown dynamic environment, the random Gaussian noise is added to the input data of long short-term memory to avoid local optimum. Besides, different unstructured environments are also considered in the training to increase the robustness of the developed network. Experiments performed on public dataset have showed that the developed network makes the robot navigate in unstructured environments safely and outperform several DRL methods.","PeriodicalId":232892,"journal":{"name":"2021 IEEE International Conference on Real-time Computing and Robotics (RCAR)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125363774","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

TactCapsNet: Tactile Capsule Network for Object Hardness Recognition 用于物体硬度识别的触觉胶囊网络

2021 IEEE International Conference on Real-time Computing and Robotics (RCAR) Pub Date : 2021-07-15 DOI: 10.1109/RCAR52367.2021.9517551

Senlin Fang, Tingting Mi, Zhenning Zhou, Chaoxiang Ye, Chengliang Liu, Hancheng Wu, Zhengkun Yi, Xinyu Wu

引用次数: 1