Knowledge Commons of Institute of Automation,CAS
Trajectory-based Split Hindsight Reverse Curriculum Learning | |
Wu, Jiaxi1,2![]() ![]() ![]() | |
2021-09 | |
会议名称 | 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems |
会议日期 | 2021-9 |
会议地点 | Prague, Czech Republic |
摘要 | Grasping is one of the most fundamental problems in robotic manipulation. In recent years, with the development of data-driven methods, reinforcement learning has been used in solving robotic grasping problems. However, grasping is a long-horizon and sparse reward task, whose natural reward only appears when the task is successfully achieved. Therefore, it brings great challenges to the deployment of reinforcement learning methods. To tackle this difficulty, we propose a new method called Trajectory-based Split Hindsight Reverse Curriculum Learning. This method of reverse learning from the goal can greatly improve the learning efficiency and the final performance of the tasks. Specifically, based on referred trajectories, the agent starts to learn in a small state space near the goal and then gradually in larger state spaces until covering the entire state space. Through split hindsight experience replay, the sampled trajectory is divided into segments that match the current subspace’s size; then, they are modified to successful trajectories to enable more efficient learning. In both simulation and real-world experiments, our method surpasses the existing methods and achieves the goal-oriented grasping tasks with higher success rates and better data efficiencies. The detailed experimental results can be viewed at https://youtu.be/7uNRzmRZhDk. |
关键词 | Reinforcement Learning Curriculum Learning |
收录类别 | EI |
语种 | 英语 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/48522 |
专题 | 多模态人工智能系统全国重点实验室_机器人理论与应用 |
通讯作者 | Qiao, Hong |
作者单位 | 1.the State Key Laboratory for Management and Control of Complex Systems, Institute of Automation, Chinese Academy of Sciences 2.School of Artificial Intelligence, University of Chinese Academy of Sciences |
第一作者单位 | 中国科学院自动化研究所 |
通讯作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Wu, Jiaxi,Zhang, Dianmin,Zhong, Shanlin,et al. Trajectory-based Split Hindsight Reverse Curriculum Learning[C],2021. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Paper1(IROS2021).pdf(5094KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论