Yuzhu Sun, Mien Van, Stephen McIlvanna, Nguyen Minh Nhat, Kabirat Olayemi, Jack Close, Seán McLoone
{"title":"机器人机械手避障的数字双胞胎驱动强化学习:自我完善的在线培训框架","authors":"Yuzhu Sun, Mien Van, Stephen McIlvanna, Nguyen Minh Nhat, Kabirat Olayemi, Jack Close, Seán McLoone","doi":"arxiv-2403.13090","DOIUrl":null,"url":null,"abstract":"The evolution and growing automation of collaborative robots introduce more\ncomplexity and unpredictability to systems, highlighting the crucial need for\nrobot's adaptability and flexibility to address the increasing complexities of\ntheir environment. In typical industrial production scenarios, robots are often\nrequired to be re-programmed when facing a more demanding task or even a few\nchanges in workspace conditions. To increase productivity, efficiency and\nreduce human effort in the design process, this paper explores the potential of\nusing digital twin combined with Reinforcement Learning (RL) to enable robots\nto generate self-improving collision-free trajectories in real time. The\ndigital twin, acting as a virtual counterpart of the physical system, serves as\na 'forward run' for monitoring, controlling, and optimizing the physical system\nin a safe and cost-effective manner. The physical system sends data to\nsynchronize the digital system through the video feeds from cameras, which\nallows the virtual robot to update its observation and policy based on real\nscenarios. The bidirectional communication between digital and physical systems\nprovides a promising platform for hardware-in-the-loop RL training through\ntrial and error until the robot successfully adapts to its new environment. The\nproposed online training framework is demonstrated on the Unfactory Xarm5\ncollaborative robot, where the robot end-effector aims to reach the target\nposition while avoiding obstacles. The experiment suggest that proposed\nframework is capable of performing policy online training, and that there\nremains significant room for improvement.","PeriodicalId":501062,"journal":{"name":"arXiv - CS - Systems and Control","volume":"26 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Digital Twin-Driven Reinforcement Learning for Obstacle Avoidance in Robot Manipulators: A Self-Improving Online Training Framework\",\"authors\":\"Yuzhu Sun, Mien Van, Stephen McIlvanna, Nguyen Minh Nhat, Kabirat Olayemi, Jack Close, Seán McLoone\",\"doi\":\"arxiv-2403.13090\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The evolution and growing automation of collaborative robots introduce more\\ncomplexity and unpredictability to systems, highlighting the crucial need for\\nrobot's adaptability and flexibility to address the increasing complexities of\\ntheir environment. In typical industrial production scenarios, robots are often\\nrequired to be re-programmed when facing a more demanding task or even a few\\nchanges in workspace conditions. To increase productivity, efficiency and\\nreduce human effort in the design process, this paper explores the potential of\\nusing digital twin combined with Reinforcement Learning (RL) to enable robots\\nto generate self-improving collision-free trajectories in real time. The\\ndigital twin, acting as a virtual counterpart of the physical system, serves as\\na 'forward run' for monitoring, controlling, and optimizing the physical system\\nin a safe and cost-effective manner. The physical system sends data to\\nsynchronize the digital system through the video feeds from cameras, which\\nallows the virtual robot to update its observation and policy based on real\\nscenarios. The bidirectional communication between digital and physical systems\\nprovides a promising platform for hardware-in-the-loop RL training through\\ntrial and error until the robot successfully adapts to its new environment. The\\nproposed online training framework is demonstrated on the Unfactory Xarm5\\ncollaborative robot, where the robot end-effector aims to reach the target\\nposition while avoiding obstacles. The experiment suggest that proposed\\nframework is capable of performing policy online training, and that there\\nremains significant room for improvement.\",\"PeriodicalId\":501062,\"journal\":{\"name\":\"arXiv - CS - Systems and Control\",\"volume\":\"26 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-03-19\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Systems and Control\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2403.13090\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Systems and Control","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2403.13090","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Digital Twin-Driven Reinforcement Learning for Obstacle Avoidance in Robot Manipulators: A Self-Improving Online Training Framework
The evolution and growing automation of collaborative robots introduce more
complexity and unpredictability to systems, highlighting the crucial need for
robot's adaptability and flexibility to address the increasing complexities of
their environment. In typical industrial production scenarios, robots are often
required to be re-programmed when facing a more demanding task or even a few
changes in workspace conditions. To increase productivity, efficiency and
reduce human effort in the design process, this paper explores the potential of
using digital twin combined with Reinforcement Learning (RL) to enable robots
to generate self-improving collision-free trajectories in real time. The
digital twin, acting as a virtual counterpart of the physical system, serves as
a 'forward run' for monitoring, controlling, and optimizing the physical system
in a safe and cost-effective manner. The physical system sends data to
synchronize the digital system through the video feeds from cameras, which
allows the virtual robot to update its observation and policy based on real
scenarios. The bidirectional communication between digital and physical systems
provides a promising platform for hardware-in-the-loop RL training through
trial and error until the robot successfully adapts to its new environment. The
proposed online training framework is demonstrated on the Unfactory Xarm5
collaborative robot, where the robot end-effector aims to reach the target
position while avoiding obstacles. The experiment suggest that proposed
framework is capable of performing policy online training, and that there
remains significant room for improvement.