首页|Research Data from University of Montreal Update Understanding of Robotics (Torq ue-based Deep Reinforcement Learning for Taskand-robot Agnostic Learning On Bip edal Robots Using Sim-to-real Transfer)
Research Data from University of Montreal Update Understanding of Robotics (Torq ue-based Deep Reinforcement Learning for Taskand-robot Agnostic Learning On Bip edal Robots Using Sim-to-real Transfer)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Fresh data on Robotics are presented i n a new report.According to news reporting originating from Quebec City,Canada,by NewsRx correspondents,research stated,"In this letter,we review the ques tion of which action space is best suited for controlling a real biped robot in combination with Sim2Real training.Position control has been popular as it has been shown to be more sample efficient and intuitive to combine with other plann ing algorithms." Financial support for this research came from Ministry of Science & ICT (MSIT),Republic of Korea.Our news editors obtained a quote from the research from the University of Montr eal,"However,for position control,gain tuning is required to achieve the best possible policy performance.We show that,instead,using a torque-based action space enables task-and-robot agnostic learning with less parameter tuning and m itigates the sim-to-reality gap by taking advantage of torque control's inherent compliance.Also,we accelerate the torque-based-policy training process by pre -training the policy to remain upright by compensating for gravity."
Quebec CityCanadaNorth and Central A mericaEmerging TechnologiesMachine LearningNano-robotReinforcement Learn ingRobotRoboticsUniversity of Montreal
NETL
NSTL
