作者: Silvio Savarese , Alexander Toshev , Roberto Martín-Martín , Or Litany , Fei Xia
DOI:
关键词: Torque 、 Robotics 、 Robot 、 Lift (data mining) 、 Artificial intelligence 、 Trajectory 、 Reinforcement learning 、 Motion generation 、 Computer science
摘要: Many Reinforcement Learning (RL) approaches use joint control signals (positions, velocities, torques) as action space for continuous control tasks. We propose to lift the action space …