Improved Reinforcement Learning Coordinated Control of a Mobile Manipulator using Joint Clamping

Many robotic path planning problems are continuous, stochastic, and\nhigh-dimensional. The ability of a mobile manipulator to coordinate its base\nand manipulator in order to control its whole-body online is particularly\nchallenging when self and environment collision avoidance is required.\nReinforcement Learning techniques have the potential to solve such problems\nthrough their ability to generalise over environments. We study joint penalties\nand joint limits of a state-of-the-art mobile manipulator whole-body controller\nthat uses LIDAR sensing for obstacle collision avoidance. We propose directions\nto improve the reinforcement learning method. Our agent achieves significantly\nhigher success rates than the baseline in a goal-reaching environment and it\ncan solve environments that require coordinated whole-body control which the\nbaseline fails.\n