Joint Space Control via Deep Reinforcement Learning
Visak Kumar,David Hoeller,Balakumar Sundaralingam,Jonathan Tremblay,Stan Birchfield,Visak Kumar,David Hoeller,Balakumar Sundaralingam,Jonathan Tremblay,Stan Birchfield
The dominant way to control a robot manipulator uses hand-crafted differential equations leveraging some form of inverse kinematics / dynamics. We propose a simple, versatile joint-level controller that dispenses with differential equations entirely. A deep neural network, trained via model-free reinforcement learning, is used to map from task space to joint space. Experiments show the method capa...
