Deep Reinforcement Learning for Robotic Controls

Presenter: Dario Mangoni on behalf of Alessandro Tasora, Engineering Professor and Digital Dynamics Lab Leader, University of Parma

This presentation address the use of the Proximal Policy Optimization (PPO) deep reinforcement learning algorithm to train a Neural Network to control a robotic walker and a robotic arm in simulation. The Neural Network is trained to control the torque setpoints of motors in order to achieve an optimal goal.