Abstract: In this work, three reinforcement learning algorithms (Proximal Policy Optimization, Soft Actor-Critic, and Twin Delayed Deep Deterministic Policy Gradient) are employed to control a two link selective compliance articulated ...