Solving the Real Robot Challenge using Deep Reinforcement Learning
This solves a robotic manipulation problem for researchers and practitioners by demonstrating a learning-based method that can handle real-world tasks without extensive domain expertise, though it is incremental as it builds on existing RL techniques.
The authors tackled the Real Robot Challenge Phase 1, where a three-fingered robot must move a cube along goal trajectories, using a pure deep reinforcement learning approach with minimal expert knowledge, and their method outperformed all other submissions, including traditional control techniques, achieving successful cube lifting via pinching grasp.
This paper details our winning submission to Phase 1 of the 2021 Real Robot Challenge; a challenge in which a three-fingered robot must carry a cube along specified goal trajectories. To solve Phase 1, we use a pure reinforcement learning approach which requires minimal expert knowledge of the robotic system, or of robotic grasping in general. A sparse, goal-based reward is employed in conjunction with Hindsight Experience Replay to teach the control policy to move the cube to the desired x and y coordinates of the goal. Simultaneously, a dense distance-based reward is employed to teach the policy to lift the cube to the z coordinate (the height component) of the goal. The policy is trained in simulation with domain randomisation before being transferred to the real robot for evaluation. Although performance tends to worsen after this transfer, our best policy can successfully lift the real cube along goal trajectories via an effective pinching grasp. Our approach outperforms all other submissions, including those leveraging more traditional robotic control techniques, and is the first pure learning-based method to solve this challenge.