Real Robot Challenge using Deep Reinforcement Learning

09/30/2021
by   Robert McCarthy, et al.
12

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. Simultaneously, a dense distance-based reward is employed to teach the policy to lift the cube to the desired z coordinate. The policy is trained in simulation with domain randomization before being transferred to the real robot for evaluation. Although performance tends to worsen after this transfer, our best trained policy can successfully lift the real cube along goal trajectories via the use of an effective pinching grasp. Our approach outperforms all other submissions, including those leveraging more traditional robotic control techniques, and is the first learning-based approach to solve this challenge.

READ FULL TEXT
research
05/19/2022

Dexterous Robotic Manipulation using Deep Reinforcement Learning and Knowledge Transfer for Complex Sparse Reward-based Tasks

This paper describes a deep reinforcement learning (DRL) approach that w...
research
07/09/2020

Robotic Grasping using Deep Reinforcement Learning

In this work, we present a deep reinforcement learning based method to s...
research
09/16/2018

Improvements on Hindsight Learning

Sparse reward problems are one of the biggest challenges in Reinforcemen...
research
08/17/2021

Diversity-based Trajectory and Goal Selection with Hindsight Experience Replay

Hindsight experience replay (HER) is a goal relabelling technique typica...
research
01/11/2020

Reward Engineering for Object Pick and Place Training

Robotic grasping is a crucial area of research as it can result in the a...
research
07/09/2022

Optimizing Bipedal Maneuvers of Single Rigid-Body Models for Reinforcement Learning

In this work, we propose a method to generate reduced-order model refere...
research
11/27/2021

Reinforcement Learning-based Switching Controller for a Milliscale Robot in a Constrained Environment

This work presents a reinforcement learning-based switching control mech...

Please sign up or login with your details

Forgot password? Click here to reset