DexPBT: Scaling up Dexterous Manipulation for Hand-Arm Systems with Population Based Training
This addresses the challenge of scaling up dexterous manipulation for robotics, though it appears incremental as it builds on existing PBT methods.
The paper tackled the problem of learning dexterous object manipulation for hand-arm robots using a decentralized Population-Based Training algorithm, which significantly outperformed regular end-to-end learning and discovered robust control policies in tasks like regrasping and grasp-and-throw.
In this work, we propose algorithms and methods that enable learning dexterous object manipulation using simulated one- or two-armed robots equipped with multi-fingered hand end-effectors. Using a parallel GPU-accelerated physics simulator (Isaac Gym), we implement challenging tasks for these robots, including regrasping, grasp-and-throw, and object reorientation. To solve these problems we introduce a decentralized Population-Based Training (PBT) algorithm that allows us to massively amplify the exploration capabilities of deep reinforcement learning. We find that this method significantly outperforms regular end-to-end learning and is able to discover robust control policies in challenging tasks. Video demonstrations of learned behaviors and the code can be found at https://sites.google.com/view/dexpbt