DexPoint: Generalizable Point Cloud Reinforcement Learning for Sim-to-Real Dexterous Manipulation
It addresses the problem of generalizable robotic manipulation for researchers and practitioners, presenting a novel framework with incremental techniques.
The paper tackles sim-to-real dexterous manipulation by training a policy with point cloud inputs and dexterous hands, achieving generalization to new objects in the same category in the real world, as demonstrated with an Allegro Hand.
We propose a sim-to-real framework for dexterous manipulation which can generalize to new objects of the same category in the real world. The key of our framework is to train the manipulation policy with point cloud inputs and dexterous hands. We propose two new techniques to enable joint learning on multiple objects and sim-to-real generalization: (i) using imagined hand point clouds as augmented inputs; and (ii) designing novel contact-based rewards. We empirically evaluate our method using an Allegro Hand to grasp novel objects in both simulation and real world. To the best of our knowledge, this is the first policy learning-based framework that achieves such generalization results with dexterous hands. Our project page is available at https://yzqin.github.io/dexpoint