Reinforcement Learning for Picking Cluttered General Objects with Dense Object Descriptors
This addresses robotic manipulation challenges in cluttered scenes, offering a novel approach for unsupervised learning, though it is incremental in combining descriptors with reinforcement learning.
The paper tackles the problem of picking cluttered general objects by proposing Cluttered Objects Descriptors (CODs) to represent object structures and training a picking policy with reinforcement learning, achieving a 96.69% success rate on unseen objects in a more cluttered environment.
Picking cluttered general objects is a challenging task due to the complex geometries and various stacking configurations. Many prior works utilize pose estimation for picking, but pose estimation is difficult on cluttered objects. In this paper, we propose Cluttered Objects Descriptors (CODs), a dense cluttered objects descriptor that can represent rich object structures, and use the pre-trained CODs network along with its intermediate outputs to train a picking policy. Additionally, we train the policy with reinforcement learning, which enable the policy to learn picking without supervision. We conduct experiments to demonstrate that our CODs is able to consistently represent seen and unseen cluttered objects, which allowed for the picking policy to robustly pick cluttered general objects. The resulting policy can pick 96.69% of unseen objects in our experimental environment which is twice as cluttered as the training scenarios.