Learning to Manipulate Deformable Objects without Demonstrations
This addresses the problem of sample inefficiency in RL for robotic manipulation of deformable objects like cloth and rope, representing a domain-specific incremental improvement.
The paper tackles deformable object manipulation by developing a model-free visual reinforcement learning approach that accelerates learning through an iterative pick-place action space and a Maximal Value under Placing (MVP) method, achieving an order of magnitude faster learning and improved coverage on real robot tasks.
In this paper we tackle the problem of deformable object manipulation through model-free visual reinforcement learning (RL). In order to circumvent the sample inefficiency of RL, we propose two key ideas that accelerate learning. First, we propose an iterative pick-place action space that encodes the conditional relationship between picking and placing on deformable objects. The explicit structural encoding enables faster learning under complex object dynamics. Second, instead of jointly learning both the pick and the place locations, we only explicitly learn the placing policy conditioned on random pick points. Then, by selecting the pick point that has Maximal Value under Placing (MVP), we obtain our picking policy. This provides us with an informed picking policy during testing, while using only random pick points during training. Experimentally, this learning framework obtains an order of magnitude faster learning compared to independent action-spaces on our suite of deformable object manipulation tasks with visual RGB observations. Finally, using domain randomization, we transfer our policies to a real PR2 robot for challenging cloth and rope coverage tasks, and demonstrate significant improvements over standard RL techniques on average coverage.