Sim-to-Real Reinforcement Learning for Deformable Object Manipulation

TL;DR

This paper demonstrates that deep reinforcement learning can be used to train policies in simulation for deformable object manipulation, such as cloth folding and draping, and successfully transfer them to real-world tasks without prior real deformable object data.

Contribution

It introduces a novel sim-to-real transfer approach for deformable object manipulation using deep reinforcement learning with domain randomization.

Findings

Policies trained in simulation successfully manipulated real deformable objects.

The approach handled multiple tasks including folding and draping.

Zero real-world training data was needed for deployment.

Abstract

We have seen much recent progress in rigid object manipulation, but interaction with deformable objects has notably lagged behind. Due to the large configuration space of deformable objects, solutions using traditional modelling approaches require significant engineering work. Perhaps then, bypassing the need for explicit modelling and instead learning the control in an end-to-end manner serves as a better approach? Despite the growing interest in the use of end-to-end robot learning approaches, only a small amount of work has focused on their applicability to deformable object manipulation. Moreover, due to the large amount of data needed to learn these end-to-end solutions, an emerging trend is to learn control policies in simulation and then transfer them over to the real world. To-date, no work has explored whether it is possible to learn and transfer deformable object policies. We believe that if sim-to-real methods are to be employed further, then it should be possible to learn to interact with a wide variety of objects, and not only rigid objects. In this work, we use a combination of state-of-the-art deep reinforcement learning algorithms to solve the problem of manipulating deformable objects (specifically cloth). We evaluate our approach on three tasks --- folding a towel up to a mark, folding a face towel diagonally, and draping a piece of cloth over a hanger. Our agents are fully trained in simulation with domain randomisation, and then successfully deployed in the real world without having seen any real deformable objects.