Making Sense of Vision and Touch: Learning Multimodal Representations for Contact-Rich Tasks
arXiv:1907.13098
The paper proposes a self‑supervised method to learn compact multimodal (vision and touch) representations that make deep reinforcement learning more sample‑efficient for contact‑rich manipulation tasks, demonstrated on peg insertion in simulation and on a real robot.
Abstract
Contact-rich manipulation tasks in unstructured environments often require both haptic and visual feedback. It is non-trivial to manually design a robot controller that combines these modalities which have very different characteristics. While deep reinforcement learning has shown success in learning control policies for high-dimensional inputs, these algorithms are generally intractable to deploy on real robots due to sample complexity. In this work, we use self-supervision to learn a compact and multimodal representation of our sensory inputs, which can then be used to improve the sample efficiency of our policy learning. Evaluating our method on a peg insertion task, we show that it generalizes over varying geometries, configurations, and clearances, while being robust to external perturbations. We also systematically study different self-supervised learning objectives and representation learning architectures. Results are presented in simulation and on a physical robot.
arXiv admin note: substantial text overlap with arXiv:1810.10191