Improved Learning of Robot Manipulation Tasks via Tactile Intrinsic Motivation
This addresses the problem of sparse reward exploration in robotic manipulation for researchers and practitioners, representing an incremental improvement with specific algorithmic enhancements.
The paper tackles the exploration challenge in deep reinforcement learning for robotic manipulation tasks by introducing a tactile intrinsic motivation reward based on force sensor readings and contact-prioritized experience replay. The result shows accelerated exploration and outperformance of state-of-the-art methods on three fundamental robot manipulation benchmarks.
In this paper we address the challenge of exploration in deep reinforcement learning for robotic manipulation tasks. In sparse goal settings, an agent does not receive any positive feedback until randomly achieving the goal, which becomes infeasible for longer control sequences. Inspired by touch-based exploration observed in children, we formulate an intrinsic reward based on the sum of forces between a robot's force sensors and manipulation objects that encourages physical interaction. Furthermore, we introduce contact-prioritized experience replay, a sampling scheme that prioritizes contact rich episodes and transitions. We show that our solution accelerates the exploration and outperforms state-of-the-art methods on three fundamental robot manipulation benchmarks.