Learning and Using Abstractions for Robot Planning
This work aims to reduce the computational burden of robot motion planning for roboticists and autonomous systems developers, offering a significant improvement in planning efficiency.
This paper addresses the computational challenge of robot motion planning, which is PSPACE-Hard for optimal solutions. The authors propose a deep-learning framework to identify critical robot configurations, using these to bias sampling distributions and significantly reduce planning times compared to current sampling-based methods.
Robot motion planning involves computing a sequence of valid robot configurations that take the robot from its initial state to a goal state. Solving a motion planning problem optimally using analytical methods is proven to be PSPACE-Hard. Sampling-based approaches have tried to approximate the optimal solution efficiently. Generally, sampling-based planners use uniform samplers to cover the entire state space. In this paper, we propose a deep-learning-based framework that identifies robot configurations in the environment that are important to solve the given motion planning problem. These states are used to bias the sampling distribution in order to reduce the planning time. Our approach works with a unified network and generates domain-dependent network parameters based on the environment and the robot. We evaluate our approach with Learn and Link planner in three different settings. Results show significant improvement in motion planning times when compared with current sampling-based motion planners.