Deep R-Learning for Continual Area Sweeping

TL;DR

This paper introduces a reinforcement learning approach for continual area sweeping in robotics, enabling robots to adaptively learn to maximize event detection rates in unknown environments, surpassing previous greedy methods.

Contribution

It generalizes the non-uniform coverage problem to less constrained environments and applies RL in a Semi-Markov Decision Process framework for improved performance.

Findings

Significant performance improvements over greedy approaches.

Effective in both abstract and high-fidelity simulations.

Applicable to service robotics scenarios.

Abstract

Coverage path planning is a well-studied problem in robotics in which a robot must plan a path that passes through every point in a given area repeatedly, usually with a uniform frequency. To address the scenario in which some points need to be visited more frequently than others, this problem has been extended to non-uniform coverage planning. This paper considers the variant of non-uniform coverage in which the robot does not know the distribution of relevant events beforehand and must nevertheless learn to maximize the rate of detecting events of interest. This continual area sweeping problem has been previously formalized in a way that makes strong assumptions about the environment, and to date only a greedy approach has been proposed. We generalize the continual area sweeping formulation to include fewer environmental constraints, and propose a novel approach based on reinforcement learning in a Semi-Markov Decision Process. This approach is evaluated in an abstract simulation and in a high fidelity Gazebo simulation. These evaluations show significant improvement upon the existing approach in general settings, which is especially relevant in the growing area of service robotics.