Residual Reactive Navigation: Combining Classical and Learned Navigation Strategies For Deployment in Unknown Environments

TL;DR

This paper introduces Residual Reactive Navigation, a hybrid approach combining learned and classical control strategies to improve mobile robot navigation in unknown, cluttered environments, enhancing efficiency and generalization.

Contribution

It adapts residual reinforcement learning to navigation, proposing a switching strategy that combines learned policies with classical controllers for better real-world performance.

Findings

Improved navigation efficiency over end-to-end methods

Effective handling of high policy uncertainty scenarios

Successful deployment in real-world indoor environments

Abstract

In this work we focus on improving the efficiency and generalisation of learned navigation strategies when transferred from its training environment to previously unseen ones. We present an extension of the residual reinforcement learning framework from the robotic manipulation literature and adapt it to the vast and unstructured environments that mobile robots can operate in. The concept is based on learning a residual control effect to add to a typical sub-optimal classical controller in order to close the performance gap, whilst guiding the exploration process during training for improved data efficiency. We exploit this tight coupling and propose a novel deployment strategy, switching Residual Reactive Navigation (sRRN), which yields efficient trajectories whilst probabilistically switching to a classical controller in cases of high policy uncertainty. Our approach achieves improved performance over end-to-end alternatives and can be incorporated as part of a complete navigation stack for cluttered indoor navigation tasks in the real world. The code and training environment for this project is made publicly available at https://sites.google.com/view/srrn/home.