Holistic Deep-Reinforcement-Learning-based Training of Autonomous Navigation Systems

TL;DR

This paper presents a holistic deep reinforcement learning approach for autonomous ground vehicle navigation, training all navigation stack entities together to improve synchronization, efficiency, and safety over existing methods.

Contribution

It introduces a comprehensive training method involving all navigation components, addressing issues like forgetfulness and non-optimal coordination in previous isolated approaches.

Findings

Outperforms baseline methods in efficiency and safety

Achieves shorter paths and fewer collisions

Improves synchronization among navigation entities

Abstract

In recent years, Deep Reinforcement Learning emerged as a promising approach for autonomous navigation of ground vehicles and has been utilized in various areas of navigation such as cruise control, lane changing, or obstacle avoidance. However, most research works either focus on providing an end-to-end solution training the whole system using Deep Reinforcement Learning or focus on one specific aspect such as local motion planning. This however, comes along with a number of problems such as catastrophic forgetfulness, inefficient navigation behavior, and non-optimal synchronization between different entities of the navigation stack. In this paper, we propose a holistic Deep Reinforcement Learning training approach in which the training procedure is involving all entities of the navigation stack. This should enhance the synchronization between- and understanding of all entities of the navigation stack and as a result, improve navigational performance. We trained several agents with a number of different observation spaces to study the impact of different input on the navigation behavior of the agent. In profound evaluations against multiple learning-based and classic model-based navigation approaches, our proposed agent could outperform the baselines in terms of efficiency and safety attaining shorter path lengths, less roundabout paths, and less collisions.