A Survey of Deep Reinforcement Learning Algorithms for Motion Planning and Control of Autonomous Vehicles

TL;DR

This survey reviews deep reinforcement learning methods for autonomous vehicle motion planning and control, comparing pipeline and end-to-end approaches, highlighting their advantages, limitations, and future research directions.

Contribution

It systematically summarizes current RL-based methods for autonomous driving, contrasting traditional pipeline and end-to-end approaches, and discusses challenges and future directions.

Findings

End-to-end RL approaches often outperform pipeline methods in performance.

Pipeline approaches offer better interpretability but may lack optimal system-level performance.

Challenges include data scarcity, generalization, and real-world deployment issues.

Abstract

In this survey, we systematically summarize the current literature on studies that apply reinforcement learning (RL) to the motion planning and control of autonomous vehicles. Many existing contributions can be attributed to the pipeline approach, which consists of many hand-crafted modules, each with a functionality selected for the ease of human interpretation. However, this approach does not automatically guarantee maximal performance due to the lack of a system-level optimization. Therefore, this paper also presents a growing trend of work that falls into the end-to-end approach, which typically offers better performance and smaller system scales. However, their performance also suffers from the lack of expert data and generalization issues. Finally, the remaining challenges applying deep RL algorithms on autonomous driving are summarized, and future research directions are also presented to tackle these challenges.