Model-free Deep Reinforcement Learning for Urban Autonomous Driving

TL;DR

This paper introduces a model-free deep reinforcement learning framework tailored for complex urban autonomous driving, demonstrating its effectiveness in dense roundabout scenarios through simulation.

Contribution

It presents a novel framework with specific input encoding and performance tricks, enabling deep RL to handle urban driving complexities.

Findings

Successfully solves dense roundabout task in simulation

Outperforms baseline methods significantly

Demonstrates potential for scalable urban autonomous driving

Abstract

Urban autonomous driving decision making is challenging due to complex road geometry and multi-agent interactions. Current decision making methods are mostly manually designing the driving policy, which might result in sub-optimal solutions and is expensive to develop, generalize and maintain at scale. On the other hand, with reinforcement learning (RL), a policy can be learned and improved automatically without any manual designs. However, current RL methods generally do not work well on complex urban scenarios. In this paper, we propose a framework to enable model-free deep reinforcement learning in challenging urban autonomous driving scenarios. We design a specific input representation and use visual encoding to capture the low-dimensional latent states. Several state-of-the-art model-free deep RL algorithms are implemented into our framework, with several tricks to improve their performance. We evaluate our method in a challenging roundabout task with dense surrounding vehicles in a high-definition driving simulator. The result shows that our method can solve the task well and is significantly better than the baseline.