Model-free Deep Reinforcement Learning for Urban Autonomous Driving
This addresses the problem of sub-optimal and expensive manual policy design for urban autonomous driving, though it appears incremental as it adapts existing RL algorithms with tricks.
The paper tackles urban autonomous driving decision-making by proposing a framework for model-free deep reinforcement learning that uses visual encoding and specific input representations. The method successfully solves a challenging roundabout task with dense traffic in a simulator, showing significant improvement over baseline approaches.
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.