End-to-End Intersection Handling using Multi-Agent Deep Reinforcement Learning

TL;DR

This paper presents a multi-agent deep reinforcement learning system enabling autonomous vehicles to navigate intersections using traffic signs, outperforming rule-based methods especially in dense traffic and generalizing to real-world scenarios.

Contribution

Introduces a multi-agent deep reinforcement learning approach for intersection navigation using traffic signs, demonstrating improved safety and generalization over rule-based methods.

Findings

Outperforms rule-based methods in dense traffic conditions

Learns intersection priorities and safe driving behaviors

Generalizes well to unseen environments and real traffic data

Abstract

Navigating through intersections is one of the main challenging tasks for an autonomous vehicle. However, for the majority of intersections regulated by traffic lights, the problem could be solved by a simple rule-based method in which the autonomous vehicle behavior is closely related to the traffic light states. In this work, we focus on the implementation of a system able to navigate through intersections where only traffic signs are provided. We propose a multi-agent system using a continuous, model-free Deep Reinforcement Learning algorithm used to train a neural network for predicting both the acceleration and the steering angle at each time step. We demonstrate that agents learn both the basic rules needed to handle intersections by understanding the priorities of other learners inside the environment, and to drive safely along their paths. Moreover, a comparison between our system and a rule-based method proves that our model achieves better results especially with dense traffic conditions. Finally, we test our system on real world scenarios using real recorded traffic data, proving that our module is able to generalize both to unseen environments and to different traffic conditions.