State Dropout-Based Curriculum Reinforcement Learning for Self-Driving at Unsignalized Intersections

TL;DR

This paper introduces a novel curriculum learning approach based on state dropout for deep reinforcement learning, improving training efficiency and performance in autonomous vehicle navigation at unsignalized intersections.

Contribution

It presents a unique curriculum design for training deep reinforcement learning agents and demonstrates its effectiveness in autonomous intersection traversal tasks.

Findings

Faster training process for RL agents.

Improved performance over non-curriculum training.

Validated on CommonRoad simulator with T- and four-way intersections.

Abstract

Traversing intersections is a challenging problem for autonomous vehicles, especially when the intersections do not have traffic control. Recently deep reinforcement learning has received massive attention due to its success in dealing with autonomous driving tasks. In this work, we address the problem of traversing unsignalized intersections using a novel curriculum for deep reinforcement learning. The proposed curriculum leads to: 1) A faster training process for the reinforcement learning agent, and 2) Better performance compared to an agent trained without curriculum. Our main contribution is two-fold: 1) Presenting a unique curriculum for training deep reinforcement learning agents, and 2) showing the application of the proposed curriculum for the unsignalized intersection traversal task. The framework expects processed observations of the surroundings from the perception system of the autonomous vehicle. We test our method in the CommonRoad motion planning simulator on T-intersections and four-way intersections.