Tactical Decision Making for Lane Changing with Deep Reinforcement Learning

Mustafa Mukadam, Akansel Cosgun, Alireza Nakhaei, Kikuo Fujimura

Abstract: In this paper, we consider the problem of autonomous lane changing for self driving vehicles in a multi-lane, multi-agent setting. We present a framework that demonstrates a more structured and data efficient alternative to end-to-end complete policy learning on problems where the high-level policy is hard to formulate using traditional optimization or rule based methods but well designed low-level controllers are available. Our framework uses deep reinforcement learning solely to obtain a high-level policy for tactical decision making, while still maintaining a tight integration with the low-level controller, thus getting the best of both worlds. We accomplish this with Q-masking, a technique with which we are able to incorporate prior knowledge, constraints, and information from a low-level controller, directly in to the learning process thereby simplifying the reward function and making learning faster and data efficient. We provide preliminary results in a simulator and show our approach to be more efficient than a greedy baseline, and more successful and safer than human driving.

TL;DR: A framework that provides a policy for autonomous lane changing by learning to make high-level tactical decisions with deep reinforcement learning, and maintaining a tight integration with a low-level controller to take low-level actions.

Keywords: autonomous lane changing, decision making, deep reinforcement learning, q-learning