Urban Driving with Multi-Objective Deep Reinforcement Learning
This addresses the problem of efficient and safe autonomous driving in urban environments for researchers and developers, though it is incremental as it builds on existing DQN methods.
The paper tackles urban autonomous driving by developing a multi-objective deep reinforcement learning agent that learns to drive on multi-lane roads and intersections while optimizing for speed, safety, and comfort, with results showing significant data efficiency improvements and zero-shot transfer to a ring road without performance loss.
Autonomous driving is a challenging domain that entails multiple aspects: a vehicle should be able to drive to its destination as fast as possible while avoiding collision, obeying traffic rules and ensuring the comfort of passengers. In this paper, we present a deep learning variant of thresholded lexicographic Q-learning for the task of urban driving. Our multi-objective DQN agent learns to drive on multi-lane roads and intersections, yielding and changing lanes according to traffic rules. We also propose an extension for factored Markov Decision Processes to the DQN architecture that provides auxiliary features for the Q function. This is shown to significantly improve data efficiency. We then show that the learned policy is able to zero-shot transfer to a ring road without sacrificing performance.