We demonstrate the first application of deep reinforcement learning to autonomous driving. From randomly initialised parameters, our model is able to learn a policy for lane following in a handful of training episodes using a single monocular image as input. We provide a general and easy to obtain reward: the distance travelled by the vehicle without the safety driver taking control. We use a continuous, model-free deep reinforcement learning algorithm, with all exploration and optimisation performed on-vehicle. This demonstrates a new framework for autonomous driving which moves away from reliance on defined logical rules, mapping, and direct supervision. We discuss the challenges and opportunities to scale this approach to a broader range of autonomous driving tasks.
@article{arxiv.1807.00412,
title = {Learning to Drive in a Day},
author = {Alex Kendall and Jeffrey Hawke and David Janz and Przemyslaw Mazur and Daniele Reda and John-Mark Allen and Vinh-Dieu Lam and Alex Bewley and Amar Shah},
journal= {arXiv preprint arXiv:1807.00412},
year = {2018}
}
Comments
Further results and demo videos can be viewed at: https://wayve.ai/blog/l2diad