Learning to Drive in a Day
This introduces a new framework for autonomous driving that could reduce reliance on handcrafted rules and extensive supervision, though it is incremental as it focuses on a limited task.
The authors tackled autonomous driving by applying deep reinforcement learning to learn lane-following policies from scratch using only monocular images and a simple reward based on distance traveled without human intervention, achieving functional driving in a handful of training episodes.
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.