Reinforcement Learning for Vision-Based Lateral Control of a Self-Driving Car | IEEE Conference Publication | IEEE Xplore