Learning Wall Following Behaviour in Robotics through Reinforcement and Image-based States | IEEE Conference Publication | IEEE Xplore