Learning to overtake in TORCS using simple reinforcement learning | IEEE Conference Publication | IEEE Xplore