A reinforcement learning approach for robot control in an unknown environment | IEEE Conference Publication | IEEE Xplore