Quadruped robot obstacle negotiation via reinforcement learning | IEEE Conference Publication | IEEE Xplore