Using Reinforcement Learning to Improve the Stability of a Humanoid Robot: Walking on Sloped Terrain | IEEE Conference Publication | IEEE Xplore