Zero-shot Reinforcement Learning on ExORL APS Walker environment v1 (test)