Apply Reinforcement Learning (RL) to enable prosthetics to calibrate with differences between humans and differences between walking environments