The use of modeling and planning as a tool to generate car and robot behavior as well as to model human behavior is discussed as part of the reinforcement learning process, and how reinforcement learning can be applied in either the real world or in simulation.