i. Briefly explain the concept of Model Predictive Control, in particular with details on the cost function. (5 marks) ii. Briefly explain the concept of Q-Learning, in particular relates rewards and a cost function. Using the maze from Figure Q5 as a supporting example, provide the maximum Q-value associate with at least half the states. Use the following rewards: 100 points for the goal (green flag), 1 for each movement, 50 on the flames. The starting point is the robot. (5 marks) Total Q5: 10 marks