Let’s talk a bit first about what we have to work with. We have a mobile base with a six degree-of-freedom arm attached to it. We are fortunate that TinMan's robot arm is constructed out of servo motors. We can tell where all of the parts of the arm are if we know where the servos are commanded to be, since we are commanding the angles of the servos. We can use forward kinematics, which is to sum up all the angles and levers of the arm in order to deduce where the hand is located at. We can use this hand location as our desired state – our reward criteria. We will give the robot points, or rewards, based on how close the hand is to the desired position and orientation we want. We want the robot to figure out what it takes to get to that position. We need to give the robot a way to test out different theories, or actions that will result in the...