Developmental Reinforcement Learning of Control Policy of a Quadcopter UAV with Thrust Vectoring Rotors