Q-learning based object grasping control strategy for home service robot with rotatable waist

In this paper, a Q-learning based object grasping strategy and control method is proposed for the home service robot with a rotatable waist. The home service robot May used in this study possesses 6-DOF arms, 2-DOF neck, a rotatable waist, the four-wheel independent steering and four-wheel independent drive mobile platform. In order to increase the coverage of grasping, this paper proposes the Q-learning controller to find the most suitable angle of waist for grasping the object By the grasping strategy, the position of end-effector is calibrated using an ultrasonic ranging module. Moreover, in order to avoid overload of servo motors, the home service robot May is able to utilize the other arm to assist when the object is really heavy. Finally, the experimental results demonstrate the feasibility and practicality of object grasping and control strategy.