Host: The Japan Society of Mechanical Engineers
Name : [in Japanese]
Date : June 08, 2016 - June 11, 2016
In this paper, we study ball juggling as a dexterous and accurate high-speed manipulation like a human. We have already achieved two-ball juggling with our high-speed robotic handarm system which has three general-purpose fingers and high-speed vision system. However, the number of times of ball-catching is not so many because the trajectory of the motion of the robot is not optimized. In this paper, we propose a trajectory generation method using smoothing splines. The motion of throwing a ball is optimized so as to minimize the torques of the joint angles. The motion of catching is controlled by visual feedback. As a result, stable throwing motion is achieved so that the success rate of ball-catching is about 90%.
