Ping-pong trajectory perception and prediction by a PC based High speed four-camera vision system

A high speed vision system is vital for a robot player to play table tennis successfully. The three dimensional ping-pong trajectory should be perceived in high precision at a high frame rate. What's more, with the beginning parts of the trajectory, the following trajectory should be predicted to save enough time for successful planning and control. In this paper, a high speed vision system is proposed. The hardware of the vision system is composed of four high speed 1394 firewire cameras and a powerful industrial PC. Color feature, shape feature and area feature are adopted for multi-feature fusion in the ball detection process. A novel physical bouncing model capable of solving rotational speed, which depends on the friction coefficient and vertical restitution coefficient, is proposed for ball trajectory prediction. Experiments show that the vision system is successfully used in our ping-pong robot.