Design of Table Tennis Picking and Serving Robot Based on Machine Vision

Abstract:
Aiming at the problems of the existing table tennis server, such as immovability, single operation and inability to integrate serving and picking up the ball, this paper designs a table tennis picking up and serving robot. The image obtained by the camera is preprocessed by median filtering, corrosion expansion, image segmentation and Hough circle detection, etc., and the table tennis is identified and acquired by combining shape and color features. This paper puts forward a method of acquiring the depth information of table tennis based on camera angle feedback ranging, which can realize the positioning of table tennis accurately and efficiently. Experiments show that the robot can accurately and efficiently identify, locate, grab and serve table tennis in different environments.