SOPC Stereo Vision Servo Robotic interaction Motion Control System

• Main Authors: Chih-yih Wu, 吳智逸
• Other Authors: Shiuh-jer Huang
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/gy9d42

碩士 === 國立臺灣科技大學 === 機械工程系 === 95 === A stereo vision servo robotic system using SOPC hardware technology is developed in this thesis. The Nios II developing board is adopted on the control structure with ALTERA Statrix II EP2S60F672C5ES FPGA chip. The CMOS color image sensor PIXART PAS106BCB283 with 356×292 pixels is adopted. First, on images acquired with two CMOS cameras connected to the Nios II development board, and then the feature of this image pair can be extracted using image processing technique. Secondly, the 3-D position information between the target object and stereo vision system can be calculated by stereo vision algorithm. Thirdly, the relation motion between the robotic end-effector and the target object can be planned to guide robot arm to catch the object. The fuzzy sliding mode control algorithm is employed on each joint doing motion control. Some tracking-and-catch games are planned to implement this stereo vision servo system on the interaction games between player and robot in 3D space. The experiments show that this stereo vision servo system can track and catch a moving target in 3D space and execute some interaction function with player under the limit SOPC computation limit.