| Vision has the advantage of broad sensing area and full information. It is one of the most important consciousness of robot. And it is one of the hotspot of robot field. In order to satisfy the high accuracy and real-time requirement of the visual servo system, we studied and designed a set of visual servo system using GPM series 2-DOF parallel mechanism of Googol Technology Co. Ltd and TMS320C6201 based on WT6201PA board of Wintech Technology Co. Ltd.In this paper, we studied intelligent robot visual technology, a method of 2D maximum between-cluster variance(2DMVB) is used in image segmentation. At the same time, we try to improve the speed of segmentation by adaptive genetic algorithm based on the elitist strategy and adjust the function by stages. The satisfied segmentation effects have been obtained. At the same time, we detail discussed the shape heart track algorithm and the visual localization technology. In this system, we fully using the powerful and high-speed data processing capability of TMS320C6201 to process the image, and reasonably using the FPGA based on the board as cooperate processor to complete filter processing. It enhanced the system processing speed Effectively.The experiment result shows that the system we have designed can track the low speed moving target rapidly and accurately. It has the certain theory significance and the practical value. |
PDF Full Text Request