Title: Vision-guided manipulator operating system based on CSRT algorithm
Authors: Suwan Bu; Liang Yan; Xiaoshan Gao; Peiran Zhao; Chee Kian Lim
Addresses: School of Automation Science and Electrical Engineering, Beihang University, Beijing 100191, China ' School of Automation Science and Electrical Engineering, Beihang University, Beijing 100191, China; Science and Technology on Aircraft Control Laboratory, Beihang University, Beijing 100191, China; Ningbo Institute of Technology, Beihang University, Ningbo 315800, China ' School of Automation Science and Electrical Engineering, Beihang University, Beijing 100191, China ' School of Automation Science and Electrical Engineering, Beihang University, Beijing 100191, China ' School of Mechanical and Aeronautical Engineering, Singapore Polytechnic, Singapore
Abstract: This paper proposed a vision guided robot operating system based on channel and spatial reliability tracker (CSRT) algorithm to improve its environmental adaptability. Firstly, the kinematics and inverse kinematics models of the manipulator are deduced, and the transformation matrices of position and orientation between the joint axes are obtained. Then the imaging model and distortion model of the camera are established, and the whole manipulator system is calibrated. The transformation matrix between pixel coordinate system and base coordinate system is obtained. Following that, the target position is determined and tracked by CSRT algorithm. Two experiments are conducted on the developed CSRT algorithm. One validated the adaptability of CSRT algorithm to the environment under normal and weak light conditions, and the other show that the visual guided manipulator grasping system with the algorithm can complete the grasping task well.
Keywords: visual guidance; CSRT algorithm; manipulator grasping; system calibration.
International Journal of Hydromechatronics, 2022 Vol.5 No.3, pp.260 - 274
Received: 14 Apr 2022
Accepted: 18 Apr 2022
Published online: 25 Aug 2022 *