In this project, a classification prototype was developed using a 5DOF robot and a camera system. The image processing system is used to detect the objects for position, orientation and color with high accuracy based on Hough-line functions. From the acquisition information collected, the robot will catch proper objects and arranges them into a storage. In order to obtain high-accuracy picking behaviors in the nonstop working conditions, proper inverse kinematics and optimal path planning algorithm were adopted. Experimental results achieved show the effectiveness and feasibility of the proposed prototype.
Authors: Vo Nguyen Thong, Nguyen Dang Hung Phu, and D. X. Ba
Ack: The research team would like to thank Mr. Nguyen Huu Phong, Mr. Le Tuan Anh, and Mr. Nguyen Dang Phuc for sharing their achievement.