Real-Time Stereo Visual Servoing of a 6-DOF Robot for Tracking and Grasping Moving Objects

Robotic systems have been increasingly employed in various industrial, urban, mili-tary and exploratory applications during last decades. To enhance the robot control per-formance, vision data are integrated into the robot control systems. Using visual feedback has a great potential for increasing the flexibility of conventional robotic and mechatronic systems to deal with changing and less-structured environments. How to use visual in-formation in control systems has always been a major research area in robotics and mechatronics. Visual servoing methods which utilize direct feedback from image features to motion control have been proposed to handle many stability and reliability issues in vision-based control systems. This thesis introduces a stereo Image-based Visual Servoing (IBVS) (to the contrary Position-based Visual Servoing (PBVS)) with eye‐in‐hand configuration that is able to track and grasp a moving object in real time. The robustness of the control system is in-creased by the means of accurate 3-D information extracted from binocular images. At first, an image-based visual servoing (IBVS) approach based on stereo vision is proposed for 6 DOF robots. A classical proportional control strategy has been designed and the ste-reo image interaction matrix which relates the image feature velocity to the cameras’ ve-locity screw has been developed for two cases of parallel and non-parallel cameras in-stalled on the end-effector of the robot. Then, the properties of tracking a moving target and corresponding variant feature points on visual servoing system has been investigated. Second, a method for position prediction and trajectory estimation of the moving tar-get in order to use in the proposed image-based stereo visual servoing for a real-time grasping task has been proposed and developed through the linear and nonlinear model-ing of the system dynamics. Three trajectory estimation algorithms, “Kalman Filter”, “Recursive Least Square (RLS)” and “Extended Kalman Filter (EKF)” have been applied to predict the position of moving object in image planes. Finally, computer simulations and real implementation have been carried out to verify the effectiveness of the proposed method for the task of tracking and grasping a moving object using a 6-DOF manipulator.