Variable Structure Control For Image-Based Visual Servoing Robotic System

Robot visual servoing is a fundamental research subject with very important theoretical research sense and industry application prospect. With the development of computer, robot, image processing and sampling, and control techniques, robot visual servoing is widely applied in the fields of industry, aerospace, medical treatment, and so on. It has become an important branch in intelligent robot field.For the control problem of image-based eye-in-hand robot visual servoing system, variable structure theory is introduced to design the controller and then many effective control strategies are presented in this dissertation. The main works are summarized as follows:Firstly, considering the eye-in-hand robot binocular visual servo system with traditional vision model, the article proposes a method of controller design based on the variable structure control theory. It gives the camera translation and rotation velocity and an adaptive update law is used to estimate the depth of selected features of the object. Under the driving of the controller, the positioning task can be completed based on the image error.Secondly, for the control problem of binocular visual servo, a variable structure controller is proposed which is designed for the new binocular stereo model presented recently. It avoids measuring and estimating the depth of object point, which can greatly improve the control performance. And then, the image-based visual system is proved to be global asymptotically stable by the Lyapunov theory. To show the effectiveness of the proposed method, simulation results are given.Finally, in consideration of robot dynamics, we propose a uncalibrated visual servoing positioning control scheme with a Jacobian matrix estimator. It does not need a priori-knowledge of the robot system, such as camera internal and external parameters. The proposed visual servoing scheme ensures the convergence of the image-features to desired position, by using the estimated Jacobian matrix, which is proved by the Lyapunov stability theory.