Visual Servo Stabilization Of Wheeled Mobile Robots In Dynamic Scenes

Visual as the most significant senses of human access to information,received the image data contains a wealth of information.The visual sensor is one of the most important robot sensors because its non-contact measurement,large amount of information,low cost and wide application range.In recent decades,the research of mobile robots around the world has been an unprecedented development,such as universe discovery,deep sea exploration,and family services.Visual servoing is an important research direction in the field of robot control.Its main idea is to use the visual sensor to obtain image information to control the mobile robot to reach the desired position/orientation.Visual servoing stabilization control is a difficult point in robot vision control and existing methods usually assume that the external scene is static.In the actual scene,the disturbance of the external environment may cause the visual features to move,and the feature model information and the scene depth information are not necessarily known,which brings challenges to the visual servoing stabilization.Therefore,it will be of great significance to complete the stabilization control task under dynamic scenes.In general,the main contributions of the dissertation are summarized as follows:1)This thesis introduces the nonholonomic system and the nonholonomic wheeled mobile robots and kinematics model is established based on the kinematics characteristics of wheeled mobile robot.The experiment platform of visual servoing system is constructed to test the designed algorithms conveniently.Meanwhile,the designed software has the strong ability of the reusability,maintainability,expansibility.2)In order to solve the problem of visual servoing stabilization in dynamic scenes,proposes a visual servoing strategy monitoring-based,which can complete the visual servoing stabilization task under dynamic visual targets.Specifically,the visual servoing strategy is designed based on the monitoring vision system,the mobile robot system,coordinate system transformation and the POSIT algorithm,which the 3D pose of the feature points is known.Finally,the mobile robot is driven to the desired pose in a dynamically changing scene.The performance of the proposed strategy is validated by both simulation and experimental results.In order to reflect the advantages of the method,the scheme is used to compare the experiment with the traditional visual servoing stabilization strategy3)A new method is proposed for wheeled mobile robot visual servoing using model-free visual targets under dynamic scenes and depth information unknow.The method is designed based on six-degree-of-freedom homography matrix,coordinate system transformation and adaptive control algorithm and so on.On this basis,set 4 or more coplanar feature points in the scene as image feedback information.Simulation and experimental results show the feasibility of the proposed strategy.4)In order to improve the environmental adaptability and intelligence of mobile robots,two sets of experimental schemes were designed in combination with wireless communication technology and the strategy is verified by the Pioneer 3 mobile robot.The experimental results are collected to verify the feasibility of the proposed approach.