A monocular visual servoing control system for mobile robot

A monocular visual servoing control system is developed. Focus on the problem for restoring depth information, a method based on fuzzy model is proposed, which is a model-independence method, and it uses the pitch degree of the camera and the position of the object in the image followed y axis of the image reference frame as the fuzzy input parameters, the fuzzy relationship matrix is built based on experiments, the fuzzy output parameter is gotten from the input parameters and the fuzzy relationship matrix by means of fuzzy operations, eventually, the depth information of objects is obtained by defuzzified output parameter. Contrast to the traditional restoring depth information methods used in the monocular vision system, the proposed method need not build the model of objects, and it do not request some additional constraints and aids also. Experiments to restore the depth information of objects using the proposed method are contrasted to employing the traditional method. The results verify that the proposed method is effective. Experiments of grasping and conveying one ball illuminate that the developed monocular visual servoing control can navigate the mobile robot exactly based on the proposed restoring depth information method.