The article is devoted to the design of control systems for moving objects using computer vision algorithms. The relevance of this research area is associated with the increasing use of autonomous vehicles for solving practically important tasks without human intervention. However, many of these tasks are impossible or inefficient to solve without using of visual information. The problem of visual positioning of a moving object equipped with an onboard video camera is considered. The purpose of the control is to provide the desired position of some observed object in the image plane of the camera. A mathematical model of the joint dynamics of a moving object and a set of observed points in the image plane is derived. Such joint dynamics are taken into account in control design because the camera, mounted on board, has restrictions on freedom of movement in space and on the speed of these movements. A multi-purpose approach to the synthesis of the control law is proposed, allowing to ensure the fulfillment of a set of requirements for the quality of processes in a closed-loop system in various regimes. The choice of adjustable elements of a multi-purpose structure is discussed. The application of the approach is illustrated by an example of control of a wheeled robot. The results of simulation performed in MATLAB/Simulink environment are presented.