A Study of Automatic Positioning Control Based on Vision Servo Control

The traditional automatic crane positioning method has low accuracy, poor stability and cannot form a true closed-loop control, and the existing crane visual positioning method generally suffers from poor anti-disturbance performance and difficulty in obtaining visual Jacobi matrix parameters. In order to solve the problem of external disturbance affecting the accuracy and stability of visual servo positioning, a mathematical model of the position posture of the crane end-effector under the disturbance condition is established, and a visual servo disturbance suppression method based on a self-anti-disturbance controller is proposed. According to the image projection characteristics, the equation of the differential relationship between the Jacobi matrix parameters and image features is derived, the adaptive update rate of the estimated Jacobi matrix parameters is designed, and the closed-loop dynamics equation is established. According to the image characteristic error, the Liapunov function is constructed and the proof of system stability is given. Simulation and experimental analysis are conducted, and the results show that the position and velocity profiles converge to zero values when the visual error converges, indicating that the method in this paper still has satisfactory positioning accuracy under visual uncertainty and constant, transient perturbations. The analysis of the comparison results with other control methods shows that the method in this paper can speed up the convergence of the visual error and has better stability while ensuring the positioning accuracy of the visual servo system, therefore, it is suitable for the automatic positioning visual servo system of cranes for hoisting.