PTZ Control System of Indoor Rail Inspection Robot Based on Neural Network Prediction Model

Abstract

With the development of the automation of electric power system, indoor inspection rail robots are applied in many transformer substations to inspect the equipment. During the inspecting of robot, the PTZ plays an important role so that the accuracy of the PTZ determines the final inspection results to a large extent. The PTZ's control accuracy is influenced by the difference of the mechanical damping and gear clearance among different PTZs. A kind of PTZ control system based on neural network prediction model was designed in this article, which can improve the control accuracy of the PTZ by auto adjust difference among different PTZs. The hardware and software was designed for the PTZ control system. The Validity and reliability of the system was verified by the experimental result, which meets the requirements of robot inspection.