Fuzzy PID Based Autonomous UAV Design for Railway Tracking

Abstract

Unmanned aerial vehicles (UAV) are frequently used in many areas today. Examination of defects in railways using UAV is a relatively new application for UAVs. By enabling the UAV to follow the railroad autonomously, the investigation reduces the cost and speeds up the monitoring process. However, in such an application, it is an important issue that the path detection and tracking algorithm works well. This study focuses on autonomous tracking of the railway using UAV. In the proposed rail tracking algorithm, firstly, a ray detection algorithm based on Gabor filter is proposed to detect the rails from the image taken by the UAV. The vanishing point was estimated by applying the least squares method to the obtained rail lines. Finally, the vanishing point was kept in the center of the image taken using the PID controller. A fuzzy system is designed to tune the PID control parameters. While the vanishing point was kept in the center of the image, the UAV was moved forward at a constant speed. In this way, the forward movement of the UAV was provided by centering the railway. The experimental results show that the algorithm works satisfactorily for the movement of the UAV by centering the railway. © 2021 IEEE.