Measurement of the Wheel-rail Relative Displacement using the Image Processing Algorithm for the Active Steering Wheelsets

초록

Abstract: The central problem of the active wheelsets steering systems of the railway vehicles is the measurement of the lateral movements of wheelsets with the noncontact manner. This paper describes the development of a real-time vision system for active steering of a railway vehicle to detect a lateral displacement between wheel and rail in various dynamic environments. An active steering system has proven its ability to reconcile the contradictingrequirements of running stability and track friendliness on curve sections and it consists of one actuator, a controller, and sensors. As one of the sensor system for detecting the wheel/rail relative displacement in active steering bogie, the image processing techniques were investigated with the aid of the vision based image processing algorithm. In this paper, we present the design and implementation of a real time vision system for a railway vehicle to measure its wheel/rail displacement. In order to extract wheel and rail features, this paper uses the various successive image processing algorithms which consist of the camera compensation algorithm against the degradation image quality, the algorithm for improving the contrast effect based on the block of the input image, the pre-processing algorithm coped with the dynamic illumination environment based on the edge, the line segment extraction algorithm of the wheel and rail using the RHT (Random Hough Transformation), and the wheel-rail lateral displacement calculation algorithm. Running test results show that our proposed lateral displacement measurement method is accurate in each sampling rate, and make it a sensor to be placed in the active steering control systems.

키워드