Research on Identification Method for Interface Flange in Automatic Docking System of Fluid Loading and Unloading Arm for Bottom Loading

Abstract

The automatic docking system of the loading and offloading arm of a tank car is the key link to realizing the unmanned operation of tank car loading and unloading. The spatial position detection of the flange port of a tank car can guide the automatic docking of the fluid loading and offloading arm and flange port of the tank car. In this paper, a flange position detection method based on image recognition was proposed. Firstly, the end state of the loading arm was analyzed to determine the expression mode of the loading arm's spatial pose so as to form a unified expression with the flange position and docked pose on the tank car. Then, for the image processing of the flange port of the tank car, this paper binarized the edge of the flange end face based on the Otsu algorithm, used the Canny algorithm for edge detection, used the least squares method to fit the image edge coordinates into a spatial circle, calculated the center coordinates and normal vector of the flange end face, and used these parameters to guide the end of the loading arm to adjust the position and attitude so that it was consistent with the position and pose of the flange port to realize docking. Then, a circular object center detection and calibration experiment, a flange end face image experiment, and an automobile tank car flange port physical detection experiment were carried out. The test results show that the spatial coordinate accuracy of the flange port diameter and center detected by this method meets the requirements of the loading arm automatic docking system, providing a research idea for the design of an automatic docking system for the loading and unloading arm of a tank car.