Abstract:

To solve the problems of slow speed and resource waste caused by manual centering of the water filling nozzle during water filling operation of an emergency sprinkler in open-pit mines, a laser radar point cloud-based automatic identification and location method for water filling nozzle of the sprinkler was proposed, so as to achieve automatic centering of the water filling nozzle and improve the efficiency and accuracy of water filling. According to the position parameters of the water filling structure, sprinkler, and laser radar, the mutual conversion formula of the coordinates of the water filling system was derived. Then, a laser radar was used to scan the on-site environment, and a three-dimensional (3D) point cloud model of the sprinkler was constructed. The point cloud data was processed using the through filtering method and the density-based clustering algorithm, respectively. The region of interest (ROI) was delineated, and the feature information of the target was extracted. Finally, the least squares method in Matlab was used to perform the optimal circle fitting on the processed point cloud model, and a fitting circle model for the water filling nozzle was established to obtain the center coordinates of the circle and compare them with the measured results. The research results show that the circle center coordinates of the water filling nozzle of the sprinkler calculated by this method are similar to the measured data, with a location error of less than 2 cm, which is within the allowable range. This proves that it can be applied to the identification and location of the water filling nozzle of the sprinkler used in open-pit mines.

Key words: open-pit mine, laser radar, coordinate transformation, point cloud data, least square method