Discussion on Multi-Sensor Fusion Obstacle Recognition Method for Mining Trucks in Open-Pit Mines
Abstract：The roads in open-pit mines have potholes and dense gravel, and accurate perception of obstacles is the core and difficulty in achieving autonomous driving of mining trucks. A fusion perception method based on LiDAR and camera is proposed for obstacle recognition. A joint calibration method between LiDAR point cloud and camera image is designed to achieve coordinate alignment between point cloud and image. A target detection algorithm combining 2D and 3D information is used, where the image detects obstacles and regresses the center point of the target, while the point cloud detects obstacles and regresses the set of target center points. A perception target fusion algorithm based on KNN algorithm and binary tree model is proposed. The KNN algorithm realizes the fusion of target center points, and the binary tree model is used to improve the efficiency of target matching.
Machine-generated Keywords：
Publication Date：2025-08-30
Online Publishing Date：2025-09-12(First online date of this platform, not the publication date of the document)
Pages：4( 70-73 )