In steel mill slag road transportation, accurate control of vehicle loading volume is critical to reducing road and bridge damage. However, traditional detection methods suffer from high cost, low efficiency and insufficient accuracy. This paper aims to propose an effective and high-precision method for loading volume detection of slag-dedicated transport vehicles.
A LiDAR point cloud-based detection method is proposed for slag transport vehicles. First, a loading volume calculation model is established using real 3D LiDAR scan data. A denoising algorithm combining local outlier factor with KD-trees is adopted to suppress dust and equipment interference. Second, a label-connected domain clustering algorithm and statistical filtering are used to segment the cab and cargo compartments. A spatial slicing method is applied to extract the loaded material point cloud. Finally, loading quality indicators and overloading judgment criteria are constructed, and high-precision volume detection is realized via triangular mesh visualization.
Experiments are conducted on 20, 24 and 45 m³ dedicated transport vehicles under various working conditions. The average relative error of the proposed method is less than 2%, indicating high measurement accuracy.
This work introduces a complete LiDAR point cloud-based framework for volume detection of slag-dedicated transport vehicles in steel mills. The method is robust to complex on-site interferences and applicable to different vehicle sizes, showing good universality for industrial slag transportation scenarios.
