Algorithm for Point Cloud Dust Filtering of LiDAR for Autonomous Vehicles in Mining Area

Xianyao Jiang,Yi Xie,Chongning Na,Wenyang Yu,Yu Meng

DOI: https://doi.org/10.3390/su16072827

IF: 3.9

2024-03-29

Sustainability

Abstract:With the continuous development of the transformation of the "smart mine" in the mineral industry, the use of sensors in autonomous trucks has become very common. However, the driving of trucks causes the point cloud collected by through Light Detection and Ranging (LiDAR) to contain dust points, leading to a significant decline in its detection performance, which makes it easy for vehicles to have failures at the perceptual level. In order to solve this problem, this study proposes a LiDAR point cloud denoising method for the quantitative analysis of laser reflection intensity and spatial structure. This method uses laser reflectivity as the benchmark template, constructs the initial confidence level template and initially screens out the sparse dust point cloud. The results are analyzed through the Euclidean distance of adjacent points, and the confidence level in the corresponding template is reduced for rescreening. The experimental results show that our method can significantly filter dust point cloud particles while retaining the rich environmental information of data. The computational load caused by filtering is far lower than that of other methods, and the overall operation efficiency of the system has no significant delay.

environmental sciences,environmental studies,green & sustainable science & technology

What problem does this paper attempt to address?

### What problem does this paper attempt to solve?

This paper primarily addresses the issue of point cloud data collected by LiDAR (Light Detection and Ranging) sensors in open-pit mining areas containing a large amount of dust particles. Specifically:

1. **Background and Challenges**:
- In open-pit mining areas, due to loose road surfaces and the dust generated by moving vehicles, the point cloud data collected by LiDAR contains noise.
- These dust point clouds severely affect the effectiveness of environmental perception, thereby impacting the safety and reliability of autonomous vehicles.

2. **Research Objectives**:
- Propose a new LiDAR point cloud denoising method aimed at quantitatively analyzing laser reflection intensity and spatial structural features.
- This method can effectively filter out sparse dust point clouds while retaining rich environmental information.
- The specific steps of the method include:
1. Using the principle of random sampling to process the original point cloud and remove the background.
2. Using laser reflectivity as a reference template to construct an initial confidence template, preliminarily filtering out sparse dust point clouds.
3. Analyzing non-ground points through the Euclidean distance of neighboring points, identifying parts of the point cloud cluster with large spatial structure information variance as suspected non-rigid bodies, and reducing the confidence values in the corresponding template.
4. Filtering dense dust point clouds based on the modified confidence template.

3. **Contributions**:
- Propose a method based on laser reflectivity and 3D feature analysis that can accurately distinguish between rigid and non-rigid point clouds.
- Propose an algorithm to quantify the transparency and density of target point clouds, comprehensively considering their optical and physical properties to identify noise targets.
- Develop a robust point cloud dust filtering algorithm suitable for large-scale dust noise and outliers, outperforming state-of-the-art point cloud denoising methods.

In summary, this paper aims to improve the perception accuracy of autonomous vehicles in complex mining environments by proposing an efficient point cloud denoising method.