Vehicle pose estimation is essential in the perception technology of autonomous driving. However, due to the different density distributions of the LiDAR point cloud, it is challenging to achieve accurate direction extraction based on 3D LiDAR by using the existing pose estimation methods. In this paper, we proposed a novel convex hull-based vehicle pose estimation method. The extracted 3D cluster is reduced to the convex hull, reducing the computation burden. Then a novel criterion based on the minimum occlusion area is developed for the search-based algorithm, which can achieve accurate pose estimation. The proposed algorithm is validated on the KITTI dataset and a manually labeled dataset acquired at an industrial park. The results show that our proposed method can achieve better accuracy than the three mainstream algorithms while maintaining real-time speed.

翻译：车辆姿态估计在自动驾驶感知技术中至关重要。然而，由于LiDAR点云密度分布不均匀，现有姿态估计方法难以基于3D LiDAR实现精准的朝向提取。本文提出了一种新颖的基于凸包的车辆姿态估计方法。首先将提取的3D聚类降维为凸包，从而降低计算负担。随后基于最小遮挡区域准则开发了一种搜索算法，可实现精准姿态估计。所提算法在KITTI数据集及工业园人工标注数据集上进行了验证。结果表明，与三种主流算法相比，本方法在保持实时速度的同时实现了更高精度。