Optimal Vehicle Pose Estimation Network Based on Time Series and Spatial Tightness with 3D LiDARs

Hanqi Wang,Huawei Liang,Linglong Lin,Jie Yu,Zhiling Wang,Fengyu Xu

doi:10.3390/rs13204123

Abstract

Vehicle pose estimation is essential in autonomous vehicle (AV) perception technology. However, due to the different density distributions of the point cloud, it is challenging to achieve sensitive direction extraction based on 3D LiDAR by using the existing pose estimation methods. In this paper, an optimal vehicle pose estimation network based on time series and spatial tightness (TS-OVPE) is proposed. This network uses five pose estimation algorithms proposed as candidate solutions to select each obstacle vehicle’s optimal pose estimation result. Among these pose estimation algorithms, we first propose the Basic Line algorithm, which uses the road direction as the prior knowledge. Secondly, we propose improving principal component analysis based on point cloud distribution to conduct rotating principal component analysis (RPCA) and diagonal principal component analysis (DPCA) algorithms. Finally, we propose two global algorithms independent of the prior direction. We provided four evaluation indexes to transform each algorithm into a unified dimension. These evaluation indexes’ results were input into the ensemble learning network to obtain the optimal pose estimation results from the five proposed algorithms. The spatial dimension evaluation indexes reflected the tightness of the bounding box and the time dimension evaluation index reflected the coherence of the direction estimation. Since the network was indirectly trained through the evaluation index, it could be directly used on untrained LiDAR and showed a good pose estimation performance. Our approach was verified on the SemanticKITTI dataset and our urban environment dataset. Compared with the two mainstream algorithms, the polygon intersection over union (P-IoU) average increased by about 5.25% and 9.67%, the average heading error decreased by about 29.49% and 44.11%, and the average speed direction error decreased by about 3.85% and 46.70%. The experiment results showed that the ensemble learning network could effectively select the optimal pose estimation from the five abovementioned algorithms, making pose estimation more accurate.

Highlights

We propose a new pose estimation network integrated with five potential pose estimation algorithms based on 3D Light Detection and Ranging (LiDAR)
The authors of this paper have proposed an optimal vehicle pose estimation network based on time series and spatial tightness with 3D LiDAR
In the SemanticKITTI dataset, compared with the other two methods, the polygon intersection over union (P-Intersection over union (IoU)) was found to improve by 5.25% and 9.67%, which showed its accuracy when there was no interference from preprocessing

Summary

Introduction

The continuous development of autonomous vehicles (AVs) has led to higher requirements for perception accuracy [1,2]. 3D object detection is one of its main research directions [3] that has received extensive attention from both industry and academia [4]. Image-based 3D object detection has been significantly improved with the development of deep learning [5,6,7,8,9], it is difficult to provide accurate obstacle object depth information for images. The access of depth information still relies on 3D point cloud data in actual applications [10]

Methods

Results

Discussion

Conclusion