High-Accuracy Vehicle Pose Estimation Method Using Roadside Nonrepetitive Scanning LiDAR

With the development of vehicle-to-infrastructure (V2I) cooperation, roadside LiDAR sensors offer a novel solution for vehicle pose estimation. This paper proposes an infrastructure-based high-accuracy vehicle pose estimation method, offering a full pipeline from raw data to final estimation results. We employ non-repetitive scanning LiDAR as the primary measuring instrument, as its novel scanning pattern can eliminate static blind zones and is more cost-efficient compared to traditional rotating multi-line LiDAR. However, due to the sparsity and irregular sampling patterns of non-repetitive scanning LiDAR, existing methods designed for rotating multi-line LiDAR fail to deliver sufficient precision. To address the challenges arising from such LiDARs, we introduce an instance segmentation method tailored for non-repetitive scanning LiDAR and propose a grid-variance-based optimization approach to construct high-accuracy vehicle geometry template. These templates are then used in a registration-based framework for continuous vehicle pose estimation. Experiments in two real-world scenarios show that our method achieves an average position error of around 7cm and a heading error of 1.1°, outperforming existing baselines in vehicle pose estimation. The dataset and source code will be available at https://github.com/ElizabethSF/Register-Loc.