Registration‐based point cloud deskewing and dynamic lidar simulation

Citations Over Time

Abstract

Abstract Point clouds captured using laser scanners mounted on mobile platforms contain errors at the centimetre to decimetre level due to motion distortion. In applications such as lidar odometry or SLAM, this motion distortion is often ignored. However, in applications such as HD mapping or precise vehicle localisation, it is necessary to correct the effect of motion distortion or ‘deskew’ the point clouds before using them. Existing methods for deskewing point clouds mostly rely on high frequency IMU, which may not always be available. In this paper, we propose a straightforward approach that uses the registration of consecutive point clouds to estimate the motion of the scanner and deskew the point clouds. We introduce a novel surface‐based evaluation method to evaluate the performance of the proposed deskewing method. Furthermore, we develop a lidar simulator using the reverse of the proposed deskewing method which can produce synthetic point clouds with realistic motion distortion.

Related Papers

• → 3D Building Scene Reconstruction Based on 3D LiDAR Point Cloud(2017)22 cited
• → Road 3D Point cloud Data Modeling based on LiDAR(2022)5 cited
• → Indoor and Outdoor Building Modeling Based on Point Cloud Data Convergence using Drones and Terrestrial LiDAR(2022)2 cited
• Susquehanna Chorale Spring Concert "Roots and Wings"(2017)
• → ИСПОЛЬЗОВAНИЕ ПОТЕНЦИAЛA СОЦИAЛЬНЫХ ПAРТНЕРОВ В ПОДГОТОВКЕ БУДУЩИХ ПЕДAГОГОВ(2024)