Covariance estimation for the Iterative Closest Point (ICP) point cloud registration algorithm is essential for state estimation and sensor fusion purposes. We argue that a major source of error for ICP is in the input data itself, from the sensor noise to the scene geometry. Benefiting from recent developments in deep learning for point clouds, we propose a data-driven approach to learn an error model for ICP. We estimate covariances modeling data-dependent heteroscedastic aleatoric uncertainty, and epistemic uncertainty using a variational Bayesian approach. The system evaluation is performed on LiDAR odometry on different datasets, highlighting good results in comparison to the state of the art.
Towards Robust LiDAR Localization: Deep Learning-based Uncertainty Estimation
Department of Computer Engineering, Amirkabir University of Technology(阿姆irkabir技术大学计算机工程系);Graduate School of Information Science and Technology, The University of Tokyo(东京大学信息科学与技术研究生院)
P3-LOAM: PPP/LiDAR Loosely Coupled SLAM with Accurate Covariance Estimation and Robust RAIM in Urban Canyon Environment
Shanghai Jiao Tong University(上海交通大学)
基于激光雷达(LiDAR)的同时定位与地图构建(SLAM)技术在自动驾驶领域引起了越来越多的关注。然而,LiDAR-SLAM遭受累积误差,这可以通过全球导航卫星系统(GNSS)显著减轻。精密单点定位(PPP)是一种不依赖于基站的高精度GNSS工作模式,在无人系统中越来越受欢迎。针对LiDAR-SLAM和PPP这两种技术的特点,提出了一种将LiDAR-SLAM和PPP相结合的SLAM系统P3-LOAM(PPP based LiDAR Odometry and Mapping)。为了更好地集成,我们通过使用奇异值分解(SVD)雅可比模型推导LiDAR-SLAM定位协方差,因为SVD提供了迭代最近点(ICP)的显式解析解,这是LiDAR-SLAM中的关键问题。然后提出了一种新的方法来评估估计的LiDAR-SLAM协方差。此外,为了提高GNSS在城市峡谷环境中的可靠性,我们开发了一种LiDAR-SLAM辅助GNSS接收机自主完整性监测(RAIM)算法。最后,我们验证P $^3$ -LOAM与UrbanNav,一个具有挑战性的公共数据集在城市峡谷环境。综合测试结果表明,P3-LOAM在精度和可用性方面优于单点定位(SPP)、PPP、LeGO-LOAM、SPP-LOAM以及UrbanNav出版商提出的松耦合导航系统等基准。