P³-LOAM: PPP/LiDAR Loosely Coupled SLAM With Accurate Covariance Estimation and Robust RAIM in Urban Canyon Environment

Light Detection and Ranging (LiDAR) based Simultaneous Localization Mapping (SLAM) has drawn increasing interests in autonomous driving. However, LiDAR-SLAM suffers from accumulating errors which can be significantly mitigated by Global Navigation Satellite System (GNSS). Precise Point Positioning (PPP), an accurate GNSS operation mode independent of base stations, gains growing popularity unmanned systems. Considering the features two technologies, PPP, this paper proposes a SLAM system, namely P 3 -LOAM (PPP LiDAR Odometry Mapping) couples PPP. For better integration, we derive positioning covariance using Singular Value Decomposition (SVD) Jacobian model, since SVD provides explicit analytic solution Iterative Closest (ICP), is key issue LiDAR-SLAM. A novel method then proposed to evaluate estimated covariance. In addition, increase reliability urban canyon environment, develop assisted Receiver Autonomous Integrity Monitoring (RAIM) algorithm. Finally, validate with UrbanNav, challenging public dataset environment. Comprehensive test results prove that, terms accuracy availability, outperforms benchmarks such as Single (SPP), LeGO-LOAM, SPP-LOAM, loosely coupled navigation system publisher UrbanNav.