Multi-GNSS PPP/INS/Vision/LiDAR tightly integrated system for precise navigation in urban environments

Abstract

Nowadays, high-precision navigation is a fundamental prerequisite for autonomous driving technology. However, it is often difficult for a stand-alone sensor to meet the needs of high-precision navigation in complex scenarios. To address this problem, we propose a tightly coupled precise point positioning (PPP)/inertial navigation system (INS)/Vision/LiDAR integration method to achieve high-precision, continuous and reliable navigation in urban environments. The multi-GNSS carrier phase and pseudorange measurements, low-cost microelectromechanical system (MEMS) inertial measurement units (IMU) records, sparse visual landmarks and extracted LiDAR edge/planar features are directly fused at the observation level through a centralized Extended Kalman Filter (EKF). The vehicle experiments in different GNSS availability conditions were conducted to evaluate the proposed method. Results indicate that our proposed PPP/INS/Vision/LiDAR integration can maintain sub-meter level positioning in both GNSS half-open-sky and difficult environments, with improvements of (50.7%, 58.6%, 54.3%) and (46.2%, 55.0%, 58.8%) relative to PPP/INS/Vision and PPP/INS/LiDAR, respectively. Moreover, both visual and LiDAR information can significantly improve the velocity and attitude estimation performance, especially for the heading.