Precise and Robust IMU-Centric Vehicle Navigation via Tightly Integrating Multiple Homogeneous GNSS Terminals

In the realm of signal interference mitigation and precise attitude estimation, the utilization of multiple homogeneous global navigation satellite system (GNSS) antennas/terminals has demonstrated notable advantages. However, the potential benefits of homogeneous sensor fusion for enhancing position estimation have been relatively unexplored. In this article, we propose a tightly coupled (TC) precise point positioning (PPP)/inertial navigation system (INS) vehicle navigation algorithm without base stations, which can integrate an arbitrary number of homogeneous GNSS terminals with the aim of improving position estimation and speeding up system recovery from interference. Specifically, for the measurement representation, the original pseudorange and carrier-phase observations from all GNSS terminals are tightly incorporated with the inertial measurement unit (IMU) data to enable optimal state estimation. On the other hand, we exploit the known geometry between terminals and the spatially atmospheric correlation of observations to compress the states, thus to accelerate system convergence. Furthermore interestingly, since each GNSS terminal is peer-to-peer rather than master-slave in the proposed IMU-centric system, it enables a seamless operation without switching even in the event of an unexpected GNSS terminal failure. Real-world experiments have demonstrated that the proposed system can achieve a root mean square error (RMSE) of approximately 0.3 m for horizontal position estimation, and 92.24% availability at a 0.5-m boundary, outperforming the prevailing methods in terms of position estimation, state convergence, and performance in extreme scenarios.