Algorithm Design of GNSS/INS Integrated Navigation for Vehicle Location in Cities

With the development of modern urban intelligent transportation system, the requirement of accuracy and reliability of autonomous navigation becomes higher and higher. GNSS/INS is the most widely used method of integrated navigation at present, but in cities, satellite signals tend to be blocked by tall buildings, resulting in fewer than four visible satellite. This paper presents loosely-coupled and tightly-coupled GNSS/INS integration models using EKF and UKF algorithm respectively. The system simulation platform is built, and the dilution of precision (DOP) is adopted to help the satellite selection which ensures the effectiveness of integrated navigation. The experimental results show that when the satellite signal is more than four, both the integration models have high navigation accuracy. When the satellite signal is less than four, using the tightly-coupled model instead of the invalid loosely-coupled model with the observable satellite information can ensure the navigation precision and improve the anti-interference ability of the system.