Tightly coupled multi-frequency PPP-RTK/INS integration model and its application in an urban environment

Abstract

The Precise Point Positioning-Real Time Kinematic (PPP-RTK) technique, which provides centimeter-level positioning with instantaneous ambiguity resolution, is considered as a potential tool for intelligent vehicle applications. However, its performance is restricted under complex urban conditions owing to intermittent signal interruptions and poor satellite geometries. Thus, a tightly coupled Multi-Frequency (MF) PPP-RTK/INS (Inertial Navigation System) model was developed with the objective of providing a stable and reliable positioning for the urban vehicle. In this model, the augmentation of INS information, third-frequency observations, precise atmospheric corrections, the fixed Extra-Wide Lane (EWL), and Wide-Lane (WL) ambiguities can be used to enhance the positioning performance of PPP-RTK. We designed urban vehicle experiments under different scenarios to validate the proposed method. The results show that PPP-RTK can be significantly improved for urban vehicle positioning by fusing the MF and INS. In urban areas, the solution availability with a horizontal positioning error within 10 cm was 96.1% for MFPPP-RTK/INS with a fixing percentage of 90.9%. Compared with the dual-frequency PPP-RTK solutions, the fixing percentage and solution availability in the MFPPP-RTK/INS was improved by 9.5% and 8.8%, respectively. Moreover, MFPPP-RTK/INS provides continuous and stable positioning and fast ambiguity recovery in GNSS-challenged environments. The MFPPP-RTK/INS could achieve a fast ambiguity re-fixing within 1 s after continuously crossing obstacles, whereas PPP-RTK could achieve the same in 10 s.

Keywords

• Triple-frequency
• PPP-RTK/INS
• urban vehicle navigation