From RTK to PPP-RTK: towards real-time kinematic precise point positioning to support autonomous driving of inland waterway vessels

Abstract PPP-RTK is Precise Point Positioning (PPP) using corrections from a ground reference network, which enables single-receiver users with integer ambiguity resolution thereby improving its performance. However, most of the PPP-RTK studies are investigated and evaluated in a static situation or a post-processing mode because of the complexity of implementation in real-time practical applications. Moreover, although PPP-RTK achieves a faster convergence than PPP, it typically needs 30 s or even longer to derive high-accuracy results. We have implemented a real-time PPP-RTK approach based on undifferenced observations and State-Space Representation corrections with a fast convergence of less than 30 s to support autonomous driving of inland waterway vessels. The PPP-RTK performances and their feasibility to support autonomous driving have been evaluated and validated in a real-time inland waterway navigation. It proves the PPP-RTK approach can realize a precise positioning of less than 10 cm in horizontal with a rapid convergence. The convergence time is within 10 s after a normal bridge passing and less than 30 s after a complicated bridge passing. Moreover, the PPP-RTK approach can be extended to outside of the GNSS station network. Even if the location is 100 km away from the border of the GNSS station network, the PPP-RTK convergence time after a bridge passing is also normally less than 30 s. We have realized the first automated entry into a waterway lock for a vessel supported by PPP-RTK and taken the first step toward autonomous driving of inland vessels based on PPP-RTK.