CLASLIB: An Open-source Toolkit for Low-cost high-precision PPP-RTK Positioning

This paper overviews the function and property of CLASLIB including its algorithm to achieve high-precision positioning. In addition, the positioning accuracy and convergence with low-cost receivers and survey-grade high-performance receivers are demonstrated. The operational service of the Centimeter Level Augmentation Service (CLAS) of Japan's Quasi-Zenith-Satellite-System (QZSS), a nation-wide satellite-based PPP-RTK (Real-Time Kinematic) service in Japan having centimeter-level positioning accuracy, began on November 1, 2018. The authors created CLASLIB, an open-source toolkit for PPP-RTK as a reference implementation. In the toolkit, an adaptive filter (AF) and partial ambiguity resolution (PAR) are introduced to improve the positioning accuracy when the ionosphere is active, such as in the event of medium-scale traveling ionospheric disturbances (MSTID), and to reduce the convergence time to obtain more precise solutions. Evaluation results indicate that the AF is more robust against the active ionosphere than the conventional EKF with constant processing noise. The positioning accuracy using survey-grade high-performance receivers is similar to that of a conventional RTK, whereas the required data rate is significantly lower. The statistics with a low-cost GNSS receiver prototype in static field test also meet the specifications of CLAS. Positioning performances in kinematic field test are also maintained with very slight degradation.