Reliable Velocity Determination Through GNSS TDCP/Doppler Combination Using an Improved FGO Framework
Haotian Wang,Shuguo Pan,Xianlu Tao,Fangyun Dong,Ling Teng,Wang Gao
DOI: https://doi.org/10.1016/j.measurement.2024.116352
IF: 5.6
2024-01-01
Measurement
Abstract:Velocity is of great significance for navigation and controlling. GNSS has become the mainstream outdoor velocity determination method by its global, all-weather, high-precision, and fast response characteristics, with time-differenced carrier phase (TDCP) and Doppler regarded as two commonly used methods for precise velocity determination using stand-alone GNSS receivers. However, TDCP aims to estimate the average velocity and is susceptible to cycle slip, while Doppler is limited by the accuracy of Doppler observations. Both methods have certain fragility in complex environments. In this contribution, a method based on factor graph optimization (FGO) using TDCP/Doppler combination is proposed. First, the proposed method considers TDCP measurements as a factor to connect Doppler measurements. This method combines the accuracy of TDCP with the usability of Doppler in complex environments, obtaining the merits of both observations. Second, an improved FGO estimator is used to obtain continuous and accuracy velocity. On one hand, the robust function of dynamic covariance estimation (DCE) avoids the need to adjust parameters for different data. On the other hand, the error envelope is generated based on the minimum detectable bias (MDB) which also proves its stronger internal reliability. Finally, kinematic tests are carried out to verify the proposed method. In urban environment, compared with Doppler and TDCP only, the estimation accuracy and usability are improved by 83.67% and 4.41%, respectively. Meanwhile, compared with WLS and EKF, the estimation accuracy is improved by 63.10% and 40.99%, respectively. It is demonstrated that the proposed method can effectively improve the velocity performance in urban environment.