多星座多频率GNSS非差非组合PPP-RTK关键模型研究

GPS-only PPP-RTK (Precise Point Positioning based on Real-Time Kinematic networks) encounters the problem of relative long time to first fix (TTFF) of ambiguity resolution and low reliability of ambiguity-fixed solutions, making PPP-RTK unable to be widely used in the field of real-time high-precision location services. This project is to study the key models and theoretical methods of multi-constellation and multi-frequency GNSS undifferenced and uncombined PPP-RTK to improve the application level of PPP-RTK in real-time precise positioning. The key issues of the project are as follows: 1) Establishing the Kalman filtering model of estimating real-time multi-frequency phase fractional cycle bias (FCB) based on the undifferenced and uncombined PPP processing, improving the computational efficiency and model flexibility for FCB estimation. 2) Investigating a hybrid strategy for PPP ambiguity resolution to solve the problem of low reliability of single ambiguity resolved method in real-time scenarios. 3) Optimizing the gird modeling of real-time high-precision atmospheric delay, providing an effective scheme for evaluating the accuracy of the grid-based atmospheric delay. 4) Refining the function and stochastic models of atmospheric delay enhanced PPP, improving the positioning performance of PPP-RTK. Based on the research results, the multi-constellation and multi-frequency GNSS undifferenced and uncombined PPP-RTK software is developed and the related products and services for PPP-RTK applications are provided, which provides effective technical and theoretical support for GNSS real-time high-precision location services.

针对单GPS系统PPP-RTK技术存在首次模糊度固定时间较长、固定可靠性不高的问题，本项目研究多星座多频率GNSS非差非组合PPP-RTK定位关键模型及理论方法，以提升PPP-RTK技术在实时高精度位置服务领域的应用水平。主要研究内容包括：1）基于非差非组合PPP，构建基于卡尔曼滤波的实时多频率相位小数周偏差（FCB）估计模型，提升FCB计算效率与估计模型的灵活性；2）研究PPP模糊度固定的混合策略，解决实时场景下单一模糊度固定方法可靠性较低的问题；3）优化实时高精度大气延迟格网模型，提供评估格网大气延迟精度的有效方案；4）精化大气延迟增强PPP的函数模型与随机模型，提升非组合PPP-RTK的定位性能。根据上述研究成果，开发多星座多频率GNSS非差非组合PPP-RTK软件并提供PPP-RTK相关产品和服务，为GNSS实时高精度位置服务提供有效的技术与理论支撑。

精密单点定位（PPP）技术能够提供全球高精度定位结果，但存在定位收敛时间长的技术难题，从而影响其大规模应用。针对PPP收敛速度慢、模糊度固定可靠性不高等问题，本项目构建了基于模糊度固定技术的大气延迟增强非组合PPP-RTK模型，以期解决上述问题。基于此，建立了基于卡尔曼滤波的实时相位小数偏差估计模型，探究该模型下合理的质量控制策略；在精细考虑伪距和载波相位硬件偏差时变特性的基础上，导出了更为严谨的非差非组合观测方程，并给出了非组合模式下两类GNSS偏差的数学表达形式，在此基础上精化PPP函数模型与随机模型；将最优整数等变（BIE）估计引入PPP模糊度估计过程中，利用GNSS模糊度整数解加权融合以获得最优浮点模糊度估计值，可有效回避模糊度错误固定风险，同时又利用了模糊度整数解信息来提升模糊度估值精度，从而提升PPP定位精度，缩短模糊度收敛时间。在本项目的资助下，发表SCI论文3篇，EI论文3篇，授权发明专利1项。本项目研究的技术与方法适用于大区域、复杂场景的实时高精度位置服务，理论意义与实用价值较突出，在无人驾驶、智能交通、精细农业等领域有广泛的应用前景。