面向多模多频GNSS高精度定位的混合整数模型完好性监测方法研究

The precise parameter estimation and validation of mixed-integer model play a crucial role in determining the high-precision and high reliability of carrier-phase differential positioning. Although multi-GNSS, including BDS, GPS, GLONASS, and Galileo, can bring a thorough refinement to the parameter estimation performance of mixed-integer model, some additional risk resources have been introduced as well, which will cause three critical issues when applying the traditional integrity monitoring technique for diversity of high-precision navigational applications, i.e., insufficient compatibility, non-rigorous validation, and low availability. In view of this, this contribution will firstly analyze the internal connection between the integrity monitoring availability improvement and the strength of mixed-integer model’s parameter estimation and validation. The mixed-integer model will be constructed under the ionosphere-free and combined noise variance restricted criterion, and the parameters of which are estimated by the partial ambiguity resolution method. The innovative allocation model of integrity risk and continuity risk is proposed to be compatible with multiple risk resources. The integrity monitoring methods have been proposed to rigorously validate the integer ambiguity correctness and ambiguity-fixed solution in position domain have been studied such that the availability of high-precision position solution can be ultimately improved. Finally, the experiments and applications will be carried out based on the several typical high-precision navigation and positioning test-beds. .The project is the extending and complementing of applicant’s recent research result, and the corresponding achievements are anticipated to not only provide the knowledge-base and technical supports for utilizing the satellite navigation-based precise positioning for safety-of-life applications in both the military and civil industry, but also has important theoretical and practical significance on the multi-GNSS data processing and BDS development.

载波相位差分定位高精度和高可靠性的关键在于混合整数模型的参数精确估计与检验。多模多频GNSS虽然可以全方位提升混合整数模型的参数估计性能，但是也引入了额外的多重风险源，导致现有的完好性监测技术在面对多类型精密导航服务时存在“兼容性不足、检验不严密、可用性低”等突出问题。针对此，本项目从分析混合整数模型的构建及参数估计与完好性监测可用性提升的内在关联入手，通过构建基于消电离层和组合观测噪声可约束的混合整数模型及其整参数部分估计方法，创新性地提出可兼容多重风险源的完好性风险和连续性风险分配模型，开展整周模糊度正确性的严密检验以及定位域固定解的完好性监测研究，以充分提升高精度定位解的可用性。在此基础上，在高精度导航定位领域开展试验与应用。.本项目是申请人已有研究成果的延续和拓展，相关研究可为卫星导航精密定位在军民领域中与生命安全相关的应用提供技术支撑，具有重要的理论和现实意义。

为解决现有的完好性监测技术在面对多类型精密导航服务时存在“兼容性不足、检验不严密、可用性低”等突出问题，本项目从分析混合整数模型的构建及参数估计与完好性监测可用性提升的内在关联入手，从模型构建、参数估计、指标体系分配、模糊度域以及定位域关键风险源的监测等方面进一步完善现有的完好性监测理论与方法，具体开展的研究工作包括：(1) 基于消电离层和组合观测噪声可约束准则的混合整数模型构以及部分模糊度解算的参数估计方法，测试结果表明所提方法构造的消电离层组合具有很好的噪声抑制能力，模糊度解算成功率与快速性均得到有效提升。（2）兼容多重风险源的完好性风险和连续性风险分配，基于四状态马尔科夫模型构建同时表征故障检测和离线保护水平双重性能的代价函数，实现兼容多重风险源的完好性风险和连续性风险分配。仿真结果表明，所提出的分配方法与传统固定分配方法相比，全球保护水平可提升57.4%。（3）严密控制误警和漏检两类错误的整周模糊度正确性检验，构建同时满足误警与漏检概率约束的检测统计量与相应门限值，实现误警与漏检两类错误的同步控制。动态车载数据测试结果表明所提方法能够同时满足误警与漏检概率需求，在定位连续性与完好性间取得最优平衡。（4）基于故障模式精简集的定位域固定解在线完好性监测，并提出多类型风险源作用下水平和垂直定位误差相关性的精细补偿方法，仿真测试结果表明，所提出基于危险误导信息概率的可用性水平可提升16.8%。.通过本项目的研究，构建了混合整数模型完好性监测理论与方法。项目丰富的研究成果可作为卫星导航精密定位应用于与生命安全相关导航服务的技术支撑，例如自动驾驶、全自动着陆等应用。