【作者】 吴万清；

【导师】 朱才连；

【作者基本信息】 中国科学院研究生院（测量与地球物理研究所） ， 大地测量学与测量工程， 2004， 硕士

【摘要】 GPS全球定位系统以其全球性、全天候、实时定位等优点显示出强大的生命力和竞争力，在航空、航天、航海及许多民用领域有着广泛的应用。将GPS导航方式与其他的导航方式相结合，以形成低成本的组合导航系统，是目前国际、国内车辆导航系统研究的重点和热点。车辆组合导航系统是由多种单一的导航定位模式组合而成，采用不同的导航定位模式进行组合可得到不同的组合导航定位系统。将RTK(Real Time Kinematic)技术与简易型单频GPS接收机和航位推算技术(Dead Reckoning简称DR)相结合，可形成基于载波相位差分的车载GPS、DR组合导航定位系统。如何处理导航系统中各种导航位置传感器的数据信息，以实现降低系统成本、提高系统定位精度的目的，成为整个导航系统设计中的关键问题。本论文针对基于载波相位差分GPS、DR组合导航定位系统的特点，设计了适合于各子系统的传感器数据处理算法及整个系统的数据融合算法： 1．针对目前在RTK动态定位中存在着整周模糊度解算速度慢、对解算数据要求高的缺点，提出了基于遗传算法的模糊度函数法搜索策略算法，通过大量的算例计算和仿真计算，该新算法能够部分解决诸多传统整周模糊度解算方法无法解决的问题，从而提高RTK整周模糊度解算速度，以满足动态定位的需要。 2．DR子系统中测量方程是非线性的，因此为了解决测量方程线性化误差对DR子系统滤波算法的影响，同时为了避免在卡尔曼滤波过程中由于数值稳定性问题而引起的滤波发散，应用了U-D分解的自适应迭代序列滤波算法对DR子系统的压电陀螺仪和车辆里程计数据进行滤波处理，仿真结果表明，该新算法能够大大提高DR子系统的导航定位精度。 3．对GPS子系统则采用普通的自适应卡尔曼滤波对其进行滤波处理。 4．在数据融合过程中：将RTK导航信息作为一部分，GPS、DR作为一部分，采用精度及可靠性优先的原则进行融合，其中应用一个容错性高的联邦卡尔曼滤波器对GPS、DR子系统的数据进行融合。更多还原

【Abstract】 Global Positioning System(GPS)is a satellite navigation system designed to provide instantaneous three-dimensional position infbrmation almost anywhere on the globe at any time, and in any weather .It have greatly applied in many areas such as aeronautics, astronautics, marine and other civil fields. Using GPS combined with other navigation methods to form integrated navigation system, has been a stress and a challenging issue in China and other countries in the world. Integrated land vehicle navigation system is composed of many single navigation methods, if we use different methods to integrate, we will obtain different integrated land vehicle navigation methods. Using Real Time Kinematic technique to integrate with sample single frequency GPS receiver and Dead Reckoning technique, will form an integrated GPS, DR land vehicle navigation system based on RTK. For the aim of reducing the cost and improving the navigation precision of the system, the most important in designing the whole system is how to dispose of the Position information data obtained by position sensors in it. Corresponding with trait of the integrated navigation system, the disposing data algorithms suitable for obtained by every sub-systems in the system and data integrated algorithm for the whole system have been designed in this dissertation:(1) For dealing with the defections of spending long time on ambiguity resolution and requiring high quality to the data in the RTK, a new algorithm named as Ambiguity Function Method Search Strategy Based on Genetic Algorithm is proposed. The new algorithm can improve the speed of ambiguity resolution in RTK by partly resolving difficulties those cannot be resolved by many traditional ambiguity resolutions, it has been proved by many examples of using which to deal with RTK survey data and computer simulation results.(2) Surveying equation of DR sub-system is not linear. A new algorithm named as U-D Factorization Adaptive Extended Iterative serial Filter algorithm is proposed for filtering the data obtained by angular rate sensor and milemeter. The new algorithm can greatly improve the positioning precision of DR sub-system by resolving the affection caused by linearization of surveying equation and avoiding divergence of Kalman filter caused by instability of values, it has been proved by computer simulation results.(3) Using Adaptive Kalman filter to filter data obtained by GPS sub-system.(4) Look upon the navigation information of RTK sub-system as one part, and the navigation information of GPS, DR as another one, using the principle of the priority of precision and reliability to integrate the two part information. A united Kalman filter to integrate information of GPS and DR sub-system has been used in the process.更多还原