摘要
为了在弱信号情况下更有效地实现定位,对辅助全球定位系统(AGPS)中2颗星定位算法进行了研究,采用AGPS定位系统加快卫星捕获和提供辅助数据.利用Marquardt算法进行位置解算,加入了步长因子,根据目标函数的变化情况自适应调整步长,解决了Gauss-Newton算法对病态方程组解算误差较大的问题.2次观测中当有2颗卫星重复出现时,平均定位偏差由703.24 m下降到116.63 m,1颗卫星重复出现时,由108.93 m下降到32.17 m,算法运算量增加不多.
A positioning algorithm based on two satellites was studied in assisted Global positioning system （AGPS） in order to realize positioning efficiently in weak signal to capture satellite and to provide auxiliary data more quickly. The positioning solution was used with Marquardt algorithm in which the step factor is used and adaptively adjusts according to the change of objective function. The errors caused in solving the singular equation groups with Gauss-Newton algorithm can be removed using this algorithm. The mean positioning bias can be decreased from 703.24 m to 116.63 m when there exist two overlapped satellites in two measures or from 108.93 m to 32.17 m when only one overlapped satellite, but the computation increases only a little.
出处
《北京邮电大学学报》
EI
CAS
CSCD
北大核心
2009年第2期39-42,61,共5页
Journal of Beijing University of Posts and Telecommunications
基金
国家自然科学基金项目(60772109)
关键词
导航原理
辅助全球定位系统
定位精度
运算量
navigation principle
assisted Global positioning system
positioning accuracy
computation
