一种GPS/BDS/INS深组合导航系统降维滤波方法

Dimension reduction filtering method for tightly-coupled GPS/BDS/INS

针对全球定位系统/北斗卫星导航系统/惯性导航系统(GPS/BDS/INS)深组合导航系统中卫星数量增加导致计算量迅速增加,难以满足高精度实时定位的问题,该文提出了一种基于星间差分的多星座GPS/BDS/INS深组合导航系统降维滤波方法。首先,通过星间差分消除时钟误差相关状态量,降低状态方程维数;然后,将滤波器分拆成多个子滤波器,通过分散估计、两步级联的方法减少计算量,即通过分散式组合方法(DIM)降低量测方程维数;最后,通过子滤波器的并行运行实现高精度实时定位需求。通过仿真实验对该文算法的定位精度和复杂度进行了分析。结果表明:当GPS和BDS可见星数目均为4时,DIM运行时间相对于集中式组合方法(CIM)减少了约31.7%;当GPS和BDS可见星数目均为5时,DIM运行时间相对于CIM减少了约28.4%。该文算法可以在保持定位精度的同时有效减小计算量。

Aiming at the problem that the increase of the number of satellites in the global positioning system/Beidou navigation satellite system/inertial navigation system(GPS/BDS/INS)tightly-coupled navigation system leads to the rapid increase of calculation,which is difficult to meet the high-precision real-time positioning,a dimension reduction filtering method of multi constellation GPS/BDS/INS tightly-coupled navigation system based on inter satellite difference is proposed here.Firstly,the state variables related to clock error are eliminated by inter satellite difference to reduce the dimension of state equation;then,the filter is divided into several sub filters,and the computational complexity is reduced by decentralized estimation and two-step cascade method,that is,decentralized integration method(DIM),so as to reduce the dimension of measurement equation;finally,high-precision real-time positioning requirements are realized through the parallel operation of sub filters.The positioning accuracy and complexity of the algorithm are analyzed through simulation experiments.The results show that when the number of visible stars of GPS and BDS is 4,the running time of DIM is reduced by about 31.7%compared with the centralized integration method(CIM);when the number of GPS and BDS visible stars is 5,the DIM running time is reduced by about 28.4%compared with CIM.This algorithm can effectively reduce the amount of calculation while maintaining the positioning accuracy.