基于伪距残差相关性的GPS/INS紧耦合抗差滤波

GPS/INS tightly-coupled robust filter based on residual correlation of pseudo-range

提出了基于GPS伪距残差相关性的GPS/INS紧耦合抗差滤波算法.理论推导了GPS伪距观测值残差的计算方法,引入了相关函数对伪距残差进行相关性分析;分析了GPS/INS紧耦合导航的动力学模型和观测模型,基于残差相关性构建了紧耦合抗差滤波算法,利用实测数据验证了算法的有效性.结果表明:相比于非抗差滤波算法和传统的抗差滤波算法,基于GPS伪距残差相关性的抗差滤波能更灵敏地探测GPS/INS紧耦合中伪距观测值的粗差,并且通过抗差滤波因子抑制粗差对于导航结果的影响,提高GPS/INS紧耦合导航精度,而且通过控制粗差能够有效避免滤波发散现象的发生.

In GPS/INS tightly-coupled navigation, gross error in GPS pseudo-range observation limits the improvement of navigation precision. Based on correlation theory, a GPS/INS tight- ly-coupled robust filter based on residual correlation of GPS pseudo-range is proposed. First the computing method of GPS pseudo-range residual is deduced in theory and correlation function is introduced to analyze the correction of pseudo-range residual. The design formulas of observation model and dynamical model are presented, the tightly-coupled robust filter algorithm is constructed on the basis of residual correlation. Finally an actual calculation was performed to test the validity of the new algorithm. The results of the experiment indicate that, compared with classic Kalman filter and traditional robust filter, robust filter based on residual correlation of GPS pseudo-range could more sensitively detect gross error in GPS pseudo-range observation of GPS/INS tightly-coupled navigation and control the influence of gross error on navigation result by the robust filter factor, which can improve tightly-coupled navigation precision. When the gross error is controlled effectively, filtering divergence will be prevented.