单天线GPS/INS经典组合模型存在的明显不足是：由于状态参数不仅包含INS的位置、速度误差,而且还包含INS的姿态误差和元件误差,存在观测信息不足,间接可测参数难以准确估计。在深入分析常用GPS/INS组合模型及滤波算法中参数估值的特性基础上,提出从两方面改善这类滤波结果：①在运动条件下的组合观测模型中附加GPS单天线测方位角约束,建立观测值与间接可测参数之间的联系;②将选权自适应卡尔曼滤波方法拓展到GPS/INS滤波中,以提高参数估计的精确性。通过对滤波结果的增益阵、位置差、速度差及姿态估计结果的分析%The main problem of the classical filtering based on the single antenna GPS/INS integrated system is difficult to estimate the indirect observable parameters accurately because the observation information is insufficient and the state model can not fit a complex kinematic environment.Through analyzing the character of the parameter estimators using the classical GPS/INS coupled model and Kalman filtering,a new method is put forward for improving the filter results.① On the motion condition,the heading angle from the single antenna GPS is added to the observation equations of the Kalman filtering,so the relationship between the observations and the indirect observable parameters is bui② The adaptive Kalman filtering by selecting the parameter weights is extended to the GPS/INS integrated system.Through analyzing the gain matrix,position and velocity differences and attitude error estimators,the results of the new method are more accurate and reasonable.
展开▼
