论文部分内容阅读
提出了一种多传感器组合导航系统的分步滤波算法。当所有传感器的观测值到来时,首先对该时刻的系统状态进行预测,然后利用常规卡尔曼滤波器和各导航传感器的观测值依次对该时刻的状态向量估计值进行更新,进而得到该时刻状态向量基于全局信息的最优融合估计。最后利用GPS/SST/高度表/SINS多组合导航系统对上述算法进行验证。仿真结果表明,该算法与集中式卡尔曼滤波算法的估计精度相同,但计算量得到降低。
A multi-sensor integrated navigation system step filtering algorithm is proposed. When the observed values of all the sensors come, the state of the system at the moment is predicted first, and then the estimated value of the state vector at that moment is updated sequentially with the observations of the conventional Kalman filter and each navigation sensor to obtain the state at that moment Vector Optimal Fusion Estimation Based on Global Information. Finally, the above algorithm is validated by GPS / SST / Altimeter / SINS multi-combination navigation system. The simulation results show that the proposed algorithm has the same estimation accuracy as the centralized Kalman filter, but the computational cost is reduced.