一种基于贝叶斯估计的地形辅助导航

采用基于蒙特卡罗近似的贝叶斯估计算法处理地形辅助导航系统中的估计问题 ,避免了对复杂地形的线性化过程。数字仿真结果表明该方法比传统的基于扩展卡尔曼滤波的SITAN算法有更高的估计精度 ,并能有效克服初始位置误差较大时 ,基于地形线性化方法易出现的滤波发散问题。

Terrain Aided Navigation Based on Based on Bayesian Estimation

An application of simulation based on Bayesian methods i e.Monte Carlo particle filters to terrain-aided navigation is described. The required posterior distribution of the state variables is represented by a set of random samples, which evolves and adapts recursively in the state space as new measurements become available. In the presented scheme, the integration is approximated by Monte Carlo methods, and no linearization of the nonlinear terrain elevation map is needed.The simulation results show that the presented method overcomes divergence problems associated with the EKF in SITAN, and has better performance of positioning.