面向直升机的地形概率密度特征匹配方法

Terrain Matching Algorithm Based on Probability Density Feature for Helicopter

随着现代战争的发展,争夺低空、超低空制空权具有重要意义,地形辅助导航是实现直升机等低空飞行器导航定位的有效手段。针对直升机低空飞行场景下地形辅助导航中的地形匹配受量测误差影响大导致匹配错误进而影响定位精度的问题,通过对传统匹配相关运算方法进行分析,引入地形高程差的概率密度作为相关性计算方法,提出了一种利用地形高程差的概率密度特征进行地形匹配的方法,并引入地形复杂度与相似度对地形匹配置信度进行评估,减小错误匹配对导航误差修正的影响。仿真实验结果表明,所提出的基于概率密度特征的地形匹配方法对量测误差具有更高的适应性,匹配准确率更高,相对于传统方法,在测量误差较大的情况下可以提高56.5%的匹配准确率。通过对地形匹配置信度进行评估,在误匹配次数较多的场景下可以将组合导航定位误差的均方根误差降低为传统方法的3.54%,有效提高了地形辅助导航系统的容错性。

With the development of modern war,it is of great significance to fight for the right to control low and ultra-low altitudes.Terrain-aided navigation is an effective means to achieve the navigation and positioning of low-altitude aircraft such as helicopters.Aiming at the problem that terrain matching in terrain-aided navigation under helicopter low altitude flight scene is greatly affected by measurement errors,which leads to matching errors and affects positioning accuracy,through analyzing the traditional matching correlation operation methods,the probability density of terrain elevation difference is introduced as the correlation calculation method,and a terrain matching method using the probability density features of terrain elevation difference is proposed.The terrain complexity and similarity are introduced to evaluate the confidence of terrain matching to reduce the impact of error matching on navigation error correction.The simulation results show that the proposed terrain matching method based on probability density features has higher adaptability to measurement errors and higher matching accuracy.Compared with traditional methods,the matching accuracy can be improved by 56.5%in the case of large measurement errors.By evaluating the confidence of terrain matching,the root mean square error of integrated navigation positioning error can be reduced to 3.54%of the traditional method in the scene with many mismatches,effectively improving the fault tolerance of terrain-aided navigation system.