摘要
针对巡飞弹弹载微机电系统(MEMS)在长航时过程中航向易发散的问题,提出了一种基于景象匹配的低精度MEMS航向修正与组合算法。首先,基于匹配点几何关系与匹配重投影设计了两种航向误差模型,前者分别根据实时图与基准图匹配点连线与图像纵轴计算反正切值,作差后计算出航向修正值;后者利用实时图与基准图的匹配关系构建重投影误差,通过优化计算出航向最优值。然后,基于上述两种模型,根据匹配点数量设计了一种航向修正与组合算法。飞行实验结果表明,长航时过程中修正后的航向误差为0.7423°(RMSE),更新频率优于1 Hz。与传统仅能定位的景象匹配算法相比,所提算法提供了一种航向修正方法,且对传感器精度依赖性较低,具备一定的通用性。
Aiming at the problem that the heading angle of micro-electro-mechanical systems(MEMS)carried by patrol projectiles is easily scattered during long endurance,a heading correction and combination algorithm for MEMS with low accuracy based on scene matching is proposed.Firstly,two kinds of heading error models are designed based on the geometric relation of matching points and matching reprojection.The former calculates the arctangent value according to the matching point line of the real-time image and the reference image and the vertical axis of the image respectively,and calculates the heading correction value after making the difference;the latter uses the matching relation between the real-time graph and the reference graph to construct the reprojection error,and the optimal heading value can be calculated through optimization.Then,based on the above two models,a heading correction and combination algorithm is designed according to the number of matching points.The flight test results show that the corrected heading angle error is 0.7423°(RMSE)and the updating frequency is better than 1 Hz during long endurance.Compared with the traditional scene matching algorithm that can only locate,the proposed algorithm provides a heading correction method with low dependence on the accuracy of the sensor and has certain versatility.
作者
朱启举
樊振辉
梅春波
杨鹏翔
卢宝峰
杨朝明
ZHU Qiju;FAN Zhenhui;MEI Chunbo;YANG Pengxiang;LU Baofeng;YANG Chaoming(Xi’an Modern Control Technology Research Institute,Xi’an 710065,China;Electronic Information School,Northwestern Polytechnical University,Xi’an 710072,China)
出处
《中国惯性技术学报》
EI
CSCD
北大核心
2023年第9期870-875,共6页
Journal of Chinese Inertial Technology
基金
国防基础科学研究项目(2019-JCJQ-2D-078)。
