摘要
无人机自主着舰视觉引导应用距离跨度大、测量精度要求高,尚缺乏有效的解决方法,针对此问题,提出一种基于多层级合作标识的无人机自主着舰视觉引导方法。该方法首先设计了基于ArUco的多层级组合新型合作标识,利用边缘检测算法对每一层ArUco二维码提取控制点,进一步通过求解PnP(Perspective-n-Point)问题得到机舰间相对位姿信息,然后对每一层二维码所解算出的位姿数据进行非线性误差补偿,利用最小二乘法(Least Mean Squares,LMS)与熵权法解算不同层级二维码的权值大小,最终通过融合不同层级二维码位姿解算结果得到最终的位姿信息,实现大距离跨度范围内的高精度位姿求解。实验结果表明,本文方法实现了大距离跨度范围内机舰间相对位姿参数的高精度测量,可为无人机自主着舰提供可靠引导信息。
The application of visual guidance for autonomous landing of UAV has a large distance span and high measurement accuracy requirements,but there is still a lack of effective solutions.To solve this problem,a visual guidance method for autonomous landing of UAV based on multi-level cooperative identification is proposed.The approach starts with the design of a new cooperative identification based on multi-level combination of ArUco.Then,the control points are extracted from each ArUco two-dimensional code by using an edge detection algorithm.Further,the relative pose information between aircraft and ships is obtained by solving the PnP(Perspective-n-Point)problem.Finally,the nonlinear error compensation is carried out for each codes solved results,the weight of different levels of ArUco codes is solved by the least mean squares(LMS)algorithm and entropy weight method,and the final pose solution result is obtained by fusing the multi-level solution results to realize the high-precision pose solution in a large distance span.The experimental results show that this method realizes the high-precision measurement of the relative pose parameters between aircraft and ships in a large distance span,which can provide reliable guidance information for autonomous landing of UAVs.
作者
丁浩峰
雷清雲
段欣葵
周颉鑫
孙晓亮
DING Haofeng;LEI Qingyun;DUAN Xinkui;ZHOU Jiexin;SUN Xiaoliang(College of Aerospace Science and Engineering,National University of Defense Technology,Changsha 410072,China)
出处
《测控技术》
2023年第8期7-14,共8页
Measurement & Control Technology
基金
大学生创新创业训练计划项目(202090002162)。
