基于协同无人机的感兴趣目标定位

Targeting of interest based on cooperative UAV

摘要:

通过无人机拍摄的光学图像求解地面感兴趣目标坐标，被广泛的应用于相关的军事以及商用民用领域。该方法的缺点是使用单架无人机时定位精度不稳定，效率也低。针对这些问题，设计了一种基于协同无人机的感兴趣目标定位系统。通过三架无人机同时观测同一地面感兴趣目标，在对目标定位的过程中，根据透视投影原理，结合无人机获取的姿态数据，通过主机数据处理系统实现机载在线实时定位。为了提高目标定位的精度，提出一种基于内三角形质心算法的协同目标定位方法。采用蒙特卡洛方法进行仿真实验，换算空间定位误差为26.6米，通过数值模拟和飞行实验验证结果表明与传统单机目标定位系统相比，协同无人机感兴趣目标定位系统可以有效利用更多的观测信息，在有效提高目标定位效率的基础上具有更高的交会精度。

Abstract:

Optical images taken by Unmanned Aerial Vehicle (UAV) to gain the coordinates of interested target on the ground, which is widely applied to the field of military and civil. But the disadvantages of this method are unstable positional accuracy and low efficiency when using a single UAV. In response to these problems, an interesting target positioning system based on cooperative UAV is designed. Through the simultaneous observation of the same ground interested target by three UAVs, in the process of target positioning, according to the perspective projection principle, and combined with the attitude data acquired by the UAV, the on-board online real-time positioning is realized by the host data processing system. What's more, in order to improve the positional accuracy, a collaborative target localization method based on inner triangle centroid algorithm is proposed. The Monte Carlo method is used to simulate the experiment. The spatial positioning error is 26.6 meters. The numerical simulation and flight test results show that compared with the traditional single-machine target positioning system, the coordinated UAV target positioning system can effectively utilize more. Observing information has higher accuracy of intersection based on effective improvement of target positioning efficiency.