无人机图像侦察目标定位方法及精度分析

摘要

In order to improve the targeting UAV speed and precision, a reliable image-based reconnaissance drones targeting algorithm is presented. Firstly, the target image information is gained by the photoelectric system of UAV, and then the target’s coordinates in the image and UAV flight parameters are combined to calculate the geodetic coordinates of targets with the use of the homogeneous coordinate transformation method. During the entire localization process, the actual errors are taken into consideration, and the Monte Carlo analysis method is applied to analyze the error according to UAV flight recorder data. Experimental results show that this algorithm can quickly and accurately locate the object, and the positioning accuracy is 12.683195 m, which can illustrate the algorithm in real-time, accuracy, reliability and feasibility.%为了提高无人机目标定位的速度和精度，提出了一种可靠的基于无人机图像侦察的目标定位算法：首先利用无人机光电系统获得实时侦察目标的图像信息，然后把目标在图像上坐标和无人机自身飞行参数相结合，通过坐标变换和几何求解等过程，构建出目标定位方程，计算出目标的大地坐标。在考虑整个定位过程中存在实际误差的情况下，根据无人机飞行时记录的数据，利用蒙特卡罗模拟法进行仿真实验。实验结果表明，此目标定位算法能快速准确地定位目标，定位精度达到了12.683195 m，说明该算法良好的实时性、精确性、可靠性和可行性。