利用地面景象信息辅助的无人机自主定位技术

Unmanned Aerial Vehicle Autonomous Positioning Technology Assisted by Ground Scene Information

摘要:

针对无人机自主定位过程中GPS定位系统失效的问题，提出了一种利用地面景象信息辅助的无人机自主定位技术，首先利用无人机所拍摄的实时航拍图像，与预先储存在无人机计算机中地面景象的数字化地形图进行匹配，从匹配结果中提取一个同名像点，结合地面景象数字化地形图所提供的数据信息获取此同名像点的地理位置坐标，根据同名像点位置与无人机位置间的几何关系，结合机载光电测量系统的坐标转换过程，实现无人机的自主定位过程。利用已知的地面同名像点的地理位置信息，反推出无人机的地理位置信息具有一定的创新性。由于整个定位过程存在实际误差，因此利用无人机飞行时记录的数据，采用蒙特卡罗法对定位误差进行仿真试验。试验结果表明该技术能够在误差允许范围内，在GPS定位系统失效的情况下完成无人机的自主定位

Abstract:

To address the problem of failure of GPS positioning system in the autonomous positioning of UAV, this paper presents a Autonomous positioning technology of UAV based on ground scene assistance. Firstly, the real-time aerial images acquired by UAV are matched with those digital topographic map of the ground scene stored in the computer in UAV. Meanwhile extracting a corresponding image points from the matching result, merging the data information in the digital topographic map to obtain the geographical coordinates of the corresponding image points. According to the geometric relationship between the location of the point and the UAV, and combine with the coordinate system conversion process of the onboard photoelectric measurement system, we can realize the autonomous positioning process of UAV. The geographic location information of the corresponding image points on the ground is known to reverse the information of the UAV. This method is the innovation point of this paper. Due to the actual error in the whole positioning process, the positioning error is simulated by Monte Carlo method based on the data recorded in the UAV flight. The experimental results show that the technology can realize the autonomous positioning of UAV when the GPS positioning system fails in the allowable range of error.