摘要
针对夜间无人机着陆时标准跑道的灯光会加重图像处理的难度问题,提出一种利用跑道灯光辅助着陆的机器视觉方法。该方法参考标准跑道灯光设置,提取跑道灯光直线特征,对EPnP方法进行扩展,将直线特征表示成跑道平面上3个控制点的加权形式,利用奇异值分解的方法求出控制点在相机坐标系中的坐标,最后使用正交化方法解算出无人机相对于跑道的位置和姿态,为无人机着陆提供了引导信息。仿真结果表明:在距机场距离较远时,该方法能有效地引导其自主着陆。
In order to assist landing of unmanned aircraft vehicle （UAV）, a vision-based method is proposed, which makes use of runway lights. Simple configure of runway lights is set up refer to standard runway lighting. EPnP method is extended to line based pose estimation after the lines made up of runway lights are extracted. Each line is expressed as a weighted sum of three control points, and the camera coordinates of the control points can be estimated through singular value decomposition （SVD）. Then, with the orthogonalization method, the position and attitudes of UAV are calculated, to provide the navigation information for UAV landing. Simulation results show that the algorithm proposed is accurate in the earlier stage of landing.
出处
《兵工自动化》
2012年第1期11-13,36,共4页
Ordnance Industry Automation
