基于机器视觉的多旋翼无人机飞行偏航控制

Flight yaw control of multi-rotor UAV based on machine vision

为了降低多旋翼无人机飞行偏航控制的超调量,并提升其抗干扰性能,提出基于机器视觉的多旋翼无人机飞行偏航控制方法。首先在无人机下方安装CCD摄像机,用于采集参考目标图像;然后利用针孔模型的相对位置解算算法,计算多旋翼无人机相对位置;最后通过比对计算位置和预先设定位置;判断无人机是否出现偏航现象。如果存在偏航现象,使用无人机姿态估计算法确定偏航角,并通过抗饱和控制器生成偏航控制量,实现对无人机的偏航控制。实验证明,该方法能够有效采集参考目标图像,获取准确的相对位置信息,具有良好的抗干扰性能。

In order to reduce the overshoot of multi rotor unmanned aerial vehicle flight yaw control and improve its anti-interference performance,a machine vision based multi rotor unmanned aerial vehicle flight yaw control method is proposed.Firstly,it installs a CCD camera below the drone to capture reference target images.Then,the relative position calculation algorithm of the pinhole model is used to calculate the relative position of the multi rotor unmanned aerial vehicle.Finally,by comparing the calculated position with the pre-set position,it determines whether the drone has yaw phenomenon.If there is a yaw phenomenon,the drone attitude estimation algorithm is used to determine the yaw angle,and the anti saturation controller is used to generate yaw control variables to achieve yaw control of the drone.Experiments show that the method can effectively collect reference target images and obtain accurate relative position information,and have good anti-interference performance.