基于视觉的无人机姿态角估计  

Vision-based estimation of attitude angles of unmanned air vehicles

作　　者：李红[1] 柴洪林[1] 彭嘉雄[2] 

机构地区：[1]华中科技大学数学系 [2]华中科技大学图像识别与人工智能研究所,湖北武汉430074

出　　处：《华中科技大学学报（自然科学版）》2007年第6期4-7,共4页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金资助项目(10477007)

摘　　要：通过提取视频图像上跑道的线性特征以及利用消失点的定义来估计无人机(UAV)的三个姿态角.先对相机的三个姿态角进行估计:利用跑道底线估计出滚动角,用消失点在图像上的位置来估计偏航角,结合消失点的定义给出俯仰角的估计,同时对偏航角和俯仰角的估计进行修正,随后给出无人机的姿态角估计.仿真试验证明了所提方法的有效性,滚动角的估计误差在0.8°之内,偏航角和俯仰角的误差在0.2°之内.In this paper, linear feature of the runway and the vanishing point are used to estimate the attitude angles of the unmanned air vehicle （UAV）. Firstly the attitude angles of the camera are given as follow, the roll is computed from the base line of the runway,the yaw （or the pitch） is gained from the position of the vanishing point in the image, and the last angle is estimated from the definition of the vanishing point, the amendment of the estimation of the yaw and pitch is proposed at the same time. Then the attitude of the UAV is estimated. Experiment shows the validity of the proposed method. the deviation of the roll is less than 0.8°, and that of the yaw and the pitch is less than 0.2°.

关 键 词：无人机 姿态角 消失点 估计 

分 类 号：V249[航空宇航科学与技术—飞行器设计]