MS-KF融合算法用于锥套跟踪

针对无人机自主空中加油过程中锥套跟踪，提出一种均值漂移-卡尔曼滤波(mean shift-Kalman filter, MS-KF) 融合算法。分析了基于均值漂移算法的锥套目标模型、相似性度量、锥套目标定位的锥套定位原理；引入卡尔曼滤波器对锥套运动状态进行预测，将锥套运动信息融合到均值漂移算法中，以保证锥套跟踪算法的稳定性和鲁棒性；给出了MS-KF融合算法用于锥套识别跟踪的流程；搭建了锥套跟踪半物理实验验证系统，分别进行MS-KF融合算法用于锥套跟踪的半物理实验验证及数值仿真分析。实验结果表明:MS-KF融合算法可以对锥套精确定位跟踪，无人机3个轴向的跟踪误差保持在0.3 m的范围内，保证了无人机自主空中加油的顺利进行。

MS-KF fusion algorithm for drogue tracking

A mean shift-Kalman filter (MS-KF) fusion algorithm for drogue tracking in unmanned aerial vehicle (UAV) autonomous aerial refueling (AAR) was proposed. The drogue tracking theory based on mean shift (MS) algorithm, including drogue aim model, similarity measurement and drogue positioning, were analyzed. By introducing Kalman filter (KF), the drogue moving information was predicted and merged in MS algorithm in order to ensure the stability and robustness of the drogue tracking algorithm. The flow block diagram of drogue tracking based on MS-KF fusion algorithm was provided and the semi physical experimental system of drogue tracking was constructed. The semi-physical verification experiment and numerical simulation analysis of MS-KF algorithm for drogue tracking were respectively conducted. Experimental results demonstrate that the MS-KF fusion algorithm can position and track drogue precisely and the tracking errors of the UAV axes keep within the range of 0.3m, which ensures the success of the UAV-AAR.