基于博弈论滤波的无人机目标跟踪  

Target tracking control of unmanned aerial vehicle based on game theory filtering

作　　者：苑瑞林[1] 朱志甫[2] 孙华[1] YUAN Ruilin;ZHU Zhifu;SUN Hua(Department of Electronic Information Engineer,Hebi Polytechnic,Hebi 458030,Henan,China;Engineering Research Center of New Energy Technology of Jiangxi Province,East China University of Technology,Nanchang 330013,Jiangxi,China)

机构地区：[1]鹤壁职业技术学院电子信息工程学院,河南鹤壁458030 [2]东华理工大学江西省新能源技术研究中心,江西南昌330013

出　　处：《中国工程机械学报》2019年第5期401-406,共6页Chinese Journal of Construction Machinery

基　　金：河南省科学技术委员会基金资助项目(jcsfA90032)

摘　　要：针对无人机飞行过程中跟踪精度差的问题,提出了一种利用雷达/红外传感器跟踪无人机的博弈论滤波技术.将无人机跟踪问题转化为一种零和动态对策,建立效用函数确定该博弈的平衡点,用于估计目标特性,包括目标的位置和速度.通过最小化测量噪声和转向指令最坏情况下的效用函数,保证了滤波器的鲁棒性.基于仿真证明了在存在模型不确定性、测量噪声和目标未知转向指令的情况下,利用博弈论技术的可行性和高性能.In this paper,a game theory filtering technique using radar/infrared sensor to track unmanned aerial vehicle is proposed to solve the problem of poor tracking precision during unmanned aerial vehicle flight.The unmanned aerial vehicle tracking problem is transformed into a zero sum dynamic game,and a utility function is established to determine the equilibrium point of the game,which is used to estimate the characteristics of the target,including the position and speed of the target.The robustness of the filter is guaranteed by minimizing the utility function of the measurement noise and the steering command worst case.The simulation proves the feasibility and high performance of game theory in the presence of model uncertainty,measurement noise and unknown steering instruction of the target.

关 键 词：雷达 红外 博弈论滤波 零和博弈 无人机 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]