摘要
针对移动目标跟踪过程中,传感器感知信息存在噪声以及运动轨迹突变导致目标观测失真甚至丢失的问题,提出了一种扩展卡尔曼滤波交互多模型算法(EKF-IMM)。该算法以交互多模型算法为主体,同时融入EKF算法做滤波处理,使得在目标跟踪过程中,不仅对目标的不同运动状态具有自适应能力,同时还能对运动状态中可能的非线性问题做更好的处理,提高算法的鲁棒性。仿真实验表明,EKF-IMM算法能很好得适应多变的目标运动,与标准KF-IMM算法相比,该算法降低了噪声对传感器的干扰,提高了定位精度。
Aiming at the problem that interferences of noise on the sensor information and motion trajectory of target observation mutations lead to distortion or even loss in the process of target tracking,proposes an extended Kalman filter interacting multiple model algorithm(EKF-IMM).The algorithm takes the interacting multiple model algorithm as the main part and integrates the EKF algorithm into the filtering process.In the process of target tracking,not only has the adaptive ability to different moving states of the target,but also can better deal with the possible nonlinear problems in the motion states,and improve the robustness of the algorithm.Simulation results show that the EKF-IMM algorithm can well adapt to changing target motion.Compared with the standard KF-IMM algorithm,this algorithm can reduce the noise interference and improve the positioning accuracy.
作者
高春艳
董鹏
高涵
王辉强
GAO Chun-yan;DONG Peng;GAO Han;WANG Hui-qiang(School of Mechanical Engineering,Hebei University of Technology,Tianjin 300130,China)
出处
《机械设计与制造》
北大核心
2020年第2期284-287,共4页
Machinery Design & Manufacture
基金
国家高技术研究发展计划（863计划）（No.2015AA043101)
关键词
目标跟踪
噪声
轨迹突变
扩展卡尔曼滤波
交互多模型
定位精度
Target Tracking
Noise
Motion Trajectory
Extended Kalman Filter
Interacting Multiple Model
Loca-tion Accuracy
