摘要
针对智能体移动方式复杂,对其进行观测的传感器测量的信息存在噪声以及目标运动轨迹发生突然的改变会导致目标观测失真甚至错误的问题,提出了一种变积容积卡尔曼滤波交互多模型算法(VICKF-IMM)。该算法将容积卡尔曼滤波与交互多模型算法相结合,并对容积卡尔曼滤波(CKF)中球面积分进行变积分转换处理。优化了其积分求解的方式,提高了整体的稳定性。Monte-Carlo仿真分析,与CKF-IMM和UKF-IMM算法相比,该算法的跟踪精度有明显的提高,并在目标运动发生突变时有更高的稳定性。
In the process of tracking targets with complex and variable motion patterns during movement,there is noise in the information measured by the sensors that observe them,and sudden changes in the target's motion trajectory can lead to distortion or even errors in the target observation.A variable cubature Kalman filter interacting multiple model algorithm(VICKF-IMM)is proposed.This algorithm combines volumetric Kalman filtering with interactive multiple model algorithm,and performs variable integral transformation on the spherical surface integral in cubature Kalman filtering.The optimization of its integral solution method has improved overall stability.Monte Carlo simulation analysis shows that compared with CKF-IMM and UKF-IMM algorithms,the tracking accuracy of this algorithm is significantly improved,and it has higher stability in case of sudden changes in target motion.
作者
王帅祥
WANG Shuaixiang(College of Mechanical and Electrical Engineering,North University of China,Taiyuan,030051)
