摘要
计算误差和模型误差是导致容积卡尔曼滤波(CKF)发散的两大主要因素,而平方根容积卡尔曼滤波算法(SCKF)只能抑制计算发散。为抑制模型误差引起的滤波发散,通过引进指数加权衰减因子实现对当前测量数据利用的加强,提出了衰减记忆平方根容积卡尔曼滤波(MASCKF)算法。并将该算法应用到光电跟踪系统的目标跟踪滤波预测中,仿真实验结果表明,MASCKF算法能有效抑制滤波发散,滤波效果优于SCKF算法。
Calculation error and model error are two main factors which cause the divergence of cubature Kalman filter(CKF),and square-root cubature Kalman filter(SCKF)algorithm can only restrain the calculation divergence.To restrain the filtering divergence caused by model error,the square-root cubature Kalman filtering algorithm based on memory attenuation(MASCKF)is proposed.In which,a weighted index is introduced as the attenuation factor to enhance the use of the current measurement data.Finally,the algorithm proposed is applied to target tracking filtering of optical tracking system.The simulation results show that,MASCKF algorithm can effectively suppress the filtering divergence caused by model error,and the effect of filtering is better than SCKF algorithm.
作者
杨旺明
YANG Wangming(China Three Gorges University,Yichang 443002)
出处
《计算机与数字工程》
2018年第3期471-474,共4页
Computer & Digital Engineering
关键词
平方根容积卡尔曼滤波
衰减记忆
光电跟踪
目标跟踪
square-root cubature Kalman filtering,memory attenuation,photoelectric tracking,target tracking
