摘要
为了解决实际线性系统中系统噪声方差和观测噪声方差未知的问题,提出了一种新的卡尔曼滤波自适应算法,利用新息序列的方差,可以在系统的自身计算过程中逐步估计并校正系统噪声方差和观测噪声方差.系统模拟显示,估计的系统噪声方差和观测噪声方差均收敛于实际的系统噪声方差和观测噪声方差,而且收敛速度比传统卡尔曼滤波要快.
In order to solve unknown system noise variance and observation noise variance in practical linear system, a new self-adaptive algorithm for Kalman filter was presented,it made use of the variance of new sequences, which could estimate and correct the system noise variance and the observation noise variance gradually in calculation process. The computer simulation shows that the estimated variance of the system noise and estimated variance of the observation noise converge to the real values more quickly than the traditional one.
出处
《南京工业大学学报（自然科学版）》
CAS
2007年第4期30-33,共4页
Journal of Nanjing Tech University(Natural Science Edition)
基金
江苏省高校自然科学基金资助项目(03KJB110037)
关键词
卡尔曼滤波
白噪声
自适应
Kalman filter
white noise
self-adaptive
