改进的强跟踪求积分卡尔曼滤波算法-《计算机技术与发展》

文章信息/Info

Author(s):: HE Shan; ZHAO Xu; SHI Xi; School of Computer Science,Xi’an Polytechnic University,Xi’an 710048,China

Keywords:: nonlinear state estimation; strong tracking quadrature Kalman filter; fading factor; softening factor

摘要:: 在非线性系统状态估计问题中,强跟踪求积分卡尔曼滤波算法在实现过程中由于对判断滤波发散的阈值设置较小,即使在系统正常情况下也会以较大概率产生渐消因子,从而导致过度调节滤波增益,使得系统状态估计不够平滑。针对该问题,提出了一种改进的强跟踪求积分卡尔曼滤波算法。该算法通过适当增大判断滤波发散的阈值,从而有效地降低了误判滤波发散的概率,增强了滤波器对系统状态的跟踪性能,并能够根据不同维数的量测方程确定弱化因子的取值,从而有效避免了凭经验选取弱化因子,具有较强的操作性。对两种算法进行实验仿真,结果表明,改进的强跟踪求积分卡
尔曼滤波算法具有更高的滤波精度,减小了系统状态估计值与真实值之间的偏差。

Abstract:: In the nonlinear state estimation,strong tracking quadrature Kalman filtering algorithm sets small threshold to judge filtering divergence leading to fading factor with high probability even when the system is normal,which causes excessive regulation of the filtering gain and makes the state estimation curve lack smoothness. For this,we present an improved strong tracking quadrature Kalman filtering algorithm. The algorithm reduces probability of misjudging filter divergence by appropriately increasing the threshold,and also can enhance the tracking performance of the filter. Thereby,it can determine the softening factor by the different dimensions of measurement equation,and thus avoids the disadvantages of the previous algorithm that determines the softening factor in accordance with experiences.It is more operable and the simulation test is carried,which shows that the improved strong tracking quadrature Kalman filtering algorithm can obtain higher filtering accuracy than the strong tracking quadrature Kalman filtering algorithm,and reduces the deviation between the system state estimation value and the real value.