目标跟踪系统自校正解耦融合Wiener平滑器

摘要

For the multisensor target tracking systems with unknown model parameters and noise, in order to solve the smoothing problem of the signals, the paper first obtained respectively the local estimates of the model parameters and the noise variances by the system identification algorithm and the related method. And the fused estimates with high reliability can be obtained by taking the average of the local estimates. Then the self-tuning decoupled fusion Wiener smoother was obtained by substituting the on-line fused estimates into the optimal decoupled fusion Wiener smoother which is based on the modern time series analysis method. The self-tuning fused Wiener smoother converges to the optimal fused Wiener smoother, and is of asymptotic optimality. So it was proved that this self-tuning smoother can solve the smoothing problem well for the system with unknown model parameters and noise variances. At last, a simulation example using Matlab software shows the effectiveness of the self-tuning decoupled fusion Wiener smoot-her.%针对带未知模型参数和噪声的多传感器目标跟踪系统,为了解决信号的平滑问题,分别利用系统辨识及相关方法得到未知模型参数和噪声方差的局部估值,并对这些局部估值求平均值作为它们的融合估值.然后将具有高可靠性的在线融合估值代入到基于现代时间序列的最优解耦融合Wiener平滑器中即可得自校正解耦融合,使自校正融合Wiener平滑器收敛于相应的最优融合Wiener平滑器,并具有渐近最优性.从而证明自校正平滑器能够很好地解决未知模型参数和噪声统计系统的平滑问题.最后利用Matlab软件仿真验证了该自校正解耦融合Wiener平滑器算法的有效性.