论文部分内容阅读
迄今对各种次优数据关联算法的性能评估通常只是在算法间相互比较的基础上进行的,缺乏各算法绝对次优程度的量化说明。本文对量测不确定性条件下多传感器融合系统状态估计的Cramér-Rao下限进行了研究,对M.L.Hernandez的成果做了进一步推广,给出了量测不确定性条件下具有不同观测噪声统计特征的多传感器融合滤波系统估计CRLB的计算公式。最后对多个参数与稳态CRLB间的关系进行了仿真实验,并指出有关文献中存在的错误。
So far, the performance evaluation of various sub-optimal data association algorithms is usually based on the mutual comparison between algorithms, and there is no quantitative description of the absolute suboptimal level of each algorithm. In this paper, the Cramér-Rao lower bound of the state estimation for multi-sensor fusion systems under measurement uncertainty is studied and the results of MLHernandez are further generalized. The statistical characteristics of different observed noises with uncertainties are given The multi-sensor fusion filtering system estimates the CRLB formula. Finally, the simulation experiments on the relationship between multiple parameters and steady-state CRLB are carried out, and the errors in the literature are pointed out.