成像匹配系统定位结果的TSU数据融合新方法

Novel TSU data fusion method applied to imaging matching position system

在采用SAR末制导导引头的成像匹配定位系统中,现有算法利用加权平均对单次匹配的多模板进行融合得到定位数据。该方法忽视了不同位置上匹配定位信息的继承性和相关性,浪费了很多宝贵信息,限制了定位精度和可靠性的提高。为克服该缺陷,本文采用数据融合思想,在不改变导引头成像及景象匹配算法的前提下,通过挖掘多次定位结果中的关联信息,提出一种时空交错(temporal-spatial-united,TSU)的数据融合算法,利用一次匹配中多模板进行空间融合,及不同时刻定位结果进行时间融合,并通过仿真验证了算法性能,结果表明,本文算法可提高末制导导引头定位精度。

In an SAR imaging matching position system applied to the last-stage guidance of aircraft, the weight average method is used to fuse multi-template matching position data commonly. But the continuity and relativity are ignored among the adjacent different imaging matching position informations, and some invaluable informations are wasted. The current method limits the position precision. In order to resolve the problem, a novel data fusion method called temporal-spatial-united data fusion （TSU-DF） is proposed. The method fuses the associated data which are mined in the position results, in other words, the TSU-DF fuses all the template position data in a single matching position firstly, and then fuses the different time point positions given by the radar imaging matching system. The simulation results show that the proposed method can improve the last-stage guidance position precision.