摘要
主被动传感器融合定位具有高度非线性,针对采用传统的线性化方法计算变量统计特性,理论定位结果与实际定位结果相差较大的情况,文章提出了基于不敏变换(UT)的数据压缩(DC)融合定位算法。首先,通过不敏变换精确计算了二维变量的统计特性,减小了非线性误差的影响;其次,针对数据压缩过程中量测信息重复利用的问题进行去相关性处理,获得较高的定位精度;最后,通过理论分析和仿真结果验证,相较于传统的线性化处理方式相差较大的情况,基于不敏变换的数据压缩融合定位方法理论结果与实际结果相吻合。
Location fusion of the measurements of the active and passive sensors is highly nonlinear. The commonly used linear method, however, may lead to worse effect. Therefore, a fusion localization method based on the unscented transformation （UT） and data compressing （DC） was proposed in this paper. Firstly, UT was taken to calculate the statistic characteristics of the two-dimensional variable, so as to eliminate the errors brought by nonlinear transformation. Secondly, since the angle measurement of the active sensor was also used in the triangulation location, the relativity between the active sensor location result and the triangulation location result was considered to improve the location precision. Theoretical analysis and simulation results showed that the location performance of the present method in this paper was superior to that of the traditional linear processing method.
出处
《海军航空工程学院学报》
2015年第5期401-408,共8页
Journal of Naval Aeronautical and Astronautical University
基金
国家自然科学基金资助项目(61372027
61102165)
关键词
主动传感器
被动传感器
非线性
不敏变换
相关性
数据压缩
active passive
passive sensor
nonlinearity
unscented transformation
relativity
data compressing
