摘要
多传感器联合定位中的核心问题是数据融合,本文对基于联合概率密度方法的数据融合进行了研究。首先证明在利用观测分布函数计算联合概率密度时,直角坐标系表示的每一个空间栅格点都对应有目标分布的概率密度值,进而形成了空间概率栅格。针对多目标检测问题,提出利用反函数法逐个检测目标的思想,避免了多目标的相互干扰,提高了算法的通用性。该方法不限传感器的种类或数量,在数据融合领域有广阔的应用空间。
The key problem in target localization with multi-sensor is data fusion and a data fusion method based on joint probability density is discussed in this paper. It shows space probability grid( SPG)of each point in Cartesian coordinates system is formed already when each space grid gets its probability density from probability distribution of target in observe space. Aiming at the problem of multi-targets detection,is proposes a method to detect targets one by one based on inverse function,which avoids mutual interference between multi-targets. As there is no limitation on sensor type or number,SPG algorithm is applicable to different kinds of sensors and can be widely used in data fusion area.
出处
《信息技术》
2017年第8期124-128,共5页
Information Technology
基金
国家863课题(2015AA7046104)
关键词
空间概率栅格
最大似然估计
多传感器
定位方法
space probability grid
maximum likelihood estimation
multi-sensor
localization method