基于高阶球面-径向准则的多传感器融合算法  

Multi-sensor Fusion Algorithm Based on High-degree Spherical-radial Rule

作　　者：余战秋 YU Zhanqiu(Department of Computer Science and Technology,Anhui Technical College of Industry and Economy,Hefei 230000,China)

机构地区：[1]安徽工业经济职业技术学院计算机科学技术系,安徽合肥230000

出　　处：《探测与控制学报》2018年第5期26-30,34,共6页Journal of Detection & Control

基　　金：安徽省自然科学研究重点项目资助(KJ2017A645)

摘　　要：针对目前多传感器信息融合精度不高的问题,提出了基于高阶球面-径向准则的多传感器融合算法。该算法将七阶球面-径向准则嵌入信息滤波框架,提出能够处理集中式多传感融合的集中式七阶容积信息滤波(C7thCIF)算法;然后,采用平均一致性算法对多传感器信息进行处理,采用C7thCIF对平均一致值进行滤波,推导了能够处理分布式多传感器融合的分布式七阶容积信息滤波(D7thCIF)算法,提高了集中式和分布式多传感器融合的精度。仿真结果表明,高阶算法能够获取比低阶算法更好的融合精度,但增加了计算复杂度。Aiming to the problem that the precision of multi-sensor information fusion is not high(need further improvement),a multi-sensor fusion algorithms based on high-degree spherical-radial rule were proposed.The centralized seventh-degree cubature information filter(C7thCIF),which ccould deal with centralized multi-sensor fusion problem,was proposed through introducing the seventh-degree spherical-radial into the frame of extended information filter(EIF).Then,average consistency algorithm was used to dispose multi-sensor information,the distributed seventh-degree cubature information filter(D7thCIF),which could deal with distributed multi-sensor fusion problem,was proposed by using C7thCIF to filter average consensus value.The dealing ability of centralized and distributed multi-sensor information fusion was enhanced.The simulation results showed that the high-degree algorithms had higher fusion precision than the low-degree algorithms,but the calculation complexity was increased.

关 键 词：平均一致性算法 球面径向准则 集中式多传感器 分布式多传感器 

分 类 号：TP953[自动化与计算机技术]