基于联邦粒子滤波器的多传感器信息融合算法  

Multi-sensor Information Fusion Based on Federate Particle Filter

作　　者：宁小磊[1] 李文博[1] 陈连华[2] 

机构地区：[1]中国华阴兵器试验中心,陕西华阴714200 [2]第二炮兵工程学院,西安710025

出　　处：《弹箭与制导学报》2011年第4期45-48,52,共5页Journal of Projectiles,Rockets,Missiles and Guidance

摘　　要：提出了一种新的SINS/CNS/GPS组合导航观测方程,并给SINS/CNS/GPS组合导航系统设计了一种新的联邦式数据融合结构。使用粒子滤波器对联邦滤波子系统的多源数据进行融合处理,从而摆脱了经典卡尔曼滤波器的限制。仿真结果表明所提算法在SINS/CNS/GPS组合导航系统的信息融合中是有效性的。A new measurement equation of SINS/CNS/GPS integrated navigation was established,and also a novel federate filter（federated particle filter,KPF） was proposed to estimate the state of non-Gaussian and non-Linear SINS/CNS/GPS system,in which particle filer was introduced to federate filter so that the information fusion of subsystem can be solved by the particle filer.By doing so,the federate filter can get rid of the disadvantage of ordinary Kalman filter.The simulation results of SINS/CNS/GPS navigation system demonstrate the feasibility of the proposed algorithm.

关 键 词：SINS/CNS/GPS 信息融合 联邦滤波 粒子滤波 组合导航 

分 类 号：V249.328[航空宇航科学与技术—飞行器设计]