摘要
针对现有的信息分配方法在滤波精度和自适应性方面存在不足的问题,设计了一种在模糊神经网络基础上改进的自适应联邦滤波器,利用模糊神经网络自适应地调整信息分配系数,从而有效抑制干扰对子滤波器工作状态的影响,使主滤波器能够充分、有效地利用子滤波器的信息。将这种自适应联邦滤波器应用到MINS/GPS组合导航系统设计中,并对系统进行了仿真。通过与基于常规联邦滤波器的组合系统的仿真结果比较可知,自适应联邦滤波器的组合系统在性能上有了显著的提高,且速度误差标准差控制在0.06m/s以内,位置误差标准差控制在3.3m以内,具有良好的导航精度。
Aiming at the deficiency of existing information sharing methods in precision and adaptability, a self-adaptive federated filter based on Fuzzy Neural Network was designed, with the information sharing coefficient adjusted by Fuzzy Neural Network, which can restrain the disturbance on sub-filter and make master-filter use information adequately. The self-adaptive federated filter was applied in the MINS/GPS integrated navigation system. The simulation was made which show that, compared to the general federated filter, the system based on self-adaptive federated filter gains remarkable improvement in navigation precision, in which the standard deviation of velocity error is within 0.06 m/s and the standard deviation of position error is within 3.3 m.
出处
《中国惯性技术学报》
EI
CSCD
2007年第6期678-681,共4页
Journal of Chinese Inertial Technology
基金
国家863高技术项目(2002AA812038)
