摘要
针对常规卡尔曼滤波在组合导航中容错性不足的问题,提出了一种基于遗传模糊推理的自适应容错滤波算法。首先建立了基于模糊推理的自适应滤波模型,利用模糊推理系统的输出对组合导航系统的量测噪声实时进行调整,以实现状态的精确估计,进而达到容错目的。接着利用自适应遗传算法对模糊推理系统的隶属度函数参数进行了优化,提高了系统的输出精度,改进了传统模糊建模中系统精度取决于专家知识是否完备的问题。最后以SINS/GPS组合导航系统为平台进行了仿真,并在系统工作中间时刻引入量测噪声故障。验证结果表明遗传模糊推理自适应滤波算法比常规卡尔曼滤波具有更强的容错能力和总体精度,在仿真中,平均位置和速度均方根误差分别降低了20.87%和41.94%。
Aiming at the shortage of standard Kalman filter for fault tolerant in integrated navigation,an adaptive filter method based on fuzzy inference system（FIS） optimized by genetic algorithm is proposed.Firstly,the adaptive filter model based on FIS is built,which adjusts really the measure noise so as to estimate the system status accurately and come to the goal of adaptive fault tolerant.Secondly,the adaptive genetic algorithm is applied to optimize the parameters of membership function so as to improve the output precision of FIS and overcome the problem of output precision depending on the expert knowledge maturity.Finally,the simulation based on SINS/GPS integration navigation system is demonstrated,which introduces the measure noise fault in the middle work time.The simulation results indicate that the proposed method has stronger fault tolerant ability and general precision,which shows that mean root mean square errors of position and velocity has decreased 20.87% and 41.94%,respectively.
出处
《中国惯性技术学报》
EI
CSCD
北大核心
2012年第3期315-319,共5页
Journal of Chinese Inertial Technology
基金
国家高技术研究发展计划(863计划)(2009AA7050411)
关键词
组合导航
卡尔曼滤波
模糊推理
遗传算法
自适应容错
integrated navigation
Kalman filter
fuzzy inference
genetic algorithm
adaptive fault tolerant
