| 摘要: |
| 智能导弹等智能化飞行器在快速跨域、高速机动飞行时，由于传感器切换、外形改变等因素，会对组合导航信息融合系统引入随机非Gauss噪声等影响，离线优化的参数往往不能满足滤波器精度的需求。自适应网络模糊推理系统ANFIS是一种将人工神经网络和模糊推理技术相结合而成,符合人类认知特点的决策方法，它可以对导航数据进行学习，实现智能决策、实时修改滤波器内部参数，对滤波器进行优化。仿真结果表明，基于ANFIS优化的智能导航自适应滤波算法可以有效减少噪声和干扰带来的影响，提高导航精度。 |
| 关键词: 智能导航 卡尔曼滤波 神经网络 模糊推理 自适应网络模糊推理技术 |
| DOI： |
|
| 基金项目:载人航天第四批预研项目(060201) |
|
| Method of Intelligent Navigation Adaptive Kalman Filter based on ANFIS |
| WU Peng,MU Rongjun,DENG Yanpeng,SUN Xuyao |
| (School of Aeronautics, Harbin Institude of Technology) |
| Abstract: |
| In order to achieve optimal filtering results, traditional data fusion filter requires that the parameters of the system are precisely known. Intelligent missile is often unsteady, strongly coupled and nonlinear when it is in the rapid cross domain flight. For this application background with complex dynamic characteristics, the parameters of offline optimization usually cannot meet the requirement of filter accuracy. The ANFIS （adaptive network-based fuzzy inference system） can be used to modify the related parameters of the filter in real-time and optimize it. The simulation result shows that the provided method improves the performance of the system to a certain extent, so that this method is feasible and effective. |
| Key words: Intelligent navigation Kalman filter Neural network Fuzzy inference ANFIS |