摘要
提出了一种基于期望模式修正(EMA)的改进交互式多模型(IMM)算法。该算法主要解决自主水下航行器(AUV)复杂工作环境下量测噪声统计特性未知或易发生变化时的状态估计问题,其核心思想是将期望模式修正机制和交互式多模型滤波算法相结合,利用状态估计过程中的获取的模型概率进行决策,得到更加接近与系统真实模式的期望模型集合,再通过期望模型集合滤波结果对固定模型集合滤波结果进行修正。与传统的交互式多模型算法相比,提出的基于期望模式修正的交互式多模型算法可以捕捉到系统模式更细微的变化。仿真结果表明,该算法可以大幅提高AUV组合导航系统的估计精度和稳定性。
An improved interacting multiple model（IMM） filtering method based on expected-mode augmentation（EMA） is proposed to solve the state estimation problem for measurement noises with unknown or randomly varying statistics properties when an autonomous underwater vehicle（AUV） is in uncertain or tough environment. The proposed adaptive approach combines the expected-mode augmentation methodology and the IMM algorithm. It mainly uses the model probabilities obtained from the IMM recursive estimating process for decision making. Compared with traditional IMM algorithm, the new IMM algorithm based on EMA methodology（EMA-IMM） can capture more subtle changes of the system mode. Simulation results show that the EMA-IMM algorithm can significantly improve the precision and stability of the navigation algorithm.
作者
王磊
程向红
李双喜
高海涛
WANG Lei CHENG Xiang-hong LI Shuang-xi GAO Hai-tao(School of Electrical and Electronic Engineering, Anhui Science and Technology University, Bengbu 233100, China School of Instrument Science and Engineering, Southeast University, Nanjing 210016, China)
出处
《中国惯性技术学报》
EI
CSCD
北大核心
2016年第4期511-516,523,共7页
Journal of Chinese Inertial Technology
基金
国家自然科学基金(61374215)
安徽高校自然科学研究重点项目(KJ2016A169)
安徽科技学院人才稳定项目
关键词
自主水下航行器
组合导航
交互式多模型
期望模式修正
autonomous underwater vehicle
integrated navigation system
interacting multiple model
expected-mode augmentation
