基于自适应鲁棒滤波的SINS/DVL动基座初始对准方法  

Initial alignment method for SINS/DVL based on adaptive robust filtering on moving base

作　　者：肖强 文笃石 段明磊 杨童 张成林 XIAO Qiang;WEN Dushi;DUAN Minglei;YANG Tong;ZHANG Chenglin(Yunnan Highway Network Toll Management Ltd.,Kunming 650100,China;School of Computer Science&Technology,Xi’an University of Posts&Telecommunications,Xi’an 710000,China)

机构地区：[1]云南公路联网收费管理有限公司,云南昆明650100 [2]西安邮电大学计算机学院,陕西西安710000

出　　处：《中国测试》2022年第5期116-122,共7页China Measurement & Test

基　　金：陕西省教育厅科研计划(15JK1679);陕西省科技统筹创新工程计划项目(2015KTCQ01-14)。

摘　　要：针对传统的基于矢量重构的鲁棒粗对准方法对DVL量测噪声敏感的问题,提出基于自适应鲁棒滤波的动基座鲁棒粗对准方法,实现动态跟踪DVL量测噪声,提高粗对准精度。首先建立姿态确定模型和观测矢量误差模型;然后分析矢量重构模型,提出基于自适应鲁棒滤波的参数估计技术,动态跟踪速度噪声,优化参数估计过程,提高矢量重构精度;接着通过最优基算法获得最优姿态四元数;最后进行仿真实验验证。实验表明,该方法较传统方法鲁棒性更强,收敛速度更快,600 s对准结束时刻准确度提高79%。Aiming at the problem that the traditional robust rough alignment method based on vector reconstruction is sensitive to DVL measurement noise,a robust rough alignment method based on adaptive robust filtering is proposed to dynamically track DVL measurement noise and improve the rough alignment accuracy.Firstly,attitude determination model and observation vector error model are established.Secondly,the vector reconstruction model is analyzed,and a parameter estimation technique based on adaptive robust filtering is proposed to dynamically track the speed noise,optimize the parameter estimation process,and improve the vector reconstruction accuracy.Then the optimal attitude quaternion is obtained by the optimal basis algorithm.Finally,a simulation experiment is carried out to verify.Experimental results show that the proposed method is more robust and converging faster than the traditional method,and the precision at the end of alignment at 600 s is increased by 79%.

关 键 词：动基座初始对准 DVL辅助 自适应噪声 鲁棒对准方法 

分 类 号：U666.1[交通运输工程—船舶及航道工程] TB9[交通运输工程—船舶与海洋工程]