基于改进扩展卡尔曼滤波的AUV单信标定位研究  

Research into AUV Single Beacon Positioning Based on Improved Extended Kalman Filter

作　　者：陈允锋 陶健龙 CHEN Yun-feng;TAO Jian-long(The Ninth Military Representative Office of the Naval Equipment Department in Shanghai,Shanghai 201206,China;Shanghai Jiaotong University,Shanghai 200240,China)

机构地区：[1]海军装备部驻上海地区第九军事代表室,上海201206 [2]上海交通大学,上海200240

出　　处：《舰船电子对抗》2023年第2期72-78,共7页Shipboard Electronic Countermeasure

摘　　要：为提高标准扩展卡尔曼滤波算法在自主水下航行器(AUV)单信标定位应用中的精确度,提出融合遗忘因子和多新息的改进扩展卡尔曼滤波算法。首先,建立单信标导航4状态量-3观测量状态空间方程;然后,在AUV位置解算时,以多新息取代标准扩展卡尔曼滤波(EKF)中的单新息,从利用当前时刻数据转为借助多个历史时刻数据更新。在此基础上,引入遗忘因子削弱历史数据误差累积,防止数据饱和现象,实现AUV位置状态的准确估计。仿真结果表明,改进算法的定位准确性和稳定性显著提升,相比于标准扩展卡尔曼滤波算法,最大定位误差减少了2.352 2 m,定位误差平均值减少了0.856 4 m。In order to improve the accuracy of standard extended Kalman filter algorithm in the single beacon positioning for autonomous underwater vehicle(AUV),an improved extended Kalman filter algorithm combining with forgetting factor and multi-innovation is proposed.Firstly,the state space equation of single beacon navigation 4 state variables-3 observations is established.Then,when the AUV position is solved,the single innovation in the standard extended Kalman filter(EKF)is replaced by multi-innovation,data updates from current moment to multiple historical moments.On this basis,the forgetting factor is introduced to weaken the accumulation of historical data error and prevent from data saturation,so the accurate estimation of AUV position state is achieved.The simulation results show that the positioning accuracy and stability of improved algorithm are significantly improved.Compared with the standard extended Kalman filter algorithm,the maximum positioning error is reduced by 2.3522 m,and the average positioning error is reduced by 0.8564 m.

关 键 词：自主水下航行器 单信标定位 扩展卡尔曼滤波 多新息 

分 类 号：U674.941[交通运输工程—船舶及航道工程] TN971[交通运输工程—船舶与海洋工程]