Underwater vehicle terrain navigation based on Maximum Likelihood Estimation

摘要：

Terrain aided navigation (TAN)is a way to improve underwater vehicle self navigation ability,which is essentially a nonlinear state estimation problem.The iterative Bayesian method based on maximum a posteriori (MAP) estimation has theoretic advantage in solving such problems.But due to the particularities of underwater vehicles,in some cases the long time continual filter is not suitable for underwater vehicles,where MAP estimation degenerates to maximum likelihood (ML)estimation,and continual filter changes to discontinuous positioning.By analyzing the shape of likelihood function,we found that the influence of false peaks in terrain matching will decrease with the increase of measuring beams, which proves the validity of ML estimation as a positioning method.Then a practical method of obtaining the measuring beams is proposed.Simulations validate these conclusions.

关键词： maximum likelihood estimation iterative Bayesian method underwater terrain navigation

作者: TIAN Feng-min XU Ding-jie ZHAO Yu-xin LI Ning

作者单位: College of Automation Harbin Engineering University Harbin,Heilongjiang Province,150001,China

会议类型: 国际会议

会议名称: 2008 IEEE International Conference on Onformation and Automation(IEEE 信息与自动化国际会议)

会议地点: 张家界

会议语种:英文

页码: 1268-1273

在线出版日期: 2008-06-20（万方平台首次上网日期，不代表论文的发表时间）