| The oceans account for 71 percent of the earth’s surface area and are rich in natural resources.Therefore,ocean exploration is of great significance.Underwater vehicles are important tools for ocean exploration.Although the SINS/DVL integrated navigation system does not have the disadvantage that the satellite navigation system cannot receive signals underwater,navigation errors exist in the SINS/DVL integrated navigation system due to the installation of SINS/DVL and the complicated and changeable underwater environment.This paper takes this as the research background,fully considers the SINS/DVL integrated navigation system installation angle,lever error and ocean current influence,applies Variable Structure Multiple Model Algorithm,deeply studies the SINS/DVL integrated navigation technology.Firstly,the common reference frames of navigation system,the fundamental principles of SINS and DVL are briefly introduced.The differential equation of the SINS is derived,and the error equation of the SINS is further derived.The error source of SINS and DVL velocity measurement error are analyzed,and models of the inertial sensor and DVL errors are built to lay a foundation for the research of SINS/DVL integrated navigation technology.Secondly,this paper expounds the Kalman filter theory,deduces the standard Kalman filter equation,and chooses the EKF as the basic filtering method of SINS/DVL integrated navigation.After deeply analyzing the installation angle and lever error caused by the inconsistency between SINS and DVL installation position,two different schemes are proposed to calibrate and compensate the installation angle and lever error,and the advantages and disadvantages of the two schemes are compared and analyzed through simulation and experiment.Finally,the SINS/DVL integrated navigation method based on Variable Structure Multiple Model Algorithm is studied.Considering the errors of SINS/DVL integrated navigation system due to the influence of ocean current on DVL velocity measurement in complex underwater environment,a first-order markov process ocean current model is constructed.At the same time,considering many advantages of the multiple model filtering theory to deal with in multiple model systems,using multi-model filter method to solve problem of enormous models set by changing ocean currents.Then in order to solve the disadvantages of SINS/DVL integrated navigation method based on Interactive Multiple Model Algorithm,which is designed to solve the influence of ocean currents on navigation performance,a SINS/DVL integrated navigation method based on Variable Structure Multiple Model Algorithm is proposed.By setting reasonable simulation conditions and adding measurement velocity noises,it is proved that the SINS/DVL integrated navigation method based on Variable Structure Multiple Model Algorithm is superior to the SINS/DVL integrated navigation method based on Interactive Multiple Model Algorithm in solving the problem of ocean current noise. |
