Research On Terrain Suitability Of Underwater SINS/TAN Integrated Navigation Systems

The underwater vehicles are an important tool for the exploration and development of oceans.The combination of TAN and SINS can effectively meet the need for autonomous long-term high-precision navigation of the underwater vehicle.Terrain suitability has a significant relationship with the terrain matching accuracy.In this paper,the suitability of underwater SINS/TAN integrated navigation is studied.The main work is as follows:(1)The working principle and error model of SINS are discussed.In addition,according to the composition and working principle of underwater TAN,three typical terrain matching methods are analyzed and compared.The basic principle of SINS/TAN is further analyzed,and the mathematical models are constructed.(2)The basic concepts and ideas of suitability analysis are explained.The basic framework of key technologies for terrain suitability is established.Based on this,a database composed of basic suitable matching features and suitable verification indicators is constructed.The suitability qualitative evaluation strategy based on screening classification is outlined.(3)The classification problem of CMAs is studied from the perspective of pattern recognition,and the basic suitable matching features extracted by FA are used as the input vectors of the classifiers.The self-organizing optimization classification of CMAs is realized by GA and SVM.In the self-organizing classification method,SVM is used as a classifier,GA is used to further select the optimal feature subset and optimize SVM parameters.In addition,a one-to-one strategy is introduced to solve the regional multi-classification problem.The results show that the classification accuracy is significantly improved and the risk of misjudgments is reduced.(4)Based on the self-organizing optimization classification method based on FA and GA-SVM,a comprehensive simulation experiment of underwater SINS/TAN integrated navigation suitability analysis is carried out,and an improved particle swarm optimization terrain matching algorithm is proposed.Based on this,the selection experiment of the adaptation zone is used to verify the validity of suitability-matching method.