Research On The Integrated Navigation Algorithm For Submarine Cable Inspection Based On Multi Source Data Fusion

The submarine cable is known as the "central nerve" of the sea and conduct data communication and power transmission between countries.The damage of the submarine cable will have a serious impact on the remote data communication.Therefore,submarine cable patrol inspection becomes a necessary task.With the development of science and technology and the progress of underwater sensor technology,underwater AUV inspection has gradually become the main way of submarine cable inspection.In the complex seabed environment,the accuracy,reliability and stability of the underwater AUV navigation and positioning system has become a necessary research topic.In order to solve the problem of underwater navigation and positioning,this paper relies on the project "Development and demonstration application of operational submarine pipeline inspection cable-free robot" to carry out the research of the integrated navigation algorithm based on the underwater submarine cable inspection AUV,which can fuse the data of the multiple sensors on board,and finally outputs reliable navigation information for the submarine cable inspection AUV.This paper firstly introduces and analyzes the design scheme of the integrated navigation system for the submarine cable patrol AUV in the project,introduces the sensor models of the global positioning system(GPS),strapdown inertial navigation system(SINS),Doppler velocimeter(DVL),depth meter,and altimeter used in the integrated navigation system in this article,respectively expounds the working principles of each sensor,and establishes corresponding parameter error models.Secondly,the filtering design is carried out for the SINS/GPS surface integrated navigation system and the SINS/DVL/depth meter/altimeter underwater integrated navigation system respectively.The adaptive fault-tolerant Kalman filtering model and federated filtering model are selected and analyzed,and the mathematical model of the sub-filter in the model is built.Then,for SINS/GPS integrated navigation system,the indirect method compact combined Kalman filter model is selected to design the adaptive fault-tolerant Kalman filter algorithm,and the GPS data is filtered and estimated after outlier detection.The results of trajectory prediction by different methods are obtained through simulation experiments.The advantages of the adaptive fault-tolerant Kalman filter algorithm are verified through error comparison,which can be used in SINS/GPS integrated navigation system.Finally,the SINS/DVL integrated subsystem is filtered by the DB Bayesian Kalman filter,aiming at the federated filtering structure selected by the underwater integrated navigation system.In the federated filtering algorithm,adaptive information distribution strategy is used to make feedback adjustment according to the precision distribution parameter information of the filtering subsystem.The navigation prediction results are obtained through simulation experiments.The federated filtering algorithm based on adaptive information distribution strategy can effectively suppress the divergence of navigation errors and improve the stability and reliability of the navigation system.