Research On Calibration And Combined Algorithm Of SINS/USBL Underwater Integrated Positioning System

China has vigorously developed various marine technologies and advanced manned and unmanned underwater vehicles.All kinds of marine engineering and underwater military activities rely on underwater vehicles that can work stably in the deep sea.In this paper,based on the project requirements,research on the calibration algorithm and integrated algorithm based on the INS(Inertial Navigation System)/USBL positioning system is carried out.The main work is as follows:(1)The USBL system and various underwater acoustic positioning technologies is studied.The error model of the USBL system is built up to analyze the effect of error on positioning accuracy.The error sources include time delay error,sound ray bending,direction-finding error,lever arm and installation error angle,etc.It can be found that the sound line bending will bring about 0.8% error to the slant distance measurement with the 70 m depth by simulating the sound velocity correction algorithm.(2)The installation error calibration of the USBL system is studied,and three calibration algorithms are introduced detailedly,inclouding,least-squares based on small-angle approximation,Gauss-Newton method based on nonlinear optimization,and Kalman filter method.The simulation experiments show that the positioning error of the navigation system can be within 2m after calibration,which has more than 80% error reduction.The calibration accuracy of the installation error angle can be within 0.1°.Among them,Kalman filter can process data in real time,and is more suitable for online calibration.(3)The INS/USBL integrated navigation algorithm is studied and designed.The loose integrated method based on position information and the tight integrated method based on azimuth and slant distance information are introduced.The slope distance of each element is added to the observation to perform the tight integration,and the depth information is used to constrain the height information of the system.With the 1% slope distance error,the positioning accuracy of loose integration model can reach 0.5% of the maximum slope distance;in contrast,the tight integration has better performance,the positioning accuracy can reach 0.3%.(4)The river experiment is carried out to test the performance of the algorithm studied in the actual environment.The result show that after installation error calibration,the positioning error for transponder can be within 2.5m,which is reduced by more than 60% compared with before calibration.The positioning error for loose integrated and the tight integration can be within 3.6m,which effectively solve the problem of pure inertial navigation acumulated error.In this paper,the design of the INS/USBL positioning system is completed,and the key technologies of underwater acoustic positioning,installation error calibration,and integrated navigation is studied.Theoretical research and experimental results show that the system has good positioning performance.