Research On Positioning Method Of Underwater Autonomous Vehicle Based On Multi-source Fusion

With the advancement of the maritime power strategy,the exploration of underwater resources and the construction of subsurface projects have drawn more and more attention.Integrated navigation and positioning technology has important application value in military and civilian fields such as underwater target tracking,marine resource development,and navigation of underwater autonomous vehicles.The federated filter positioning method based on multi-source information fusion,as an important content in integrated navigation,will directly affect the performance of the navigation system.Therefore,in the research of underwater autonomous vehicle positioning technology,it is important to improve the federated filtering algorithm based on multi-source information fusion.In the federated filter integrated navigation system based on multi-source information fusion,the filtering estimation result of the navigation subsystem has a direct impact on the performance of the integrated navigation system.The performance of integrated navigation system can be improved by improving the accuracy of navigation subsystem and the federated filtering algorithm.Based on this,we try to improve the filtering algorithm of federated filtering and its navigation subsystem.Based on the multi-source information fusion technology in underwater vehicle combination positioning,this thesis proposes a strapdown inertial navigation/Doppler log/ultra-short baseline combined navigation and positioning method based on improved federated filtering algorithm.Firstly,in the design of the SINS/Doppler Logging(DVL)integrated navigation subsystem,based on the error source of the Strapdown Inertial Navigation and the influence mechanism of the positioning result,the filter gain compensation algorithm is applied to the traditional filtering to improve the algorithm.So that the SINS/DVL integrated navigation and positioning method based on filter gain compensation and adaptive filtering technology is proposed.Then,in the research process of the underwater ultra-short baseline positioning(USBL)system,there are different degrees of errors for the estimated phase difference of the signal between the array elements,which will result in low positioning accuracy.From the measurement error source and positioning result,the minimum mean square error is used as the best estimation criterion based on the influence mechanism.We use Kalman filter algorithm to achieve signal denoising,obtain high-precision phase difference information,and propose an ultra-short baseline positioning method based on Kalman filter algorithm.Finally,this thesis uses strapdown inertial navigation system,Doppler log and ultra-short baseline positioning system,and applies the federal filtering algorithm to fuse their navigation information,and designs the SINS/DVL/USBL integrated navigation based on improved federated filtering algorithm.The positioning method realizes the fusion of multi-source sensor information of underwater vehicles.Based on the design of this thesis,the underwater integrated navigation positioning model was designed and completed.In order to verify the performance of the system,the performance of the improved federated filter was tested.The improved federal filtering algorithm designed in this thesis was compared with the traditional federal filter and the SINS/DVL integrated navigation system.The experimental results show that the positioning method based on multi-source information fusion designed in this thesis can deal with the AUV motion state mutation problem well under the complex tasks,and can also suppress the divergence of errors in long-range conditions.And the new system can maintain high precision and stability.