Study On The Key Techniques Of Integrated Long Baseline/Ultra Short Baseline Underwater Acoustic Positioning And Tracking

Underwater target positioning technology plays a very important role in ocean exploitation.Due to the propagation characteristics of acoustic signal in water,underwater acoustic positioning system is still the preferred solution for underwater target positioning,among which the integrated LBL/USBL underwater acoustic positioning system is the most frequently studied and applied.In this dissertation,several key technologies in the integrated LBL/USBL underwater acoustic positioning system are studied to improve the accuracy of underwater acoustic positioning.The obtained research results have important theoretical and application value for underwater target positioning and tracking technology.The main contents and innovations of the dissertation are as follows:(1)Aiming at the problem that the accuracy of generalized cross-correlation time delay estimation technique is reduced under low SNR,a new generalized cross-correlation weighted function is proposed.The new weighted function is based on PHAT weighted function,and the power spectrum characteristic of the input signal is used to expand the amplitude of the cross-power spectrum function at the signal frequency,which enhance the component of the acoustic source signal in the signal power spectral density,and further improve the SNR.Simulation results show that the proposed weighted function can effectively improve the accuracy and stability of the time delay estimation technique.(2)The distribution of effective sound speed is studied and an iterative algorithm of effective sound speed estimation based on BELLHOP model is proposed.Based on the accurate time delay estimation,the BELLHOP model is used to calculate the fastest path of acoustic signal propagation,and the effective sound speed between the transmitter and receiver is calculated in an iterative way.The simulation results show that the proposed algorithm can reduce the computational complexity while assuring the accuracy of effective sound speed estimation,and achieve the balance between range estimation accuracy and computational complexity by adjusting the threshold,which improves the effectiveness and adaptability of the sound speed correction algorithm.(3)In view of the inability to obtain sound speed profile data in some extreme cases,an improved algorithm for effective sound speed estimation and underwater static target positioning is proposed based on integrated LBL/USBL underwater acoustic positioning system.The algorithm assumes that the effective sound speeds between the underwater target and different buoys are not equal,and uses the redundant positioning information in the LBL/USBL system to establish the optimization function.Then the particle swarm optimization algorithm is utilized to solve the optimization function to obtain the effective sound speed.(4)An improved particle swarm optimization based on region division is proposed.The algorithm assigns each particle into different regions according to their distance from the global optimum,and the inertia weight and learning factors of each particle are adaptively adjusted according to different principles in order to balance the optimization ability and convergence speed of the population.Adaptive mutation operations are also introduced to avoid the population falling into local optimal solutions.Simulation experiments show that the improved particle swarm optimization can improve the convergence speed while assuring the optimization performance,which improves the performance of the integrated LBL/USBL underwater acoustic positioning system under unknown sound speed profile.(5)Based on the study of high-precision time delay estimation and effective sound speed estimation,the positioning technique in integrated LBL/USBL underwater acoustic positioning system is studied and a data fusion algorithm based on outlier rejection are proposed.Firstly,the algorithm analyzes the positioning results of each underwater acoustic positioning system.Secondly,the Grubbs criterion is utilized to identify and reject the outliers.Finally,the distance-based weighted data fusion is used for the remaining effective positioning points to obtain the final positioning result.The simulation results show that the algorithm can significantly improve the positioning accuracy of the LBL/USBL system for underwater static targets,and ensure that the LBL/USBL system has high positioning accuracy in the entire positioning area.(6)A R-T-S interval smoothing algorithm based on Unscented Kalman Filter is proposed to solve the tracking problem of the system.The algorithm makes full use of the multiple positioning informations obtained by the USBL system through multiple buoys,which further reduces the influence of measurement error.The simulation experiments prove that the algorithm can obtain better filtering performance than using only the Unscented Kalman Filter,and ensures that the LBL/USBL system has a high positioning and tracking accuracy for underwater moving targets in the case of large error from USBL system positioning.