| The importance of the ocean,as a contested place for great powers,is self-evident.China has a maritime land area of more than 3 million square kilometers,and accelerating the development of the marine economy and realizing a maritime power is an important development goal of China.As the basic support and prerequisite for all marine activities,the seabed geodetic control network is the first basic problem to be solved at present.To establish a high-precision submarine geodetic control network,we must first solve the problem of highprecision underwater acoustic positioning and positioning.This paper conducts systematic research on underwater sound velocity error correction and positioning,and the main research work and results are as follows:(1)Sound speed profile data is the basic data for high-precision position determination of underwater reference points.Aiming at the actual problem that the measured depth of sound speed profile is less than the position depth of the underwater reference point transponder,a calculation method of extrapolating the ssp using a polynomial and other function model is proposed,which effectively makes up for the problem of insufficient depth of the measured ssp data,and the experimental results of the measured data of different sound segments show that the function model fitted with the measured data at a depth of 2000～2500m extrapolates the sound velocity at a water depth of 2500～3100m,and the fitting extrapolation effect of the second-order function is the best.(2)Aiming at the problem of low computational efficiency caused by large observation redundancy of ssp,the geometric characteristics of ssp are used to improve the Douglas-Puck algorithm through the local perspective,and the adaptive dilution of ssp observation data significantly improves the calculation efficiency of ssp,and the experimental results show that when the threshold is 0.02～0.1m/s,the horizontal direction accuracy of the improved DP algorithm is comparable to that of the original DP algorithm,but the reduction rate is higher.When the threshold is 0.1～2.0m/s,the horizontal accuracy of the improved DP algorithm is significantly higher than that of the original DP algorithm,and the reduction rate is comparable to that of the original DP algorithm.(3)Aiming at the problem that it is difficult to calculate the incidence angle of the voice line accurately,the Newtonian iterative algorithm is improved by introducing the downhill factor,and the experimental results show that when the incidence angle of the voice line is50°～65°,the accuracy of the Newtonian iterative method improved by the downhill factor is33% higher than that of the original Newtonian iterative method.(4)Underwater high-precision positioning must construct a random model that is compatible with the time-delay observation data.In this paper,the calculation effects of the positive proportional function,the incidence angle cosine function,the incidence angle exponential function,the incidence angle piecewise cosine function and the incidence angle piecewise exponential function taking into account the weight function of the sound line incidence angle are compared and analyzed,and the deep-sea experiment shows that the segmented cosine function and the segmented exponential function have the highest positioning accuracy of the seabed earth reference point when the weight function is used.(5)Aiming at the problem that the unit weight variance factor of multiple types of timedelay observation data in actual observation is inconsistent and cannot be accurately known,and there is inevitable abnormal error in the observation data,an underwater localization solution model combining Helmert variance component estimation and IGG-III tolerance estimation is constructed,while suppressing the influence of observation anomaly error and balancing the contribution of different types of observation data to the localization results.It is76.24% higher than the least squares method and 34.95% higher than the Helmert variance component estimation method. |
PDF Full Text Request