Research On Ultra-Short Baseline Localization And Calibration Technologies For Complicated And Compact Array

This is an era of ocean large-scale exploitation.As a basic guarantee,underwater acoustic positioning technology play an important role in marine development.Ultra-short baseline(USBL) acoustic positioning systems has small size and can be easily deployed,so they can improve the efficiency of oceanic engineering to a great extent.Using USBL system to determine the target's geometric coordinate needs the collaboration of GNSS,attitude sensor,and acoustic positioning system.However,some factors restrict the accuracy of USBL systems,such as acoustic positioning errors and installation errors between sensors.In order to improve the positioning accuracy of USBL,a systematical research on high-precision ultra-short baseline positioning technology is conducted,including three aspects: high-precision acoustic positioning for the USBL array with arbitrary configuration,array geometry errors correction,and angular misalignment calibration.Due to the geometric approximation of the positioning model,the existing algorithms also have systematic errors.To solve the problem,a generalized USBL acoustic localization algorithm is proposed for the complicated and compact array.Depending on the coordinate transformation theory,the proposed method establishes an observation equation of target bearing for each baseline.By selecting different combinations of baselines,the target bearing can be calculated through the least squares.The algorithm not only is able to be applied to the array with complicated geometry,but also greatly reduces the system positioning error.The simulation result shows that the proposed acoustic positioning algorithm has less systematic errors and can provide more accurate results than the existing ones for both 2-dimensional and 3-dimensional arrays.The noncoincidence between the transducer geometry center and its radiation center lead to array geometry error.In order to correct array geometry error,a high precision error correction algorithm is proposed.The algorithm utilizes effective sound velocity theory to determine the element position in a measurement coordinate system,which reduces the dimension of the unknown parameter space and improves the estimation accuracy and robustness.Moreover,the numerical analysis shows that different elements have subequal positioning error.Therefore,the algorithm converts the element positions to baseline lengths so that positioning errors can be removed,and then reconstructs array geometry according to the baseline length to achieve high-precision correction.The results of anechoic pool measurement show that the proposed method is more accurate and robust.And the field test result indicates that the positioning accuracy is greatly improved with after array geometry is calibrated through the proposed method.Because of the influence of environmental factors,such as wind and wave,the actual ship trajectories used to calibrate the angular misalignment is easy to be distorted,which make the calibration result inaccurate.To solve this problem,a novel angular misalignment calibration algorithm based on matrix decomposition is proposed.The algorithm forms observation equations of the angular misalignment at each point in the trajectory,so that the result is free of the distorted trajectories.Also,the matrix decomposition is used to remove the system residual error caused by the nonlinear calibration model during the Taylor expansion.The simulation results show that the estimation accuracy of the angular misalignment obtained by the proposed algorithm is completely unaffected by the track shape.The lake experiment data shows that the three proposed algorithms improve the positioning accuracy of USBL system to a great extent.