Research On Detection Method Of Multi-beam Echo Sounder Installation Biases

Multi-beam Echo Sounder(MBES)was the mainstream instrument for underwater terrain acquisition currently.And installation biases often caused problems such as tilt,translation or rotation of the terrain.In view of the current MBES installation biases detection using Patch test method that had a lot of limits of requiring special terrain,detection process complexity,decoupling,and an overall detection method of installation biases was proposed.This method took into account the coupling between the installation biases.By analyzing the influence of each installation biases on the terrain,models of the influence of installation biases on the sounding were formed.With the help of the terrain image and the point cloud data formed by the common coverage area of the round-trip line,a method of installation biases overall detection was proposed,which realized the integrated solution of the installation biases.Therefore,simplified the detection process,reduced the special requirement on the terrain,and impove the detection accuracy and efficiency.Main work and contributions in the thesis were as follows:(1)Research on the soundings calculation was carried out,and the soundings compution process was formed.Considering the deficiencies of decoupling,searching for special terrain and complicated detection steps in Patch test,a method for calibrating the installation biases using soundings coordinate difference of the round-trip line was proposed.(2)Analyzed the influence of installation bias on topographic measurement results,and drived error models such as water depth error ratio and along-track error ratio,then formed a comprehensive error model for the influence of installation biases on soundings,and solved the problem of the magnitude of the influence of the installation biases on the terrain.(3)Base on the common coverage area of the round-trip line,the method of solving the installation biases by SIFT matching terrain elevation(water depth)image and by ICP point cloud registration was proposed,which solved the problem of high precision acquistion of installation biases.Compared with the results of Caris,the detection accuracy was obtained better than 0.1°and 0.08°respectively.Comparing the both methods,it was considered that when the terrain was flat but there was microtopography,the image matching method used to solve the installation biases was better.When the local shape had a large slope or the height difference(water depth difference)changed greatly,the point cloud registration method used to calculate the installation biases was better.