| Underwater positioning technology plays an essential role in the oceanic exploration and scientific research, and without accurate positioning the oceanic information is useless. Compared with positioning technologies used on the sea surface, underwater positioning technology is lagged behind much more, especially for the development of long baseline (LBL) acoustic positioning technology. Although the research on LBL has been implemented in China, the development progress is slower than that in advanced countries. Meanwhile, due to the commercial blockade few technologies of LBL were imported from advanced countries. Thus we are still resting on learning foreign LBL technologies from the expensive commercial LBL system, and the establishment of our own LBL system with independent intellectual property has not been finished yet. The current situation indicates that plenty of works are required in the development of domestic precise underwater positioning technology and system of our own. In order to promote the progress of LBL research in China, in this thesis domestic and foreign theories and methods of LBL positioning are investigated and summarized completely, a set of systematic theory and method is formed, and a software is developed for processing LBL high-accuracy underwater positioning data, which filled gaps in the LBL field in China. The achieved results in this research could benefit the further development and applications of the domestic LBL positioning system.In this thesis, the following studies are carried out:1. System and theory of LBLBased on the analysis of current LBL systems, the theory of LBL positioning, the process of the positioning and data processing are depicted in detail.2. Correction of sound ray, the inversion of sound velocity profile and the construction of sound velocity field.(1) Based on the theory of sound propagation, a method of sound-ray correction is proposed, which reduces the complexity of the algorithm in sound ray tracing, improves the efficiency of ranging calculation, and thus realizes the fast tracing of the sound ray propagation.(2) Using the history SVPs and additional constraints of sounding time as well as empirical orthogonal function (EOF) algorithm, an inversion algorithm for achieving precise sound velocity profile is developed, which provides a way to get precise sound velocity without actual SVP. (3) Based on limited Sound velocity profiles, a method for constructing local sound-velocity field model is proposed by means of space-curved surface interpolation of EOF coefficients. The model accuracy achieved by the method in a experiment is better than0.5m/s. Compared with traditional method based on SVP space interpolation, the new method presents significant advantages.3. Study on methods of coordinate-datum transmission and the integrated data processing of underwater acoustic array network(1) By studying the relationship between the vessel frame system (VFS) and geographic reference system (GRF), a model for transferring the3-D solution of GNSS to that of vessel-mounted acoustic transducer is obtained, which establishs the foundation to transmit the3-D coordinate datum form sea surface to the underwater acoustic-array network.(2) The absolute calibration technique for the underwater acoustic-array network is studied and a round-calibration method is proposed. The highest-accuracy positioning solution can be achieved for acoustic-array element while the circle radius equals water depth. Without sound velocity profile, the centimeter-level accuracies in determining location of acoustic-array element and transiting coordination datum from sea surface to underwater can be got by the round-calibration method.(3) The methods of underwater acoustic array network adjustment are studied and several network-adjustment models are proposed for different situations. The unconstrained3-D net adjustment model is provided for the trilateration network. Besides, a2-D adjustment idea and model as well as3-D net adjustment model with additional depth constrains are put forward too. As to the trilateration network adjustment, we introduce the round calibration method for the absolute position and provide the2-D/3-D adjustment methods and models for the array calibration with additional part of the absolute coordinates. These studies on net adjustment models realize the precise determination of the relative/absolute position of the underwater acoustic array element.4. The effects of the structure factors of underwater acoustic array on LBL positioningAfter the systematic study on the influence of the shape of array network, height difference of adjacent array elements and precision of array element on LBL positioning, the square network shows the appropriate covering area with the highest position accuracy for array element, which could used for the seamless splicing in the construction of array network, and could also be used as a basic network shape in the construction of underwater array network. At the same time, it is better to arrange each array element on flat and open seabed for improving the array element position accuracy; it is also suggested that the height difference of adjacent array elements should be controlled within10m in the water lays of thermocline and seasonal fluctuations, while within50m in the deep channel layer and a positive gradient layer. Meanwhile, to meet high-accuracy LBL positioning, the calculation model of optimum length of underwater acoustic array network is given, and the array element should be higher a order of magnitude than that of LBL tracking and positioning for better position accuracy. With the research on optimum LBL sailing route, the ideal sailing with higher positioning accuracy is got at locations near array elements within the regular triangle or square when water depth is shallower than500m water depth, while at the center of these basic array networks when water depth is deeper than500m. However, the water-depth devision could be reduced to200m when a regular pentagons or a regular hexagon is used as basic underwater array network.5. LBL Positioning Model(1) For different measurement conditions, the corresponding LBL positioning models are studied. For ranging observation, the3-D rendezvous-positioning model is constructed. For observations of distance and depth, the method of projecting positioning and the2-D rendezvous-positioning model are developed. Based on above relative network adjustment, the absolute coordinate of array element from round-calibration method is introduced into net adjustment. In association with these models, the accuracy assessment and error ellipse of these adjustment models are also provided.(2) For space rendezvous, plane intersection, space rendezvous added depth constraints and space rendezvous in independent coordinates, these models are studied through experiments and comparing analysis. Due to asymmetric distribution of the transponder and transceiver in space, the vertical solution of LBL positioning is instable in the space rendezvous model; therefore, pressure sensor is suggested in high-accuracy LBL positioning and the water depth should be used as a constraint condition in space rendezvous positioning.(3) For conditions of no sound velocity profile or existing significant impact of sound velocity error on LBL positioning, a differential positioning method is proposed. Based on the correlation of the sound-velocity error influence in adjacent ranging observations, and with the help of single-difference used in GPS RTK positioning, the differential error equations are constructed among adjacent ranging observations in the differential positioning method, which significantly improves the LBL positioning accuracy in an experiment.(4) INS is also introduced into LBL positioning. In order to fulfill the fusion of LBL positioning solution and INS solution, a model of robust Kalman filter is constructed, which realizes the mutual compensation and checking between them and improves accuracy and reliability of the LBL tracking. 6. Software development of LBL synthetic positioning data processingBased on the theoretical research of LBL and combined with analysis of the software requirements, the software framework for LBL synthetic positioning data processing is built and the software is developed, which is the first set of LBL positioning data processing software of China.7. Test, data processing and analysisThe LBL positioning test was carried out in the Songhua Lake and the experimental data was also processed and analyzed. Compared with GPS RTK positioning, the results show that the positioning accuracy of lake basin array network, tracking positioning within underwater array network and outside the underwater network reached centimeter level, decimeter level and meter level, respectively. And the results meet with the requirement of underwater tracking positioning, and also verify the correctness of the theoretical methods and the reliability of the software system proposed in this thesis. |
PDF Full Text Request