The Simulation Study On 3D Underwater Positioning In The Deep Ocean

Underwater GPS positioning system,which is also called GPS Intelligent Buoys system, provides positioning and navigation services for the underwater objects by laying out GPS buoys on the ocean surface to form a geodetic datum. China has successfully developed an underwater GPS positioning system in 2001. And many experiments carried out from the 15 to 110 meters depth of Thousand-island Lake, Sanya and Fuxian Lake have proved the feasibility and capability of this system in shallow sea. This paper is focus on a further research to find whether the system could work in the deep ocean, and what accuracy it can reach.In the paper, the sound speed profile was obtained by the sound experiential formula using the data of seawater temperatures and salinities which are measured by the Argo plan. Subsequently these data were substituted into the acoustic wave propagation law in order to analysis the usability of the hyperboloid model in the deep sea, and the cover area of the signal. Finally, the observation data of the hyperboloid model were simulated to discuss and analyze the effect of positioning errors on the outside fitting positioning accuracy. The details of this paper are as follows:1. The analysis of underwater acoustic velocity based on Argo dataAccording to the underwater acoustic ray theory, the propagation path of an acoustic signal depends on its velocity. And the underwater acoustic velocity is a function of seawater temperature, salinity, and depth or pressure. This paper analyzed the sound speed profile using the observation data of Argo plan which are gotten from a water depth of 2000 m.2. The effective acoustic positioning area in the seawaterDue to existence of the sound shadow zone in the ocean surface, some original sound signals could not arrive at the sea surface. According to the Snell law, there is an angular limitation when radial transmits from the sound source to the ocean surface. This limited emission angle sound signal is determined in the paper. In addition, the layout area of GPS Buoys and acoustic position area of some networks at the case study area of 47.956oN and -128.971oW where the sound source is at about a water depth of 2000 m were also determined.3. The feasibility of application of hyperboloid model in the deep oceanThe effectiveness of using hyperboloid model in shallow water has been investigated by some researchers and the results show that the hyperboloid model could achieve high precision. However, more researches are still needed since sound ray curve which affects positioning accuracy will became larger along with the growing of water depth. This paper provides detailed curve degree of the signal at a water depth of 2000 m and 8000 m in the case study area, and the numerical experiments show that the hyperboloid model can be used in the deep ocean.4. Positioning accuracy analysis of the hyperboloid modelBase on the sound wave ray propagation law, the simulation method of hyperboloid model observation data is proposed in this paper. With this method the data at a water depth of 2000 m in research area were simulated. The case study data which cover 10km×10km area with the resolution of 50 m×50 m were calculated using the least squares method. The results show that the outside fitting accuracy of underwater GPS system is less than 4 m in horizontal direction and 10 m in vertical direction.Because it is difficult to survey the velocities of sound radial in each ray, the same velocity along the vertical transmit path was used follow. The results show that using the same velocity could greatly improve the precision outside of the buoys and lightly reduce the precision inside.In addition, relative time error was added to the observation data. The results show that the time errors which are less than 10-4s have little influence on the positioning precision. Most of the effects in horizontal accuracy are from -0.5 m to 0.5 m, while the vertical affects are smaller than 1 m.Furthermore, different emission angle of signal has different propagation path. The bigger the angle is, the larger the sound ray curve will be. Therefore, the cutoff angle is introduced to underwater position to restrict this curve. The results show that when the cutoff angle is 75o, the horizontal accuracy is less than 1 m, and the vertical one is less than 4 m, and therefore the position quality is improved.5. The influences of random errors on the positioning resultsThis paper employed multi-epoch observation of a point at a water depth of 2000 m, and random errors at 10-3 magnitude order were put into these observation data which were the arrival times of signals. Utilizing these data to execute positioning calculation, the results show that the positioning error is about 2 m at X axis, 2 m at Y axis, and 10 m at Z axis.