Research Of Underwater Navigation Based On Full Tensor Gravity Gradient

An underwater path planning method considered both matching area selecting andunderwater aided navigation based on gravity gradient is a new underwater passive navigationtechnology, it has become one of the hot research.Gravimeter and gravity gradiometer, gravitygradient reference map and related matching algorithms are the three elements of the gravitygradient assisted navigation.Route planning and adaptive area selection is the prophase workof gravity gradient aided navigation. Therefore, this thesis have done the work base on the fulltensor gravity gradient aided navigation as follows：Firstly, the preparation methods of the gravity gradient reference map is introduced. Thereference map of gravity gradient is an indispensable and important component of gravitygradient aided navigation. There are high precision of the gravity gradiometer in foreigncountries. But because the measured gravity gradient data cost is too high. And some areacannot be measured. High precision gravity gradient maps at present can not cover the world.But China has the world’s high precision terrain data. This paper introduces the method toobtain the gravity gradient reference maps based on the spherical harmonic model. Then itintroduces the method to obtain the gravity gradient reference maps based on terrain elevationmodel. And it discusses the influence of density factors on the preparation of reference map.Secondly, the gravity gradient feature and the method of adaptive area selection areintroduced. The environment of Seabed terrain is complex. It extracts and analyzes the gravitygradient features to find the obvious characteristics regional. It is helpful to improve thepositioning precision of gravity gradient aided navigation system. This paper extracts thefeatures of full tensor gravity gradient, These features include the mean, variance, energy andentropy. In order to improve the matching area selection generalization ability. The paperuses the support vector machine to select the adaptive area.Finally, according to the different scenarios of the adaptive and non adaptive area, itpresents a method of dynamic route planning underwater while it considers a adaptive areaselection and target detection simultaneously. Adaptive area have obvious characteristics, It issuitable for navigation. The obvious characteristics of adaptative area often complex terrain,there may be a reef. In the non adaptive area, it needs to filter tracking strategy for navigationand positioning. but it may encounter small reef or other moving Submersible. Therefore, theunderwater dynamic route target detection simultaneously is proposed in this paper,Simulation results show that the expected purpose can be achieved through this method.and it is correct and effective.