Research On Multi Sensor Information Fusion Technology In Collaborative Navigation Network

With world gradually paying more and more attention on ocean issues, the unmanned surface vessel develops with an unprecedented rapid speed. And it will play an important role in future naval warfare. In order to adapt to future war in more complex marine environment,Compared with traditional navigation, the collaborative navigation technology of USV is facing huge challenge.This topic will start with multi-sensor information fusion technology in Collaborative navigation network. Firstly, it briefly introduces current situation of research on USVs.Through analyzing the domestic research on development of US Vs, we can know that although our country has gained great development in recent years,there are still large gaps compared to United States and Isra. For the purposes of collaborative navigation network of unmanned vessels, the way to shorten the gap is not only to improve the accuracy of collaborative sensor networks, but also to improve the accuracy of information fusion optimal estimation criteria. Thus the information fusion and structure of multi-sensor data fusion are mainly reported and analyzed in this thesis.Secondly, this thesis aims to analyze the information sources in cooperative navigation network of USVs. It also performed simulation and analysis according to the suitable mathematical model established under information feature. In view of important position, the thesis reports and analyzes the principle of inertial navigation systems, modeling and simulation in detail. Modeling and simulation inertial navigation systems, GPS and DR provided data sources to later achievement of optimal estimation algorithm of information fusion.Finally, in this paper, the optimal estimation criteria information fusion in navigation network is analyzed, and different Kalman filters were designed based on data fusion subsystem features. By using a linear Kalman filter to estimate the system variables, GPS and inertial navigation system achieved good results. Also, it achieved satisfied results using nonlinear Kalman filter and particle filter to estimate the state variables.For the purposes of navigation of unmanned vessels collaborative networks, positioning accuracy and quality of the navigation information could be highly improved by using multi-sensor data fusion technology.