Research On AUV Beacon Navigation And Positioning Technologies Based On Nonlinear Filtering

In recent years,the navigation and positioning technologies based on beacons ranging-only have been more and more widely used in the field of underwater unmanned system,and have gradually become a research hotspot and one of the main development orientations in the field of underwater navigation.Besides,simultaneous localization and mapping algorithms have been more mature in the field of land robot navigation,but less applied in the field of underwater beacon navigation.In this thesis,based on dead reckoning system,a priori known fixed beacons navigation system and SLAM algorithms,the navigation and positioning technologies of AUV beacon based on nonlinear filtering method have been discussed,and finally realize the design of range-only SLAM navigation algorithms based on unknown random multi-beacon.The main research contents of this thesis are summarized as the following three parts:In the first part,the model of dead reckoning system has been established,and the error propagation of the system has been studied.Firstly,the modeling of external electronic compass and doppler velocimetry sensor has been completed.Then,the principle of dead reckoning and the characteristics of error accumulation with time have been analyzed.After that,the dead reckoning model of AUV has been established and applied to the longitude,latitude and altitude position update algorithms.At last,aiming at the dead reckoning method,the error propagation equation of the dead reckoning system has been developed and solved in detail.At the same time,the influence of the doppler velocimeter scale factor error,installation error matrix and the angle measurement error of compass on the error propagation caused by the position update of dead reckoning have been simulated and analyzed.In the second part,the model construction and solution of fixed beacon navigation system under the condition of known a priori has been deeply discussed and analyzed.Firstly,choose the modeling and solution of the long base-line underwater acoustic positioning system as the research main line,the reverse solution principle of the spherical intersection model based on the least square method has been discussed.Then,after analyzing the shortcomings of the least square static positioning method and the observability of the single beacon ranging integrated navigation system,this paper has proposed an EKF based integrated navigation model of beacon ranging and dead reckoning to realize the real-time dynamic positioning of AUV and restrain the error divergence of dead reckoning.Finally,with the help of simulation experiments,the effects of the selection of AUV motion path,the number of beacons and the measurement error of beacon ranging sensor on the navigation and positioning performance of AUV have been deeply discussed and analyzed.In the last part of this thesis,the SLAM navigation problem based on prior unknown random multi-beacon has been studied.Firstly,the navigation problem of slam with random multi-beacon has been described in detail.Then,the realization difficulties of slam navigation in the case of range-only have been analyzed.At last,based on the extended kalman filter method,the RO-EKF-SLAM navigation algorithm based on unknown random multiple beacons has been studied and designed.At the same time,this thesis has simulated AUV to perform a circular motion model,used for analyzing the performance of the algorithm.