Research On The Geomagnetic Bio-inspired Navigation For AUV Based On Optimization Searching Strategy

For the limitation of geomagnetic data acquisition with high precision and large range for ocean environment,this thesis mainly studies the geomagnetic bio-inspired navigation problem without a priori geomagnetic map for autonomous underwater vehicles(AUVs).Geomagnetic bio-inspired navigation method is inspired by the geomagnetic navigation behavior of animals which mainly depend on their sensitivity to the variation trend of geomagnetism.Essentially,the process of geomagnetic bio-inspired navigation can be regarded as a multi-objective convergence process from the present position to the desired position of the geomagnetic components.This thesis presents the geomagnetic bio-inspired navigation method based on optimization searching strategy.The main contributions of this thesis are summarized as follows:Firstly,for the multi-objective searching problem with no knowledge of the objective function characters,this thesis presents the geomagnetic bio-inspired navigation method based on the evolutionary strategy of random walk for geomagnetic components searching.The movement direction of the vehicle is determined in accordance with the changes of the objective function,and the searching performance is analyzed.Finally,compared to two classic algorithms,the proposed method obtains the better robustness performance.Secondly,aiming at the problem which is time-consuming and difficult in naviga-tion steering mechanism,this thesis presents a new search strategy based on animals'tendency behavior.From the perspective of geomagnetic trend sensitivity,it is clear that the mechanism of magneto-taxis movement behavior whether it is far away the stimulus source or not.This thesis studies the classical information tropism algorithms which can be applied in solving the multi-objective searching problem of geomagnetic bio-inspired navigation,and analyzes the performance of algorithms.In this thesis,the search strategy is optimized,and the algorithm based on fuzzy decision is proposed.Thirdly,aiming at the problem of the searching space malposition caused by geo-magnetic anomalies and ocean current,the improved multi-objective searching methods based on behavior constraint are proposed.Considering the existence of the geomag-netic anomalies and the ocean current,which has significant influence on the geomagnetic navigation system for AUV,it often disrupts the distribution of the geomagnetic field.This thesis introduces the environmental monitoring operator and the energy function respectively for two cases above.The results show the improved multi-objective searching methods can overcome the interference from the geomagnetic anomalies and the ocean current.Finally,XQ-ROS mobile robot is used to establish the geomagnetic navigation exper-iment platform.An experimental verification work is carried out for the multi-objective searching method based on geomagnetic trend sensitivity.According to the characteristics of the indoor geomagnetic environment,the thesis studies the feasibility of the proposed geomagnetic navigation methods.Then,the proposed optimization searching algorithms are tested,and the results demonstrate the reliability and feasibility of the proposed algorithms.