Error Analysis And Algorithm Research Of Mutil-AUV Cooperative Navigation

Autonomous Underwater Vehicle(AUV),as a booster for marine development and exploration,can be used for ocean exploration in high-risk areas of the deep sea,and has received extensive attention from countries around the world.Especially in the cooperative navigation of AUV system,by reducing the underwater vehicle configuration,enhancing information sharing between AUVs and improving navigation accuracy,it has become a new research hotspot in the field of underwater navigation.Based on the multi-AUV collaborative system,this paper studies the error and algorithm of the navigation and positioning of the system.The research work of the dissertation mainly includes the following aspects:1?There are two cooperative navigation positioning methods,single-leader and multi-leader.The positioning principles are introduced,positioning model established.For the single-leader mode,the information source is less,which leads to poor observability of the system,and the observability analysis is carried out to pave the way for the simulation verification.2?Based on the underwater acoustic ranging of cooperative navigation system,the factors affecting the multi-AUV cooperative navigation accuracy are analyzed,and the error analysis from the AUV dead reckoning is carried out in cooperative navigation.The factors affecting the precision of cooperative navigation,including sound velocity error,time delay,clock synchronization,and ocean current influence,is analyzed quantitatively or/and qualitatively by studying their effect on the ranging accuracy.3?Two cooperative navigation algorithms based on the Extended Kalman Filter(EKF)and the Unscented Kalman Filter(UKF)are studied.Based on the basic principles of the two algorithms,combined with the cooperative navigation motion model,the steps of the cooperative navigation positioning algorithm are designed.The two algorithms are compared in terms of filtering accuracy and filtering stability.The effectiveness of two cooperative navigation filtering algorithms is verified by theoretical analysis and simulation.4?In order to achieve the localization of a single-leader system in the influence of ocean current,a cooperative navigation filter algorithm based on improved particle swarm optimization(PSO)and UKF is proposed.The idea of selection,intersection and mutation in genetic algorithm is integrated into the improved particle swarm optimization algorithm.In this paper,the algorithm is used to optimize the noise matrix in UKF to realize the cooperative navigation of single-leader under the influence of ocean currents.And simulation cases are designed to verify its better effectiveness.