Multiple Autonomous Underwater Vehicles Cooperative Localization Based on Asynchronous Relative Observation

The asynchronous relative observations are incorporated in the cooperative localization system of multiple Autonomous Underwater Vehicles (AUVs) in unknown environment in this paper. First, the motion model of multi-AUV cooperative localization system is established. Then, the asynchronous relative observation is obtained when two AUVs meet and the corresponding measurement model is introduced. The state and the covariance matrix of this AUVs group can be easily distributed and calculated on each AUV platform node by using the distributed extend Kalman filtering (DEKF) algorithm. The cooperative localization with the proposed DEKF algorithm was implemented and tested in the simulated environment. The theoretical analysis and simulation results showed that the asynchronous relative observation between AUVs can be effectively fused by the DEKF method.