Objectives  In order to solve the problem of inertial navigation systems(INS)and doppler velocity logs(DVL) on multiple autonomous underwater vehicles(AUV) failing or having no sensors, combined with the traditional extended Kalman filter estimation, a cooperative positioning method based on dual-model is proposed.

  Methods  A relative motion model and two-leader state space model are established. The velocity information of the following AUV is estimated using the relative motion model in underwater acoustic communication, and then the multi-AUV coordinated positioning state space model with the two-leader model is applied to further improve the robustness and accuracy of the coordinated positioning system. Semi-real simulation experiments are then carried out using sea test data.

  Results  The results show that the leader-follower multi-AUV cooperative localization method based on the dual-model can estimate the position of the following AUV in real time without INS or DVL on the AUV.

  Conclusions  This method can ensure that the positioning accuracy of the cooperative localization system is within the allowable range, and reduce the cost of the multi-AUV cooperative localization system.