Robust Identification Algorithm for Unmanned Underwater Vehicles Dynamics Model Parameters

Abstract

As the exploration of marine resources continues to deepen, the Unmanned Underwater Vehicles (UUV) have gradually become the focus of industries such as military, fishery, seabed prospecting and marine platform monitoring. At present, the identification technology of the dynamic model parameters of the UUV is not mature yet, and the robustness is poor. In order to solve the problem that the identification algorithm is susceptible to underwater noise when the robot is navigating underwater, this paper integrates the Huber loss function into the recursive least square method, and proposes a robust identification algorithm to improve the control of the underwater robot Accuracy. Finally, the dynamic model of the underwater vehicle is simulated in Simulink to obtain a reasonable input signal, and then this algorithm is used to simulate and identify the model under Gaussian noise with ‘outliers’. The simulation results show that the algorithm has strong robustness while ensuring the identification accuracy.