Online adaptive parameter identification of an unmanned surface vehicle without persistency of excitation

This paper is concerned with the parameter identification of an unmanned surface vehicle (USV) with fully unknown coefficients. An online adaptive parameter identification method is proposed for the USV to identify its model parameters without the condition of persistence of excitation. Specifically, a composite adaptive update law is developed based on a truncated integral filtered regressor, such that the model parameters can be estimated online, adapting to possible parameter changes. Then, an extended state observer is used to monitor the total estimation errors caused by environmental disturbances and parameter estimation errors. A salient feature of the proposed method is that the acceleration information is not required and only linear velocities and yaw rate are used for identification. Moreover, the convergence of the estimation errors is assured under the condition of interval excitation only. The stability of the online parameter identification method is proved by Lyapunov stability analysis. Simulations and experiments are carried out to demonstrate the effectiveness of the proposed online identification method for the USV.