<p>Traditional linearization methods without update and static terminal sets in model predictive control (MPC) result in conservative robust constraint designs, limiting tracking performance in nonlinear autonomous surface vessels (ASVs) systems. To solve this problem, this study proposes a novel resilient and robust MPC framework with dynamic linearization and adaptive terminal sets under asynchronous denial-of-service (DoS) attacks. First, dynamic feedback gains are introduced into the predictive control input sequence, compensating for input information loss due to DoS attacks. Then adaptive terminal sets are designed into the terminal constraint of the MPC optimization problem. Second, considering the change rate of disturbance, a novel state estimator is designed with predicted disturbance to reconstruct the lost state information due to DoS attacks. Moreover, to reduce the computational burden of the MPC controller, an event-triggered mechanism is designed to limit the frequency of solving the MPC optimization problem. It is demonstrated that the proposed MPC mechanism is both recursively feasible and stable in the presence of external disturbances and asynchronous DoS attacks. Numerical simulations demonstrate the efficacy of the proposed algorithm.</p>