Joint state-disturbance observation with encoding-decoding mechanisms for uncrewed surface vehicles: A neural network-based approach

Key Points

• Enhanced observation of state-disturbance using advanced neural networks improves real-time predictions.
• The study shows a significant increase in prediction accuracy, achieving over 90% accuracy in testing.
• Adopting encoding-decoding mechanisms, this analysis incorporates various disturbances for better responsiveness.
• Future applications of this method may enhance the autonomous navigation of uncrewed surface vehicles.