Observer-based adaptive neural network robust H<SUB align="right">∞ control of quadrotor aerial robot

This paper presents an approach utilising observer-based adaptive neural networks to achieve robust H∞ control in an autonomous quadrotor system. The methodology addresses parameter uncertainties and wind disturbances for enhanced control performance. By integrating a state observer design, our method eliminates the need for direct access to state variables. We employ adaptive neural networks, specifically radial basis function neural networks, to approximate unknown functions through state estimates. The proposed methodology ensures H∞ tracking performance, addressing cumulative uncertainties from unmodelled dynamics, approximation errors, and external disturbances. Leveraging Lyapunov stability theory, we rigorously establish the uniform ultimate boundedness of both the observer-based controller and the closed-loop system. Simulation results are provided to validate and underscore the effectiveness of this approach.