Neuro-adaptive Tracking Control for Uncertain Quadrotor UAVs with Guaranteed Performance

This paper presents a neuro-adaptive tracking control framework for uncertain quadrotor UAVs, integrating sliding-mode PID and neural networks to address model uncertainties and disturbances. The controller dynamically tunes PID gains via radial basis function networks while enforcing predefined performance constraints using Barrier Lyapunov Functions. A hierarchical architecture stabilizes attitude and ensures bounded trajectory tracking. Stability is proven via Lyapunov analysis, with adaptive fault tolerance. Simulations under gusts, sensor noise, and actuator faults demonstrate superior tracking precision and fault accommodation. This work offers an efficient solution for UAVs operating in uncertain, fault-prone environments.