Neuro-Fuzzy-Based Adaptive Sliding Mode Control of Quadrotor UAV in the Presence of Matched and Unmatched Uncertainties

Sliding Mode Control (SMC) is a popular nonlinear controller for quadrotor UAVs due to its robustness, fast response, and ability to handle complex dynamics. However, it suffers from chattering, sensitivity to modeling errors, and poor performance in the presence of unmatched uncertainties. In this paper, a novel Neuro-fuzzy adaptive sliding mode control is designed and proposed for the position and attitude control of quadrotor UAVs. The proposed method combines SMC with the learning capabilities of Artificial Neural Networks (ANN) and the decision-making abilities of Fuzzy Logic Control (FLC). Firstly, the quadrotor flight dynamics is derived using Newton's quaternion formalism. Secondly, conventional SMC is designed, and the system's stability is validated using Lyapunov stability analysis. Finally, the designed SMC equivalent control part is estimated online by ANN, while its switching control part is estimated by FLC. To verify the controller's performance, extensive software-in-the-loop simulations have been conducted in various scenarios. The results show that the proposed controller effectively tolerates matched and unmatched uncertainties and has better tracking and disturbance rejection capabilities with minimal control effort compared to fuzzy-based SMC and conventional SMC. Therefore, the suggested controller is very suitable for quadrotor UAV applications that require high tracking precision despite varying operating conditions.