Adaptive TLBO-Based Fuzzy Logic Controller for Quadrotor Trajectory Tracking

This paper presents an Adaptive Fuzzy Logic Controller (AFLC) optimized using the Teaching-Learning-Based Optimization (TLBO) algorithm for controlling a quadrotor-type autonomous aerial vehicle (AAV) operating under varying perturbations. In the proposed AFLC-TLBO approach, the fuzzy rules and membership functions of the fuzzy logic controller (FLC) are dynamically updated to maintain high performance under external disturbances. A reference model is used to generate the desired quadrotor outputs, from which an objective function is computed based on tracking errors. When this function exceeds a predefined threshold, the TLBO algorithm is activated to retune the FLC parameters in real time. The controller performance is evaluated under multiple disturbance scenarios and compared with a conventional FLC. Simulation results demonstrate significant improvements in disturbance rejection capability. In altitude control, the proposed method reduces the peak z-error by 92% and 67% during the first and second perturbations, respectively, while decreasing the recovery time by approximately 95% and 29%. In two disturbed trajectory-tracking scenarios, the proposed method achieves overall RMSE reductions of 42% and 33%, respectively, relative to the standard FLC. These results show that the proposed method significantly enhances disturbance rejection and confirm its suitability for real-time control of nonlinear systems under dynamic conditions.