Partical Swarm Optimization Based Adaptive Fault-Tolerant Control for Quadrotor Attitude Stabilization

This paper introduces a novel approach for achieving optimal attitude control of a quadrotor in the presence of uncertainty, external disturbances, time-varying sensor faults, and nonaffine nonlinear actuator faults. The proposed method utilizes the Particle Swarm Optimization (PSO) technique to approximate adaptive parameters and fuzzy initial values, resulting in a reliable controller that can quickly compensate for changes in the starting point. Fuzzy systems are employed to estimate unknown nonlinearities, nonaffine nonlinear actuator faults, and time-varying sensor faults. To address approximation errors and external disturbances, a robust control term is incorporated. The issue of an algebraic loop is resolved using a Butterworth low-pass filter. Additionally, the proposed robust scheme effectively handles external disturbances without relying on approximations, and the controller is dynamically updated through online reconfiguration. The stability of the entire closed system is analyzed based on Lyapunov theory. Simulation scenarios are presented to demonstrate the efficacy and benefits of the proposed approach.