Real-time Fault Tolerant Attitude Stabilization Control of a Quadrotor in the Presence of Actuator Fault

In this paper, a finite-time fault tolerant controller based on sliding mode algorithm, as a robust control method, is presented to control the attitude stabilization of the quadrotor system in the presence of actuator fault and uncertainty. The controller is designed based on the nonlinear model of a quadrotor and its stability analysis is performed according to the Lyapunov stability theorem. Also, regarding some weaknesses of MEMS sensors such as partly high noise and bias error, an extended Kalman filter is designed and implemented in order to merge sensors data and reduce the noise effect on the outputs. To validate the controller performance, the experimental tests is implemented on a full-scale quadrotor in real-time. The evaluation of the designed strategy is carried out in different scenarios, no fault in the actuators and a partial loss of effectiveness of an actuator as well as uncertainty in the quadrotor parameters. The experimental results reveal the superiority of the sliding mode tolerant strategy over feedback linearization in the presence of various faults and uncertainty effects.