MUDE-based control of quadrotor for accurate attitude tracking

Abstract In this paper, a modified uncertainty and disturbance estimator (MUDE)-based attitude controller for quadrotors based on a high-accuracy actuator dynamic model is proposed. The actuator dynamics are approximated by a first-order plus time-delay dynamic model, and both static and dynamic models of the actuator are identified using a motor test platform. The actuator model is further exploited as a virtual sensor to provide the actuator-generated torque estimates for the controller. By these estimates, an MUDE is constructed to compensate for not only the uncertainties and disturbances in attitude dynamics but also the time delay and disturbances in actuator dynamics. The systematic consideration of actuator disturbance rejection significantly improves the overall attitude tracking performance during maneuvers, which is clearly distinguished from other existing solutions assuming ideal actuators. Furthermore, the stability and performance analysis of the closed-loop system is presented, showing that the estimation error and the closed-loop tracking error can be reduced by tuning a single parameter. Finally, the proposed control, as well as classic uncertainty and disturbance estimator-based control and cascade PID controller, are implemented and compared on a real quadrotor platform. The experimental results show the effectiveness and advantage of the proposed control method with respect to signal tracking and disturbance rejection.