Adaptive dynamic control for trajectory tracking with a quadrotor

Abstract

This work proposes an adaptive dynamic controller for an unmanned aerial vehicle (UAV) to track a desired trajectory. Initially, reference velocities are generated by a controller that is based only on the kinematic model of the UAV. Subsequently, new control actions are calculated to compensate for the internal dynamics of the robot. Then the model parameters that characterize the robot dynamics are updated during navigation, characterizing an adaptive controller. In this way, the performance of the flight application with the quadrotor is improved, since the control errors are minimized. The stability of the proposed control system is proven, based on the Lyapunov theory. Finally, simulated results are presented, demonstrating the good performance of the controller even without any previous knowledge of the values of the parameters of the UAV dynamics.