Robust vision-based control of an underactuated flying robot tracking a moving target

Abstract

In this paper a robust image-based visual servo (IBVS) scheme is presented for a quadrotor tracking a moving target. The objective is to consider the full dynamics of the system to design vision-based controllers for the translational motion and the yaw rotation of this unmanned aerial vehicle. Passivity properties of the dynamics of the image features obtained from perspective image moments in a virtual image plane are used to solve the problem of designing full dynamic IBVS controllers for these robots. Nonlinear robust controllers are designed to deal with uncertainties in the dynamics of the image features and also uncertainties in the dynamics of the robot. The controllers need neither 3D information of the target model nor yaw information of the robot. Stability analysis guarantees that the states of the system are uniformly ultimately bounded. Simulation results are presented to validate the designed controllers.