Observer-Based Time-Varying Backstepping Control for a Quadrotor Multi-Agent System

This paper studies the formation tracking problem for a multi-agent system composed by a set of quadrotor UAVs. Parrot AR.Drone 2.0 quadrotors are used as agents of the system. The AR.Drone 2.0 features an internal controller to stabilize the angular dynamics. This controller is modelled and identified through the least squares method. The proposed control strategy is designed using a time-varying version of the backstepping technique for each agent. For the implementation of the control law, it is assumed that each agent measures only the leader and its own positions, while the leader also knows the desired trajectory the system must follow. Linear and angular velocities of the agents are estimated using suitable Luenberger observers. The proposed control strategy allows the leader agent to converge asymptotically to a predetermined flight trajectory while the follower agents converge asymptotically to their own trajectories defined by the leader position and a constant formation vector. The theoretical results are validated through real-time experiments.