Nonlinear Model-based Position Control for Quadrotor UAVs

Position ontrol of Mi ro Air Vehi les (MAV) is hallenging, be ause position measurements by global navigation satellite systems or laser s anners are typi ally available at mu h lower rates than the ontrol frequen y. Furthermore, the transient response of lassi PID ontrollers is either slow or indu es overshoot. In this work, we address this issue by a model-based ontrol approa h. We model and identify the dynami s of the MAV and use this knowledge in a nonlinear as aded ontroller to generate time-optimal traje tories. The proposed method is evaluated in simulation and two real MAVs.