Linear Parameter-Varying Controller Design for a Nonlinear Quad-Rotor Helicopter Model for High Speed Trajectory Tracking

This paper presents an approach to position control of a quadrotor helicopter using Linear Parameter-Varying (LPV) synthesis techniques. The use of LPV control proves beneficial when tracking high speed maneuvers that require large tilt angles. For larger deviations from the hover position, a linear approximation is no longer adequate and the performance resulting from Linear Time-Invariant (LTI) controllers is negatively impacted and can even lead to instability. The proposed method uses the tilt angles as scheduling parameters for the position subsystem and the yaw rate for the orientation subsystem while compensating for the yaw angle via a rotation of reference and feedback signals, thereby simplifying the model and allowing for high accuracy approximation in a large operating envelope. Contrary to previous LPV control approaches, no Jacobian linearization takes place, which enables larger tilt angles and therefore faster trajectories to be followed.