Experimental Modeling of a Quadrotor UAV Using an Indoor Local Positioning System

This paper reports the results obtained from an experimental modeling of a vertical take-off and landing type unmanned aerial vehicle (UAV). The considered UAV is an ar.drone quadrotor UAV which consists of a rigid body with fours arms, each of which is mounted with a rotor for motion and flight control. To complement onboard sensors installed on the UAV (i.e. ultrasonic sensor & IMU), an indoor local positioning system is used in the experiment to obtain the global positioning data of the quadrotor UAV. The modeling is performed using system identification technique whereby the UAV is subjected to certain inputs and its real-time output is observed and then used to estimate the model. From the experiment, it was found that a fifth order linear state space equation model captures the dynamics of the considered quadrotor UAV sufficiently good.