Flight Testing, Data collection, and System Identification of a Multicopter UAV
Over the last few decades, multicopters, a type of Unmanned Aircraft Vehicles (UAVs), have become increasingly popular for a wide variety of missions ranging from surveillance of disaster hit areas to precision agriculture. With an overall reduced size and elimination of a human pilot onboard, UAV technologies are easily accessible but also have certain limitations in comparison to manned airplanes and helicopters. Due to their ability to fly at slower speeds and agility, muticopters offer several advantages for many applications. However, multicopters are a recent addition to aerospace vehicles, and theory of their flight dynamics is not fully developed. In order to utilize their full potential, it is essential to design control systems that increase their performance and improve handling qualities. This, in turn, requires an accurate flight dynamic model of the vehicle. Due to their complexity and lack of fully developed theory, system identification techniques are required for the identification of flight dynamics models of multicopters. This presentation talks about the identification of the bare airframe flight dynamics model of a Matrice 600 multicopter in hover using frequency response method. The vehicle is flown for frequency sweep flight data, which is then converted to frequency response using CIFER (Comprehensive Identification from FrEquency Response) software. The frequency responses are then used to identify the parameters of transfer function and state-space models. The identified models are verified using separate set of flight data. The presentation will show and discuss the identify models and simulation results.