System Identification of Quadcopter by Flight Data Acquisition

This paper mainly focuses on evaluating a low cost high-frequency flight data acquisition system (FDAC) for system identification and aerodynamic data collection on small to mid-sized Rotary Unmanned Aerial Vehicles (RUAVs). The system is low weight, low power and operates up to 150Hz sampling rate. The data acquisition system is a low cost, open-source autopilot system called PixHawk and is available commercially off the shelf. The FDAC combines the large variety of sensor streams into a unified high-fidelity state data stream that is recorded for later system parameter identification and simultaneously forwarded to a separate processing unit, such as an autopilot. Evaluation of this FDAC is done by identifying a stable Quadcopter UAV platform. First the quadcopter dynamics are identified from the roll, pitch and yaw attitude data from the recorded flight data on an onboard memory card. The identification is done by using Instrumental Variable(IV) method. The comparison of the parameters identified from flight data and the actual quadcopter behavior will evaluate the FDAC. Then the FDAC can be used for the Parameter identification of any multi-rotor UAV.