Real-time validation and comparison of fuzzy identification and state-space identification for a UAV platform

Unmanned aerial vehicles (UAVs) have been playing an increasingly important role in military and civilian applications. Identification of UAV model is an important process in the controller design. In this paper, identification of the attitude dynamics of UAV is investigated. Two different identification techniques for attitude dynamics of UAV are applied, verified and compared together. The first method is based on an error mapping approach, while the second one is based on fuzzy system approach. The main features of the two identification methods are discussed and compared. The identification algorithms are programmed onto the microcontroller and a real time validation was performed using the in-house developed hardware in loop simulation (HIL) tool. The performance of both identification approaches is evaluated based on the flight data. Real time simulation results show that the fuzzy identification approach is better than error mapping approach