Model Generation and Performance with CIFER

This paper describes the generation of a mathematical model for the Thunder Tiger Raptor 50 V2 remote control helicopter using a Comprehensive Identification method based on FrEquency Response (CIFER). The scope of this paper is to point out the advantages and the problems of creating a canonical helicopter model for the purpose of flight simulation and robust algorithm design with a method based on the frequency response of the system. This is compared with a more traditional approach using a linear parameter varying model (LPV). With CIFER, it was possible to develop accurate and simple models of the helicopter which sufficiently predicts the actual helicopter response. Three different models were obtained. One is a singleinput single-output transfer function hover model, either simulating the pitch rate, roll rate, or yaw rate response. The second model is a six degree of freedom multiple-input single-output state space model of the Raptor 50 V2 in hover. The third model is a six degree of freedom multiple-input single-output state space model of the helicopter in forward flight. The responses from the models are compared with actual flight data. The similarity of the simulation with responses implies a great capability of the model to predict the helicopter response.