Rotorcraft Parameter Identification from Real Time Flight Data

Rotorcraft system identification involves the derivation of the model structure and parameters, such as aerodynamic stability and control derivatives, from flight-test results. Accurate rotorcraft system identification often requires higher order mathematical models. However, system identification becomes difficult and complicated as the number of modeling degrees of freedom are increased. In the present work, a new method for rotorcraft parameter estimation based on the application of a radial basis function network is proposed. The radial basis function networkbased technique does not require a mathematical model of the helicopter, and the rotorcraft parameters can be directly computed from the flight data. The radial basis function network is found to give results in the same range as obtained from conventional parameter estimation techniques, such as the maximum likelihood method. Results obtained using the radial basis function network approach are compared to published research on the BO 105 helicopter and found to be in good agreement.