The main difficulties in modeling yaw dynamics of a helicopter arise from the high nonlinearities,cross-couplings and dynam-ic uncertainties of these aerocraft.This paper proposes a new identification approach for yaw dynamics modeling through modes partition method(MPM) with a concentrated search space limited by implicit human factors.Working from first princi-ples and basic aerodynamics,the nonlinear equations of motion for yaw dynamics is derived.The model is linearized and transformed into a combination of dynamic modes,whose coefficients are identified from real-flight data through distributed genetic algorithm(DGA).The effectiveness of the approach is validated in terms of the identified model which can accurately capture the dynamic characters of the helicopter.Time-and frequency-domain results clearly demonstrate the potential of MPM in modeling such complex systems.
