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. 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 .