In order to reduce the deviation between the finite element model of helicopter airframe and the actual model， a helicopter airframe model updating method based on the response surface method is adopted， which uses a hypersurface function to approximately replace the actual and implicit complex function relationship， avoiding the traditional parameter updating method of calling the finite element program in each iteration. Meanwhile， the computational efficiency is improved by reducing the number of sample points through Latin hypercube sampling. Taking the 10% scaled down helicopter fuselage model as an example， the first six natural frequencies of its actual model and the finite element model are obtained through experiments and finite element analysis. The design parameters are determined by sensitivity analysis and the response surface method is used to update the parameters of the fuselage model. The average error of the natural frequencies between the predicted results of the finite element model and the experimental test results is reduced from 3.54% to 1.03%， effectively improving the prediction accuracy of the finite element model. The revised results show that the finite element model updating method based on the response surface method can be well applied to the parameter updating of helicopter airframe model.