A method for identifying the backbone curve parameters of the modified Ibarra-Medina-Krawinkler （IMK） model by using quasi-static test data of reinforced concrete columns and thus improving the simulation accuracy of reinforced concrete frame structures is proposed in this paper. In this method, a robust unscented Kalman filtering algorithm based on singular value decomposition（robust SVD-UKF algorithm） is introduced to suppress the influence of gross error of the observation on the parameter identification, and the particle swarm optimization algorithm is adopted to automatically optimize the initial covariance matrix, the process and measurement noise matrices. The identification of backbone curve parameters of the modified IMK model is realized using MATLAB，in which the symmetric and asymmetric hysteresis behavior of the columns in the positive and negative direction is considered. The effectiveness of the proposed method is verified by model backbone curve parameter identification based on the measured hysteretic curves of reinforced concrete columns and its application in the nonlinear simulation of frame structures.