An iterated filtering method is presented to improve the update stage of nonlinear filtering. First, we develop a generalized iterative framework for sigma point kalman filter by utilizing Gauss-Newton algorithm. A simplified iterated square root cubature kalman filter (SCKF) is proposed with application to tightly coupled GNSS/INS, where the linear combination of innovations is used in measurement update. Numerical experiment and field test results indicate that SCKF has similar performance with extended kalman filter. Compared with the non-iterated methods, iterated SCKF improves the heading by 23.6% with two iterations, and get a faster convergence rate regarding heading and velocity.