In the acquisition process of the optimal estimation of ballistic parameters, Calman filtering is a strong engineering application and is used widely. A state initial value estimation method base on polynomial fitting was proposed to mitigate the adverse effects of initial state of filtering on Calman filtering process. This method obtains the initial value of filter state through the selection and fitting estimation of pre filtering ballistic data. In the filtering process under the initial value of this estimated state, the speed of convergence is faster and the fluctuation of the filter residuals is gentle. The simulation results show that the estimation method can shorten the acquisition time of the optimal ballistic parameters by 50%.