In this paper, sparse collocation approach is used to develop optimal feedback control laws for Optimal Control Problems (OCPs) involving the terminal state constraints. The effective collocation process is accomplished by utilizing the recently developed Conjugate Unscented Transformation to provide a minimal set of collocation points. In conjunction with the minimal cubature points, an l1 norm minimization technique is employed to optimally select the appropriate basis functions from a larger complete dictionary of polynomial basis functions. Finite time attitude regulation problem with terminal constraint is considered. Numerical simulations demonstrates the effectiveness of the proposed approach in derving the feedback laws for terminally constrained OCPs.