In this paper, a modified sequence quadratic programming method with nonmonotone line search technique is presented. The algorithm has no demand on initial point, moreover it avoids using a penalty function or a filter. So it is more flexible and easier to implement.  To avoid Maratos effect, a revised direction is computed by solving a linear system. Under some reasonable conditions, the global convergence is shown.

constrained optimization, KKT point, sequential quadratic programming, global convergence.