The use of an improved genetic algorithm to solve a constrained cruise missile routing problem is explored in two-dimensional space. In modern war, the threat areas with which the cruise missiles confronted were represented as circles. A route was defined as a series of points through which the cruise missile sequentially travels from the starting point to the ending point. The physical constraints of total route length maximum angle and minimal step length were included and appear in the fitness function. In order to be valid, a route must go from start to finish without violating any constraint, and can't thrill through the threat circles. The computer simulation results indicate that the genetic algorithm can solve the problem of routing of the cruise missile low altitude penetration, and has a certain value of theory and application.