Application of Improved Genetic Algorithm in Cruise Missile Route Planning

Abstract

To address the shortcomings of the traditional genetic algorithm in cruise missile route planning, which is prone to “premature maturation” and premature convergence to a local optimal solution, an improved genetic algorithm is proposed that introduces an adaptive operator and a variation ratio strategy. The algorithm processes the individual fitness values in the population by ranking ratio technique, and then adopts a selection strategy combining elite selection and roulette algorithm to add the feasible routes with the best fitness values directly to the children at each evolution, and then roulette selects the remaining feasible routes, which improves the global optimal search performance of the algorithm in the trajectory planning. Meanwhile, an adaptive crossover operator is used to dynamically select the crossover probability based on the individual fitness values of the parents. Finally, the two algorithms are applied to the established map model for route planning separately, and the simulation results show that the path solved by the improved genetic algorithm reduces three planning waypoints and 10.74 % of the range compared with the traditional genetic algorithm, and the global search performance of the route planning process applying the improved genetic algorithm is significantly better than that of the traditional genetic algorithm.