Genetic vector ordinal optimization algorithm based on RSM for UCAV attack planning

The problem of generating optimal air-to-ground attack plan for the UCAV is studied. In order to deal with the difficulties about time-consuming evaluation of the objective cost values and the large solution space, a genetic vector ordinal optimization algorithm (GVOOA) based on response surface methodology (RSM) is proposed. In this paper, several constraints including flight capability constraint, battlefield constraints and weapon constraint, are considered. And two optimal cost functions are built using Monte Carlo simulation criteria of the weapon attack operation, and then the attack planning problem is transformed into a simulation multi-objective optimization problem (SMOOP). In order to improve the convergence performance, an approximate model building approach based on RSM is presented, and a combining vector ordinal optimization (VOO) with NSGA-II is designed. The proposed approach is demonstrated using a typical air-to-ground attack mission scenario. The simulation results indicate that the proposed approach is better in the solving speed compared to the traditional approach, under the near optimal performance.