Parameter Optimization of Genetic Algorithm Utilizing Taguchi Design for Gliding Trajectory Optimization of Missile

Abstract

The present study aims to establish a genetic algorithm (GA) method to optimize gliding trajectory of a missile. The trajectory is optimized by discretizing the angle of attack (AOA) and solving optimal control problem to achieve maximum gliding range. GA is employed to resolve the optimal control problem to achieve optimized AOA. A Taguchi’s design of experiments was proposed contrary to full factorial method to ascertain the GA parameters. The experiments have been designed as per Taguchi’s design of experiments using L27 orthogonal array. Systematic reasoning ability of Taguchi method is exploited to obtain better selection, crossover and mutation operations and consequently, enhance the performance of GA for gliding trajectory optimization. The effects of GA parameters on gliding trajectory optimization are studied and analysis of variance (ANOVA) is carried out to evaluate significance factors on the results. Crossover function and population size are observed as highly impacting parameter in missile trajectory optimization accompanied by selection method, crossover fraction, mutation rate and number of generations. Artificial neural network (ANN) method was also applied to predict the significance of GA parameters. The results show that the gliding range is maximized after GA parameter tuning. Simulation results also portrayed that with optimal AOA, gliding distance of missile is improved compared to earlier one. The numerical simulation shows the efficiency of proposed procedure via various test scenarios.