Parameter optimisation of Genetic Algorithm Utilising Taguchi Design for Gliding Trajectory Optimisation of Missile

The present study aims to establish a Genetic Algorithm (GA) methodology to optimise the missile gliding trajectory. The trajectory optimisation was carried out by discretising the angle of attack (AOA), subsequent transformation of the optimal control problem to a nonlinear programming problem (NLP), and resolving the optimal control problem to attain a maximised gliding range. GA is employed for resolving optimal control problem. Taguchi design of experiments was proposed contrary to the full factorial method to ascertain the GA parameters. The experiments were designed as per Taguchi's L27 orthogonal array. The systematic reasoning ability of the Taguchi method is exploited to obtain better selection, crossover, and mutation operations, and consequently, enhance GA performance. An analysis of variance (ANOVA) is performed to evaluate the influencing factors in the results. Crossover function and population size are observed as impacting parameters in trajectory optimisation, accompanied by selection, crossover fraction, mutation rate, and number of generations. An Artificial Neural Network (ANN) approach was enforced to anticipate the significance of GA parameters. Based on Taguchi design of experiments, analysis of variance, and artificial neural network methods the optimal parameters of GA were selected. It is observed that the maximum gliding distance is achieved after GA parameter tuning. It is noticed from the simulation results that the missile gliding range is enhanced in comparison to earlier ones. The simulation results also show the efficiency of the proposed procedure through different test cases.