| description abstract | A genetic algorithm (GA), a well-known numerical method, is widely applied in different areas of optimal studies. It is found that if the solution-search space is wide or if the selected fitness function is highly nonlinear, the GA’s solutions can strongly depend on the set parameters, which include population size, crossover rate, mutation rate, and the remaining size of the parent in the GA. This paper combines the Taguchi experimental method, which serves as a rough search tool, with the GA, which serves as a fine search tool, to find the best combination of the GA parameters for different flight-control problems. The purpose of such a combination is to make control more robust and closer to the optimal solution. To demonstrate this new idea, the writers consider its application to different flight-control problems for the F-16 fighter by using autostabilization, linear quadratic regulator (LQR) and | |
