Constrained Stochastic Tournament Selection in Flight Control Problems

Engineers in some practical design engineering domains are reluctant to incorporate GAs into their procedures because of their execution time and the closeness of their results to the global optimum. Some works on the design of a fly-by-wire control system, have sought to improve the performance of GAs for the efficient and effective computation of the controller gains to ensure optimal aircraft performance over its full flight envelope. The present work reports the design of a new tournament selection method for the application of a GA to a non-constrained optimization model of the controller gains problem. The results show the advantage of the approach over the application of other GAs architectures to an unconstrained optimization model of the controller gains design problem