Robust Control Design for Aircraft Controllers via Memetic Algorithms

This article develops a novel method based on the memetic algorithm for optimal feedback control design. The aircraft controllers are designed to produce desired handling qualities. The robust properties of the linear quadratic regulator are exploited to obtain the desired aircraft handling qualities with an output feedback constrained linear quadratic regulator subspace based on a cost function minimization method. Numerical examples of lateral autopilot design are optimized with the memetic algorithm. The memetic algorithm designs lateral feedback gains based on the quadratic performance index to maintain heading and roll attitude. The memetic algorithm is compared to the eigen-structure assignment, genetic algorithm and gradient-based optimization methods [21] and genetic algorithm. The memetic algorithm can derive a better solution with fewer iteration steps. Most importantly, the memetic algorithm can obtain a global optimum using mutation and thus avoid being trapped in local optima. The simulation and experimental results demonstrate that the proposed method produces useful, feasible and good robustness.