Genetic algorithm based LQR control of hovercraft

Abstract

A hovercraft can move on any rough surface or water, it uses propulsion force of air and is supported by cushion that has high compressed air in it. Hovercraft is suitable for many military and ordinary purposes. In this paper a LQR controller is designed for hovercraft. The performance of LQR depends upon selection of Q and R matrix the optimal value of Q and R is found using genetic algorithm with a new method based on function of first derivative of position is used as a fitness function. The results show that the fitness function not only stabilizes system very efficiently but the step response is also very satisfactory.