Creating swing-up patterns of an acrobot using evolutionary computation

Abstract

In recent years, the various controller design schemes have been examined for nonlinear mechanical systems with uncertainties. As one of such mechanical systems, acrobots attract attention in last decade, which belong to nonholonomic systems. Several controller design schemes have been proposed for the acrobots. In this paper, a new approach to control the acrobot is considered, in which a suitable swing-up pattern is created using the real-coded genetic algorithm(GA). Then, the energy and the extra motion which are totally consumed in the swing-up motion, are evaluated on the fitness function. The effectiveness of the newly proposed scheme is numerically verified on some simulation results.