Dynamic fuzzy control and system stability for the Acrobot

This paper discusses the swing-up control of a two link robot mechanism known as the Acrobot. Previously, a technique using meta-rules was proposed to compensate for external disturbances when using a dynamic fuzzy controller. In this paper, stability of the meta-rule system is analyzed from a global asymptotic stability viewpoint based on fuzzy characteristic matrices. These matrices are generated from the first phase of a three phase learning process involving the combination of genetic algorithms, dynamic switching fuzzy systems, and meta-rule techniques with the purpose of developing a better performing fuzzy controller.