A variable universe fuzzy control algorithm based on fuzzy neural network

In the conventional variable universe fuzzy control system, it was hard to select the function models of the contraction-expansion factors and parameters of the models. A new variable universe fuzzy control algorithm, to achieve the contraction-expansion of universes by fuzzy neural network in place of the function models, was proposed in this paper. The algorithm, with the semantic characteristics of fuzzy inference and learning ability of neural network, can reasonably adjust universes. Finally, the control algorithm is applied to a quadruple inverted pendulum system. The experiment results show that the proposed algorithm can avoid choosing the function models of contraction-expansion factors and parameters of models, and improve response speed and control precision of the controlled system.