Adaptive fuzzy control for balancing double inverted pendulums

The stabilization control of a double-inverted-pendulums(DIP) system connected by a spring is considered in this paper. The DIP system is a complicated, nonlinear, unstable two-input two-output system with strongly coupled interconnections. Takagi-Sugeno(T-S) fuzzy systems are used to deal with the unknown system uncertainties. Combining “dynamic surface control(DSC)”approach with “minimal learning parameters(MLP)” algorithm, a novel decentralized robust adaptive control scheme with simple structure and lightened computation load is developed, which is easy to be implemented in applications with only one online-learning parameter in each subsystem and guaranteed stability. In addition, the potential controller singularity problem is removed. Simulation results for balancing the DIP system validate the effectiveness and excellent transient performance of the proposed scheme.