Adaptive self-constructing fuzzy neural network controller for hardware implementation of an inverted pendulum system

A tracking control of a real inverted pendulum system is implemented in this paper via an adaptive self-constructing fuzzy neural network (ASCFNN) controller. The linear induction motor (LIM) has many excellent performances, such as the silence, high-speed operation and high-starting thrust force, fewer losses and size of motion devices. Therefore, the experiment is implemented by integrating the LIM and an inverted pendulum (IP) system. The ASCFNN controller is composed of an ASCFNN identifier, a computation controller and a robust controller. The ASCFNN identifier is used to estimate parameters of the real IP system and the computational controller is used to sum up the outputs of the ASCFNN identifier. In order to compensate the uncertainties of the system parameters and achieve robust stability of the considered system, the robust controller is adopted. Furthermore, the structure and parameter learning are designed in the ASCFNN identifier to achieve favorable approximation performance. The Mahalanobis distance (M-distance) method in the structure learning is also employed to determine if the fuzzy rules are generated/eliminated or not. Concurrently, the adaptive laws are derived based on the sense of Lyapunov so that the stability of the system can be guaranteed. Finally, the simulation and the actual experiment are implemented to verify the effectiveness of the proposed ASCFNN controller.