Multiple fuzzy logic systems: a controller for the flexible pole-cart balancing problem

This paper examines the applicability of fuzzy logic based algorithms to the control of complex and highly nonlinear systems. An online fuzzy logic controller has been developed to control a cart balancing a flexible pole under its first mode of vibration. The controller design includes 5 fuzzy logic systems (FLS) and a single rule based evaluator that centers the pole on the truck. Results of physical experiments show that the controller not only balances the flexible pole indefinitely but also brings the cart to the center of the track. The controller can also easily adapt to disturbances from the external environment (e.g. moving or shaking the track randomly, elevating the height of the track on either side, pushing the pole in any direction, preventing the pole from moving further by putting an obstacle in its path). The operation of the system can also be initialized anywhere in the track. The controller is sufficiently fast to balance the system from an initial angle of 20 degrees