Design and implementation of a novel fuzzy controller with dsp for rotary inverted pendulum

Fuzzy controllers are used in many applications because of their rapid design by translating heuristic knowledge, robustness against perturbations, and smoothness in the control action. It is known that as the number of membership functions increases fuzzy control becomes smoother. However, since the number of rules is an exponential function of the number of inputs and number of membership function, the methodology is difficult to implement when the number of rules is very large. To overcome such a limitation, this paper proposed an information fusion based fuzzy control algorithm, which fuses the angle and velocity as one input to reduce the complexity of the controller and enhance the robustness. By doing so, the fuzzy controller is simplified and the RAM memory requirements are reduced. In order to test the performance of the proposed algorithm, a DSP based rapid control prototyping system is also developed to facilitate the development of real-time control system. Real-time control experiments with Quanser´s rotary inverted pendulum demonstrated the efficiency of the designed fuzzy controller embedded into a stand-alone processor based on a fixed-point DSP.