Intelligent Control of an Inverted Pendulum by Self-Tunable Fuzzy PI-type Controller
Djamel Eddine Chaouch, Hichem Maaref · 2011
The work describes an automatically online Self-tunable PI-type fuzzy inference system with application of control approach to nonlinear system.In this paper, we describe a neuro-fuzzy controller as a STFIS (Self Tuning Fuzzy Inference System) optimized online using the rules of Takagi-Sugeno type, and a back propagation learning algorithm which is used to minimize a cost function that is made up of a quadratic error term and a weight decay term that prevents an excessive growth of parameters.Then STFIS PI-type controller is synthesized.The application is done on a system of inverted pendulum to follow a predefined reference trajectory in the presences of disturbance.Some simulations with robustness tests are performed, which demonstrates the feasibility of the proposed control strategy.Results of simulations containing tests of robustness are presented and realized in MATLAB environment.