Genetic Optimization of Fuzzy Sliding Mode Controllers: An Experimental Study
Mariagrazia Dotoli, Paolo Lino, Bruno Maione, David Naso, Biagio Turchiano · 2003
Fuzzy Sliding Mode (FSM) techniques are successful in controlling nonlinear plants and reducing both control action and computational effort. However, their design is non-trivial, since it involves choosing the sliding parameter affecting the overall control speed, the input/output scaling gains, as well as the membership functions of the signal labels. This paper describes a procedure applying genetic algorithms to optimize the design of FSM controllers. Besides some simulation results, various laboratory experiments are reported, where the designed controllers are applied to the stabilization of an inverted pendulum. We employ the proposed methodology with the objective of stabilizing the pole in the upwards unstable position while simultaneously controlling the connected cart and minimizing the settling time, the cart travel and the required control action. Several conclusions are drawn out, with regard to the controller complexity and the system performance. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.