Bio-inspired Control Based on a Cerebellum Model Applied to a Multivariable Nonlinear System

In this paper, a novel bio-inspired control system is applied over a nonlinear multivariable system in real time. The control system is based on a bi-hemispheric neural network and a generalized polynomial controller. A ball and plate system is selected to perform the experiments since it adequately describes the brain interaction performed through the cerebellum that implies the sensor (human eyes or artificial vision system) and the actuator (human motor system or servo-motors related with the angles of the plate). The proposed method is evaluated by analyzing the settling time and steady-state error in comparison with a nonlinear controller obtained by using state feedback control law which is computed through extended linearization. The proposed approach is evaluated in a simulated and real environment, where obtained results show that the proposed method reduces the steady-state error of the nonlinear multivariable system.