The use of feedforward neural networks to cancel nonlinearities of dynamic systems

This work presents the experimental results of a nonlinear system subject to feedback linearization using artificial neural networks. The linearization function is directly estimated from actual input-output data.