Utilização de redes neurais no controle da velocidade de um veículo experimental

TL;DRAbstract

This work studies the application of projects that use neural networks for identification systems and control of speed, to make the system control more robust to the parametric and measure variations, and mainly to compensate the non-linear effect of the gain related to the inherent range of operation to the speed systems.For simulation and test an automodel with a DC motor was used.Several libraries of the Matlab/Simulink were used in the control stations to assist the activities of simulation.The results show an excellent performance of the RNA's in the identification of non-linear elements, and promise a significant reduction of the error of the predicted value of the speed.The results gotten with the neural controller for linearization feedback were acceptable and enough for our objective.

Chat with Paper

AI Agents for this Paper