Pronóstico del área de contacto de los neumáticos de un vehículo vía redes neuronales recurrentes

In this paper, a predictive soft sensor via a Recurrent Neural Network is proposed for sensing the area between the tires and the terrain, which is one of the most important issues in the automotive industry, in fact, keeping a tire contact patch suitable ensures the comfort and maneuverability of the vehicle. The core of the soft sensor is an recurrent neural network, the training of a recurrent neural network is complicated for algorithms based on the gradient, because of the apparition of some valleys on the error surface. As an alternative, the training process is interpreted like an optimization problem and it is solved by an differential evolution algorithm. This soft sensor has been validated using a test rig based on the frustrated total internal reflection phenomenon. 171 Research in Computing Science 116 (2016) pp. 171–184; rec. 2016-03-14; acc. 2016-05-10