Side-Slip Angle Estimation by Artificial Neural Networks for Vehicle Dynamics Control Applications

Abstract:
State estimation is a major aspect of vehicle dynamics control systems. Whenever states cannot be measured, they have to be estimated in order to control them. In addition to state estimators based on physical model knowledge, data-driven models are used to a greater extent. Especially artificial neural networks exhibit a great potential to increase the estimation quality. In the present contribution, an artificial neural network is set up to estimate the side-slip angle. The knowledge about the side-slip angle is required in order to affect the self-steering behavior within a central predictive control. In the validation, the artificial neural network is compared to a physical model for estimating the side-slip angle.