Recurrent Neural Network Model for On-Board Estimation of Side-Slip Angle

Measuring a vehicle’s side-slip angle can be useful to analyze the angle between the vehicle’s longitudinal axis and its speed direction (known as its “lateral dynamics”). Determining this value can lead to further insights regarding the vehicle’s stability, understeer and oversteer conditions, lateral forces during cornering, or tire wear estimation.

In their article titled “Recurrent Neural Network Model for On-Board Estimation of the Side-Slip Angle in a Four-Wheel Drive and Steering Vehicle,” Tiziano Alberto Giuliacci, Stefano Ballesio, Marco Fainello, Ulrich Mair, Dr. Julian King present a methodology for estimating the side-slip angle of a vehicle in real-time using virtual sensors and a recurrent neural network model. Their machine learning model was trained and tested using on-road experimental data, and its broader application may be present a cost-effective and reliable option for estimating the side-slip angle in standard vehicles, opening further avenues for optimization.