Validation of machine learning approaches for estimating wheel fatigue loads at the front suspension of a race car during track driving

The wheel loads of a race car have been estimated in view of structural dura-bility assessments. First, the front left double-wishbone suspension of a rear-wheel-drive race vehicle has been instrumented; then, wheel loads have beenestimated by means of four approaches: (i) a geometric matrix (GM) method,(ii) a feedforward neural network (FNN) approach applied to the fully instru-mented suspension (FIS), (iii) a FNN approach involving a reduced number ofsensors (the partially instrumented suspension (PIS)) and an inertial measure-ment unit (IMU), and (iv) a linear modeling approach (LM). After havingtrained the FNNs by using suspension signals acquired in a racetrack as inputsand related tire forces measured with the GM method as targets, the FNN-based methods have been validated on three different racetracks by comparingthe estimated loads with those estimated by means of the GM method. Accord-ing to the results achieved, the FNN approaches are effective for the estimationof the wheel forces.