Valerio Cubeddu
Preview-based traction control using self-adaptive artificial intelligence algorithms.
Rel. Aldo Sorniotti, Fabio Alberti, Gianluca Frison. Politecnico di Torino, Corso di laurea magistrale in Ingegneria Meccanica, 2026
Abstract
Road irregularities do not only produce vertical vibrations. They also cause oscillations of the longitudinal acceleration, which are smaller but still important for passenger comfort. Traditional solutions work on the suspension and therefore act only on the vertical direction. In-wheel motors offer a new possibility, because their torque is fast and can be controlled at each wheel separately. This thesis presents a nonlinear model predictive controller (NMPC) that reduces these longitudinal oscillations and, at the same time, limits the wheel slip when the road has low friction. The controller uses a quarter-car internal model and receives information about the road profile ahead.
Since the powertrain of the test vehicle has a pure delay of about 77 ms, this delay was included in the internal model through a state augmentation
Relatori
Anno Accademico
Tipo di pubblicazione
Numero di pagine
Informazioni aggiuntive
Corso di laurea
Classe di laurea
URI
![]() |
Modifica (riservato agli operatori) |
