Traction control oriented torque-to-slip identification for powered two-wheelers

In this paper a torque-to-slip model, oriented to the development of a traction control system based on wheel slip control for powered two-wheelers, is identified. A black-box identification approach is employed to find model parameters, using I/O experimentally measured data. Different tests are employed to analyse different features of the inquired dynamics. In particular, a PRBS based identification unearths an average model; step responses are employed to study the effect of the vehicle lean angle and narrow band sine sweeps better quantify the features of an underdamped mode. The resulting model is thus a family of linear models with lean angle and velocity as scheduling parameters. The identified model is proven accurate and suitable for control system design, thanks to a closed-loop experimental validation.