A robust MPC approach for the rebalancing of mobility on demand systems

A control-oriented model for mobility-on-demand systems is here proposed. The system is first described through dynamical stochastic state-space equations, and then suitably simplified in order to obtain a controloriented model, on which two control strategies based on Model Predictive Control are designed. The first strategy aims at keeping the expected value of the number of vehicles parked in stations within prescribed bounds; the second strategy specifically accounts for stochastic fluctuations around the expected value. The model includes the possibility of weighting the control effort, leading to control solutions that may trade off efficiency and cost. The models and control strategies are validated over a dataset of logged trips of ToBike, the bike-sharing systems in the city of Turin, Italy.