Ubiquitous sensing in smart cities enables large-scale multi-source data collected in real-time, poses several challenges and requires a paradigm-shift to data-driven cyber-physical systems (CPSs) that integrates optimization, control and machine learning. For instance, how to capture the complexity and analyze the dynamic interplay between urban-scale phenomena from data, and take actions to improve service efficiency and safety, is still a challenging problem in transportation systems. My work about real-time robust resource allocation problem in smart cities falls in the intersection of machine learning, optimization and control. We design a data-driven dynamic robust optimization framework for autonomous ride-sharing and carpool systems, matching vehicle supply towards both current and predicted future demand. With spatial-temporal uncertainty of demand prediction, we prove and develop computationally tractable methods that provide probabilistic guarantees for the system’s worst-case and expected performance. Dynamic pricing and hierarchical carpool algorithms are also designed for travel time reliability during peak hours. We show that the performance of the ride-sharing system is improved based on world taxi operational data. For electric vehicles such as E-buses and E-taxis, we consider both the mobility pattern and charging pattern to co-design the charging and passenger pick-up algorithms. Heterogeneous data, uncertainties of the dynamic status of the EV system, tradeoff among objectives and benefits of different components of the system are the main focus for current work.
"CPS: Small: COLLAB: Improving Efficiency of Electric Vehicle Fleets: A Data-Driven Control Framework for Heterogeneous Mobile CPS", funded by NSF.
[C1] Data-Driven Distributionally Robust Electric Vehicle Balancing for Mobility-on-Demand Systems under Demand and Supply Uncertainties (IROS2020) [link]
[J1] Data-Driven Distributionally Robust Electric Vehicle Balancing for Autonomous Mobility-on-Demand Systems under Demand and Supply Uncertainties (TITS under review)
Oral presentation at IROS 2020
https://www.youtube.com/watch?v=OQqytc2zTEE
As electric vehicle (EV) technologies become mature, EV has been rapidly adopted in modern transportation systems, and is expected to provide future autonomous mobility-on-demand (AMoD) service with economic and societal benefits. However, EVs require frequent recharges due to their limited and unpredictable cruising ranges, and they have to be managed efficiently given the dynamic charging process. It is urgent and challenging to investigate a computationally efficient algorithm that provide EV AMoD system performance guarantees under model uncertainties, instead of using heuristic demand or charging models. To accomplish this goal, this work designs a data-driven distributionally robust optimization approach for vehicle supply-demand ratio and charging station utilization balancing, while minimizing the worst-case expected cost considering both passenger mobility demand uncertainties and EV supply uncertainties. We then derive an equivalent computationally tractable form for solving the distributionally robust problem in a computationally efficient way under ellipsoid uncertainty sets constructed from data. Based on E-taxi system data of Shenzhen city, we show that the average total balancing cost is reduced by 14.49%, the average unfairness of supply-demand ratio and utilization is reduced by 15.78% and 34.51% respectively with the distributionally robust vehicle balancing method, compared with solutions which do not consider model uncertainties.