As cities move toward “smart city” paradigms they will engage more and more with private mobility operators. Examples of these efforts include the Mobility Data Specification from LA DOT. Data sharing for such ecosystems require mechanisms to preserve the competitive privacy of operators’ operational policies. In our latest paper, we develop one such mechanism using concepts of constrained entropy maximization and inverse optimization constraints to ensure synthetic queries exhibit certain desirable features. This work will be presented at ISTTT23 in Lausanne and is funded by NSF CMMI-1652735. 

Link to paper: https://www.sciencedirect.com/science/article/pii/S0968090X1831622X

Recent news suggest the importance of studying the roles of public transit and mobility-on-demand services in competition and cooperation. We developed an integrated dynamic platform that can incorporate transit rides as a leg in the MOD service and showed that for certain scenarios (such as Manhattan trips from Long Island via LIRR) the MOD services will naturally converge to providing feeder access because it is more cost-effective. Joint work (NYU/LISER) with Tai-Yu Ma, Saeid Rasulkhani, and Sylvain Klein. Funding support from the Luxembourg National Research Fund (INTER/MOBILITY/17/11588252) and NSF CMMI-1634973.

https://doi.org/10.1016/j.tre.2019.07.002

A major issue in the development of reinforcement learning algorithms for autonomous vehicles is the need to make them reflect the preferences of travelers better. One of the ways to do that is to incorporate user schedule preferences under reliability-based route selection criteria into the learning mechanism. This work led by Jinkai Zhou investigates the potential of such an integration. We used data collected from queries from Google Maps to mimic airport shuttle services to train a multi-armed bandit algorithm to see how it is impacted by the consideration of on-time arrival reliability. The work was funded by NSF CMMI-1652735.

https://journals.sagepub.com/doi/full/10.1177/0361198119850457