Michael W. Levin

A general framework for modeling shared autonomous vehicles with dynamic network-loading and dynamic ride-sharing application

•We present a framework for modeling shared autonomous vehicles (SAVs) compatible with a general class of traffic simulation.•We implement our framework using a cell transmission model simulator on a city network.•SAVs can greatly increase congestion, and therefore road congestion should be included in all SAV models.•We compare polynomial-time heuristics for dynamic ride-sharing and preemptive relocation for SAVs.

Enhancing the Validity of Traffic Flow Models with Emerging Data

Modeling the impact of connected and automated vehicles (CAVs) on the environmental sustainability, mobility and safety of roadway traffic at the local link level or the regional network level requires a significant amount of currently non-available data. Multiple CAV test-beds and data collection efforts utilizing the latest sensing and communication technologies have been however publicized over the past few years. Such efforts have been led by the industry and public agencies in the US and abroad. Accordingly, (1) researchers and practitioners should be aware of the type and quantity of data needed to calibrate and validate traffic models while taking into account the impact of CAV technological specifications, the driver behavioral characteristics and the surrounding driving environments. (2) Moreover, the gap between such emerging data needs and the data made available to researchers or practitioners should be identified. This chapter summarizes the presentations of speakers that are investigating such gap during the Automated Vehicles Symposium 2017 (AVS17) held in San Francisco, California on July 11–13, 2017. These speakers participated in the break-out session titled “Enhancing the Validity of Traffic Flow Models with Emerging Data”. The corresponding discussion and recommendations are presented in terms of the lessons learned and the future research direction to be adopted. This session was organized by the AHB45(3) Subcommittee on Traffic Flow Modeling for Connected and Automated Vehicles.

Dynamic transit lanes for connected and autonomous vehicles

Transit lanes provide dedicated right-of-way to transitxa0vehicles, but reduce the number of lanes available to otherxa0vehicles. Several studies have implemented intermittent bus lanes, which are sometimes reserved for transitxa0but otherwise are available for general traffic. However, their efficiency suffers from the difficulties of communicating accessibility to drivers. We extend this concept by proposing dynamic transit lanes for connected autonomous vehicles, in which infrastructure continuously updates vehicles on lane accessibility. We present a cell transmission model of dynamic transit lanes in which the number of lanes available to general traffic changes in space and time in response to the presence or absence of transitxa0vehicles. In order to extend the concept of transit signal priority in the context of connectedxa0autonomous vehicles and integrate it with dynamic transit lanes, we also modify the reservation-based intersection control system for autonomous vehicles to prioritize transit. Numerical results from small test cases show that the dynamic transit lanes and transit intersection priority allow transit to move nearly at free flow on the corridor despite congestion. Results from the downtown Austin city network using dynamic traffic assignment show that bothxa0transit and general trafficxa0would experience significant benefits in realistic settings.