Integrating connected vehicles into urban traffic management systems

Abstract

Connected intelligent transport systems contain a wealth of data accessible to traffic signal controllers. However, algorithms that use data from a connected environment do not fully exploit the potential of this new data source. Instead, traffic signal controllers rely on speed and position data to supplement data from infrastructure. This research aims to understand which data that are available from connected vehicles are useful for integrating with existing traffic signal control systems in urban environments. Vehicle positions and speeds fit well into our current understanding of traffic theory, but more abstract data such as passenger counts and stop frequencies may offer new ways to optimise traffic signal controllers to reduce traffic delays.

The contributions of this research include 1) A traffic signal control algorithm which combines position information from connected vehicles with data from existing inductive loops and signal timing plans to perform decentralised traffic signal control to reduce delays at existing urban intersections. The algorithm adapts to scenarios with low numbers of connected vehicles and degraded infrastructure, an area where existing traffic signal control strategies are limited. 2) A framework for testing connected traffic signal controllers based on a large urban corridor in the city of Birmingham, UK. The testing framework overcomes the limitations of existing research by implementing a large-scale, realistic simulation case study, which accounts for mixed-mode traffic, multiple levels of traffic demand, degraded loop detector coverage, non-ideal wireless communications, and a full 24-hour simulation period. 3) A greedy stage sequence optimisation algorithm that abstracts for arbitrary connected vehicle data. 4) A method for introducing implicit stage coordination to greedy stage optimisation paradigms. 5) Insights that show coordination is redundant when signal controllers have accurate data and can react fast enough to traffic changes.

This research shows how data connected vehicles can be exploited to improve urban traffic signal control, and how using connected vehicle data differs from traditional sources. The outcomes of this research have a significant impact on the implementation of connected intelligent transportation systems and policy for the transportation industry.

More information

Identifiers

ORCID for Craig Benjamin Rafter: orcid.org/0000-0003-3411-114X

ORCID for Thomas Cherrett: orcid.org/0000-0003-0394-5459

ORCID for Bani Anvari: orcid.org/0000-0001-7916-7636

ORCID for Benedict Waterson: orcid.org/0000-0001-9817-7119

Catalogue record

Export record