CykelTAKO - Bicycle traffic analysis on signalized intersections
To quantify how intersections contribute to congestion in bicycle traffic, and how infrastructure design, flow and heterogenous speed preferences shape queuing and platoon formation.
A significant cause of delays in bicycle traffic is linked to signalized intersections, especially in cities with high flows, yet recommendations for designing infrastructure to cope with a high demand are largely lacking today. At signalized intersections, queuing creates platoons which may lead to a higher proportion of bicyclists being constrained downstream of the intersection than upstream. Due to the large heterogeneity in speeds among bicyclists, delays may occur even at low flows if overtaking is restricted, which may further increase delays over long distances after a crossing, in addition to the direct delay of the red signal at the crossing. Empirical knowledge on what contributes to delay is limited, particularly for Swedish conditions.
The project investigates the queueing behavior of bicyclists at signalized intersections and its contribution to the formation of platoons in bicycle traffic after the intersections. The specific research questions are:
- Compared to motorized traffic, bicyclists waiting at red signals may deviate from strict FiFo-queues, which may result in certain types of bicyclists positioning themselves at the front of the queue. How do infrastructure (namely bicycle path width), flow, and heterogeneity in a population of bicyclists influence positioning when waiting at red signals, and their behavior when the light turns green?
- How does the behavior of bicyclists influence the clearance of the queue?
- How are platoon formation and proportion of constrained bicyclists influenced by queuing behavior, flow, heterogeneity in a population of bicyclists, and bicycling infrastructure design (before, at, and after the crossing)?
To investigate these research questions, video data collection via drones is proposed at 10–20 selected intersections during rush hours to observe the behavior of bicyclists over a stretch of at least 50 meters before and after intersections. Automated tracking technology is used to extract the trajectories of cyclists, while side-mounted cameras on the ground document the type of bicycles. This provides a basis for studying queuing behavior and platoon formation.
The project is expected to provide:
- quantification of the share of impeded cyclists downstream of signalized intersections,
- input data for default values, for example saturation flows and acceleration distributions for bicycle traffic,
- a method for automatic identification of bicycle type in real traffic, enabling more detailed analyses by bicycle type,
- data and insights that can be used to develop and calibrate simulation models, and to give recommendations for the design of signalized intersections.
Contact:
Guillermo Pérez Castro, guillermo.perez.castro@vti.se