MAITE

The Third Transportation Student Research Symposium

Abstract Detail

Name	School	Title of presentation
Charisma Choudhury	MIT	Modeling Anticipatory Driving Behavior

Driving behavior models are used within microscopic traffic simulations to predict driving maneuvers. Existing models usually assume that drivers react to current traffic conditions. However, in reality, drivers may plan a set of actions based on previous, current and anticipated future conditions and make a sequence of choices to execute the chosen plan. For example, a driver familiar with the network and traffic conditions may target lanes based on his path-plan and predicted delays. Maneuver to the target lane however may not be possible due to conflicts with other vehicles and only the immediate lane choice of the driver is observed. Similarly, a driver who has decided to merge but cannot do that immediately may try to anticipate the maneuvers of the drivers in his target lane and make decisions based on that. Such anticipatory behaviors are more prominent in congested and incident affected traffic conditions and ignoring this phenomenon can result over prediction of congestion.
The paper focuses on methodology on modeling anticipatory driving behavior and the associated unobserved decisions and demonstrates this through two scenarios: arterial lane selection and freeway merging. Both models are estimated using detailed trajectory data. The improvements in the simulation capability are validated using the microscopic traffic simulator MITSIMLab.

Back to Abstract List page

Events Page

Visit the New England Section of ITE events page for news about other upcoming events around New England, including a Mass ITE Chapter meeting .

with comments and questions.

Last updated January 9, 2009