Engineering/research internship, Spring-summer 2012
PATH, Berkeley University, California, USA (where the internship shall be supervised by Pr Steven Shladover)
An important practical aspect of traffic data analysis is the way it is obtained. Until recently most information used to come from magnetic detection loops buried into some main roads. Nowadays, due to wide usage of GPS navigation, more and more data are collected from fleet of probe vehicles (aka floating car data). The individual travel times of one single probe vehicle being potentially a very noisy estimator of the traffic state, these are usually aggregated over time, space and vehicles. Interesting open research topics are therefore related to the quality/fidelity/characteristics of obtained traffic data depending on sampling and aggregation strategy, which can depend on the application requirements.
The robotics lab (CAOR) has been working on traffic datamining since nearly 3 years, within a collaborative project, with two INRIA labs, funded by French National Research Agency (ANR). This work is also conducted in a research collaboration recently started with MobileMillenium team of Berkeley University, which has a large real traffic database on San Francisco region. Another Berkeley lab, the PATH, has a large experience in using traffic simulations for testing various Intelligent Transport Systems applications. CAOR and PATH have begun a collaboration, funded by CalFrance, on use of probe vehicles in traffic related applications.
The internship work shall be focused on possible use of probe vehicles traffic data in applications related to traffic signal management and command. More precisely, the aim is to analyze how the probe vehicles data can be used as a complement to standard loop detectors, to improve control of traffic signals. The improvement could in particular be related to the estimation quality of the traffic situation, either for continuous adapting of signal cycle and duration, or for global switching of traffic lights regime in a whole arterial area. This last aspect could be of particular importance in the context of Integrated Corridor Management, which could imply significant re-routing of traffic from congested freeways into neighbouring arterial roads. The work will require use of existing traffic simulations tools (such as VISSIM, and/or some mesoscopic simulation software), and statistical analysis of historical real-world traffic data.