In detail, two key objectives are considered. The first is the determination of the traffic state by collecting information obtained from V2X-heartbeat messages, as well as on-board and personal devices, such as PDAs which are fused for obtaining local and global network states. The second one is to use this information for making traffic lights adaptive to the current traffic state. Approaches from self-organization will be used in order to avoid problems arising due to the incomplete knowledge about the network state. Besides these results, COLOMBO will deliver prototypes for novel intersection monitoring, including a model-based determination of vehicular emissions, and incidents detection, as well as methods to determine a cost- and benefit-optimal placement of the developed solutions in a given road network. Additionally, vehicular pollutant emission will be targeted, by developing models for vehicular population in 2020, including electrical vehicles as well as by developing a model for an emission reducing driving behaviour. The work is accompanied by an extension of European tools commonly used within traffic- and V2X-community, namely iTETRIS and SUMO, the US-American tool ns-3 and the European pollutant emission model PHEM.
V2X-based Traffic Surveillance assuming low Equipment Rates
Conventional traffic detection systems are based on inductive loops or camera detectors, which are able to
measure the number of passing vehicles at a particular road cross section or small area.
The first COLOMBO objective is to prove that valuable information about the traffic state can be provided by using data gathered from different forms of car-to-car and car-to-infrastructure cooperative communication, even when assuming low penetration rates.
A new Concept of self-organizing Traffic Light Control System
COLOMBO will develop a traffic light control system directly inspired from research in swarm intelligence. Swarm intelligence is a discipline that studies natural and artificial systems that are composed of a large number of (typically identical or very similar) individuals (“agents”), which coordinate using decentralized control and self-organization. Recent research results have shown that the principles underlying many natural swarm intelligence systems can be exploited to engineer artificial swarm intelligence systems that show many desirable properties and effective solutions.
Evaluation of Traffic Efficiency and Environmental Impacts
COLOMBO uses simulations to evaluate the developed algorithms. This will be achieved by joining simulators of different domains into one execution system. The following software will be used:
- iTETRIS as the middleware which synchronises the following simulators
- SUMO, an open source road traffic simulation
- ns3, an open source communication simulation
- an open source optimization tool kit, to be developed during the project
- models of the applications developed within the work on the first two objectives
In addition, the commercially available emission simulation PHEM will be involved.