Highlights An ensemble-based system is used as a decisional entity for efficient traffic control. The approach requires a reduced number of sensors located along principal roads. We solve the problems related to training, testing and evaluation of the ensemble. The method is implemented within the framework of wireless sensor/actuator network. We present a case study to show the preciseness of our method.
Abstract Rapid advances in computing, sensing and telecommunication technology offer unprecedented opportunities for artificial intelligence concepts to expand their applications in the field of traffic management and control. Our methodology gravitates around a powerful decision-making method: ensemble-based systems. This technique is used to accurately classify the near future traffic conditions and to make efficient decisions for adapting the traffic lights sequences within an urban area to optimize the traffic flows. The proposed approach requires only measurements provided by traffic sensors located along the principal roads entering the zone. This reduced number of sensors are considered to be enough relevant for classifying the near future state of the traffic and moreover, their measurements can be validated through analytical/hardware redundancy. Our methodology is meant to be implemented within the framework of a wireless sensor and actuator network and is confirmed by computer simulation, including normal or abnormal traffic conditions, for the central part of the city of Timisoara-Romania.
Ensemble based traffic light control for city zones using a reduced number of sensors
Tramsportation Research, Part C: Emerging Technologies ; 46 ; 261-273
2014-06-25
13 pages
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Ensemble based traffic light control for city zones using a reduced number of sensors
Online Contents | 2014
|SAFETY Temporary traffic control zones
British Library Online Contents | 2006
Traffic control adjacent to safety zones
Engineering Index Backfile | 1928
|