Grant: Discovery Grants Program - Individual
University of Calgary
$24,000.00
Integrated corridor control with incident detection and prediction algorithms
EF 2012-13
Résumé
The objective of a real-time traffic control system is to optimize the efficiency of traffic networks by responding to the dynamic and random nature of traffic on a real-time basis. Fluctuation of traffic depends on two main factors: 1) the control inputs that are directly related to corresponding control devices such as traffic lights and variable message signs (VMS) and 2) the disturbances such as demand variation and incidents whose values cannot be manipulated, but may be possibly measurable (e.g. demand), detectable (e.g. incident) or predictable over a certain time horizon (e.g. traffic flows). Thus, to achieve optimum control, traffic networks have to be continuously monitored, demand and incidents detected and predicted when possible, remedial and pro-active actions taken and control strategies injected in the system to fulfill a set of objectives all in real time. This research will integrate the previously mentioned processes to exchange information for optimum control of urban corridors. The primary goal of the work proposed is to reduce congestion effectively within corridor networks by developing an adaptive integrated corridor control for both recurring and non-recurring congestion by estimating/predicting external disturbances such as demand, incidents, and congestion conditions. The main control objective is to optimize all key controls (traffic signals) and routing (e.g. VMS) concurrently to achieve a set of criterion, such as minimizing total vehicular delay in the network, total queue length, increasing throughput and increasing safety.
The formulated optimal-control problem will be resolved in real time by applying three tools from Artificial Intelligence as follows: 1) Parallel Genetic Algorithms (PGAs) as the main optimization methods that provide traffic-responsive queue management and delay schemes, 2) Neural Networks (NN) for short term congestion estimation/prediction, and 3) Data Fusion algorithms to detect incidents based on machine vision and detectors information.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
Alberta, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
371620-2009
Name nom
Kattan, Lina
Department département
Civil Engineering
Organization id
11
Institution établissement
University of Calgary
Province en
Alberta
Province fr
Alberta
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2012
Competition year année de concours
2009
Award amount
24000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
1061
Committee name en
Civil Engineering - A
Committee name fr
Génie civil - A
Area of application code
700
Area of application group en
Transportation systems and services
Area of application group fr
Services et systèmes de transport
Area of application en
Transportation systems and services
Area of application fr
Services et systèmes de transport
Research subject code
1003
Research subject group en
Civil engineering
Research subject group fr
Génie civil
Research subject en
Transportation engineering
Research subject fr
Génie du transport
Partie
4
Nb partie
5
Application title
Integrated corridor control with incident detection and prediction algorithms
Keyword
NSERC
Application summary
The objective of a real-time traffic control system is to optimize the efficiency of traffic networks by responding to the dynamic and random nature of traffic on a real-time basis. Fluctuation of traffic depends on two main factors: 1) the control inputs that are directly related to corresponding control devices such as traffic lights and variable message signs (VMS) and 2) the disturbances such as demand variation and incidents whose values cannot be manipulated, but may be possibly measurable (e.g. demand), detectable (e.g. incident) or predictable over a certain time horizon (e.g. traffic flows). Thus, to achieve optimum control, traffic networks have to be continuously monitored, demand and incidents detected and predicted when possible, remedial and pro-active actions taken and control strategies injected in the system to fulfill a set of objectives all in real time. This research will integrate the previously mentioned processes to exchange information for optimum control of urban corridors. The primary goal of the work proposed is to reduce congestion effectively within corridor networks by developing an adaptive integrated corridor control for both recurring and non-recurring congestion by estimating/predicting external disturbances such as demand, incidents, and congestion conditions. The main control objective is to optimize all key controls (traffic signals) and routing (e.g. VMS) concurrently to achieve a set of criterion, such as minimizing total vehicular delay in the network, total queue length, increasing throughput and increasing safety.
The formulated optimal-control problem will be resolved in real time by applying three tools from Artificial Intelligence as follows: 1) Parallel Genetic Algorithms (PGAs) as the main optimization methods that provide traffic-responsive queue management and delay schemes, 2) Neural Networks (NN) for short term congestion estimation/prediction, and 3) Data Fusion algorithms to detect incidents based on machine vision and detectors information.