|
|
|
Title: Neural Networks for Travel Time Prediction on Interrupted Flow Facilities
Accession Number: 01456600
Record Type: Component
Availability: Find a library where document is available Abstract: This article focuses on a specific application of an artificial intelligence (AI) paradigm through a case study that uses neural networks (NNets) for travel time estimation and prediction on arterials. It is shown how the state–space notion of roadway traffic can be used with NNets to model travel time on urban arterials. The state–space representation of complex dynamical systems such as arterials provides enough insight to model it accurately. State–space NNets (SSNN) is a generic form of recurrent neural networks (RNN), and is particularly suited for the purpose of travel time predictions on arterials. Furthermore, when SSNN are combined with conditional independence (CI) graphs, a specialized statistical technique, to analyze the independence and interaction among variables involved in the travel time process, a more efficient and robust travel time prediction process emerges whereby fewer variables are needed. This modeling approach relies on data that are easily available in the field and hence has great potential in advanced traveler information systems (ATIS) and advanced traffic management systems (ATMS) applications.
Monograph Accession #: 01456594
Language: English
Authors: Abu-Lebdeh, GhassanPagination: pp 42-56
Publication Date: 2012-11
Serial: Media Type: Web
Features: Figures; References; Tables
TRT Terms: Subject Areas: Data and Information Technology; Highways; Operations and Traffic Management; I73: Traffic Control
Files: TRIS, TRB, ATRI
Created Date: Dec 10 2012 9:45AM
More Articles from this Serial Issue:
|
