Real-time surveillance of traffic conditions on urban freeway corridors using spatially discrete presence detectors is addressed. Using a finite-dimensional (macroscopic) fluid-analog model for freeway vehicular traffic flow, an extended Kalman filter is proposed as a data-processing algorithm to obtain minimum variance estimates of spatial mean speed and density. It is shown that certain model parameters associated with available roadway capacity can be estimated on-line with a variation of the extended Kalman filter, and furthermore, that the time signatures associated with these estimates provide quantitative information concerning the presence of anomalous (i.e., incident) traffic events. Performance of the surveillance algorithm is evaluated using a detailed, multi-lane microscopic vehicle simulation which retains a stochastic mix of driver-vehicle types and passing. Complexity problems associated with the computer implementation of the extended Kalman filter are addressed, and techniques for decentralized realization are proposed.
Traffic Surveillance Data Processing in Urban Freeway Corridors Using Kalman Filter Techniques
1978
198 pages
Report
Keine Angabe
Englisch
Transportation & Traffic Planning , Road Transportation , Freeways , Urban areas , Traffic engineering , Data processing , Surveillance , Algorithms , Estimates , Vehicular traffic , Real time operations , Traffic control , Methodology , Partial differential equations , Mathematical models , Computerized simulation , Traffic surveillance , Traffic flow , On line systems , Kalman filters , Traffic incident detection , Payne model , Traffic management
Multimodal Optimization of Urban Freeway Corridors
NTIS | 2006
|Integrated Traffic Control for Incidents in Freeway-Arterial Corridors
British Library Conference Proceedings | 1997
|Optimum Volume of Freeway Corridors
Transportation Research Record | 2020
|Improved Kalman Filtering Approach for Estimating Origin-Destination Matrices for Freeway Corridors
Online Contents | 1994
|