On Line Estimation of Turning Movements and Saturation Flows in PRODYN

Abstract

PRODYN is the french real time urban traffic control algorithm developed by CERT and assessed on ZELT experimental field test in Toulouse. It is based on Dynamic Programming sub-system optimization and on Decentralized Coordination. The real-time optimization is implemented on a rolling horizon and state variables like queues are estimated by Bayesian technics. As PRODYN still requires manual introduction of traffic parameters like turning movement ratios (TMR) and saturation flow rates (SFR), we have developed real time estimation algorithms for those parameters using data from existing magnetics loop sensors.In this paper we present, first, the results of a study on simulation which shows that control efficiency is strongly affected by parameter variations. Then, we present different TMR estimation methods based either on Least-Square minimization or on Kaiman filtering technic. The originality of some methods consists in using traffic light information and considering not only equality but also inequality constraints on TMR. Algorithms are compared, on simulated sensor data, in order to get the one which works best in real-time. The last part of the paper concerns the study of three SFR estimation methods. The first one is based on an identification of the flow-occupancy logarithmic relationship for a single sensor located at the traffic light stop line. The two other are based on the use of various sensors and on concomitant estimation of vehicule queues and SFR. The second method uses actual PRODYN sensors and the third method is based on a new location for loops.Finally, these estimation algorithms have been proved to be efficient and fast enough to be integrated in PRODYN architecture in order to improve the overall control. This architecture will be implemented and tested on ZELT system soon.