Neural Networks in Freeway Control

by Yorgos J. Stephanedes, Univ of MN, Minneapolis, United States,
Xiao Liu, Univ of MN, Minneapolis, United States,



Document Type: Proceeding Paper

Part of: Pacific Rim TransTech Conference: Volume I: Advanced Technologies

Abstract:

Hierarchical control strategies have been proposed for freeway optimum control since the 1960's. Multi-layer hierarchical control optimizes system performance to determine a nominal metering rate and adjusts this nominal rate based on current traffic conditions. In this paper we present a heuristic simulated annealing algorithm to optimize the ramp metering rates in a freeway system in a 3 hour peak period off-line. Following optimization we train a backpropagation neural network to learn the optimized relationship between the ramp metering and the traffic variables, and apply the trained neural network to perform freeway ramp control.

Subject Headings: Neural networks | Control systems | Highways and roads | Algorithms | Traffic models | Traffic management | Ramps (road)

Services: Buy this book/Buy this article

 

Return to search