A neuro-fuzzy approach for ramp metering

Ramp metering, in the context of ITS systems, offers several operational features for improving traffic flow, traffic safety and air quality by the regulation of input flow to a freeway or highway. A general fuzzy ramp control algorithm is described and an example is examined. This fuzzy control algorithm is arranged into a neuro-fuzzy architecture and the membership functions are learned by a neural network based method from the US demand-strategy method.