Even relatively small changes in traffic intensity or the capacity of busy sections of the road network can provide a sufficiently large reduction in traffic delays. Therefore, it is necessary to predict the location of the stressed points of the transport network in a timely manner, and apply one of the developed algorithms for redistributing traffic flows. Intersections and road sections in front of traffic lights are considered as possible stress points. The condition of each point in the transport network can be described using an input vector: information on the average speed of vehicles, traffic intensity and density, percentage of heavy vehicles, availability of repair zones, etc. There is also information about 6 fuzzy classes characterizing the degree of tension of the point (“free movement”, “stable movement”, “constant limited movement”, “constant movement with high density”, “saturated movement”, “congestion”). Each point can be assigned to a certain class with some degree of belonging [0, 1]. An algorithm for a neuro-fuzzy classifier using a fuzzy inference mechanism in solving the classification problem by a neural network with clear/fuzzy connections will be proposed.
Adaptive Traffic Control Through a Neuro- fuzzy Classifier for Predicting Traffic Network Tension Levels
2024-06-20
213232 byte
Conference paper
Electronic Resource
English
Neuro-fuzzy techniques for traffic control
Tema Archive | 1998
|Neuro-Fuzzy Techniques for Traffic Control
British Library Conference Proceedings | 1997
|Fuzzy traffic control: adaptive strategies
Tema Archive | 1993
|