Traffic control model based on multilayer adaptive fuzzy probabilistic neural network

To control traffic flows under congestion conditions, it is proposed to use a neuro-fuzzy control model based on a multilayer neural network. Since traffic flows can contain different types of transport units, their classification and adaptation in the management of such objects is required. Traffic...

Full description

Saved in:
Bibliographic Details
Main Authors: Varlamova, Lyudmila, Nabiev, Timur, Gubkina, Anna, Kienko, Galina, Tashpulatova, Nadira
Format: Conference Proceeding
Language:English
Subjects:
Adaptive control
Artificial neural networks
Classification
Fuzzy control
Fuzzy logic
Multilayers
Neural networks
Traffic congestion
Traffic control
Traffic flow
Traffic models
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:To control traffic flows under congestion conditions, it is proposed to use a neuro-fuzzy control model based on a multilayer neural network. Since traffic flows can contain different types of transport units, their classification and adaptation in the management of such objects is required. Traffic flows are distinguished by their instability and unsteadiness, a large number of disturbing influences, an adaptive approach to control is required. It is necessary to create a classifying neuro-fuzzy system that will allow producing a simple and effective classification method under the condition of mutually overlapping classes and propose an architecture of a classifying fuzzy probabilistic network that will classify vehicles passing an intersection, both from the point of view of Bayesian and fuzzy classification at the same time. The network should be simple to implement and suitable for processing incoming observations in a sequential online mode. Probabilistic neural networks introduced by D.F. Spekht are a fairly effective tool for solving the problem of classifying input parameters.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0090154