Open AccessFactors Affecting Road Rating
Author(s) -
Fei Wang,
Bokuan Zhang,
Ruishu Gong
Publication year - 2020
Publication title -
frontiers research of architecture and engineering
Language(s) - English
Resource type - Journals
eISSN - 2591-7595
pISSN - 2591-7587
DOI - 10.30564/frae.v3i1.1658
Subject(s) - artificial neural network , speed limit , analytic hierarchy process , computer science , limit (mathematics) , stability (learning theory) , artificial intelligence , machine learning , simulation , transport engineering , mathematics , operations research , engineering , mathematical analysis
The decision of traffic congestion degree is an important research topic today. In severe traffic jams, the speed of the car is slow, and the speed estimate is very inaccurate.This paper first uses the data collected by Google Maps to reclassify road levels by using analytic hierarchy process. The vehicle speed, road length, normal travel time, traffic volume, and road level are selected as the input features of the limit learning machine, and the delay coefficient is selected. As the limit learning machine as the output value, 10-fold cross-validation is used. Compared with the traditional neural network, it is found that the training speed of the limit learning machine is 10 times that of the traditional neural network, and the mean square error is 0.8 times that of the traditional neural network. The stability of the model Significantly higher than traditional neural networks.Finally, the delay coefficient predicted by the extreme learning machine and the normal travel time are combined with the knowledge of queuing theory to finally predict the delay time.