DocumentCode :
131705
Title :
Study on Traffic Flow Base on RBF Neural Network
Author :
Xiaoying Li
Author_Institution :
Coll. of Electr. & Inf. Eng., Changsha Univ. of Sci. & Technol., Changsha, China
fYear :
2014
fDate :
10-11 Jan. 2014
Firstpage :
645
Lastpage :
647
Abstract :
According to charging lathing standard type classification volume of traffic blowing the expense at enormous and the data is inaccurate, But the volume of traffic may obtain easily according to the traffic survey classification from each kind of intelligent toll station. The function neural networks method will charge with because of radial base the rate of flow data change becomes volume of traffic inquiring into a data, Building-up changes a model, Using the MATLAB programming, Obtaining each kind of vehicle type error distinguish ratio and total error distinguish ratio. It can utilize fully thereby advantage of the fee-collecting station, Every fee-collecting station all has the precise writer that various motorcycle type passes in the process charging, it can cut down the cost getting the traffic survey data, Enables the traffic survey the data to obtain the full use.
Keywords :
pattern classification; radial basis function networks; road traffic; traffic engineering computing; Matlab programming; RBF neural network; fee-collecting station; intelligent toll station; lathing standard type classification volume; motorcycle type; radial basis function neural network; total error distinguish ratio; traffic flow; traffic survey classification; traffic volume; vehicle type error distinguish ratio; Educational institutions; MATLAB; Mathematical model; Presses; Radial basis function networks; Training; Vehicles; RBF neural network; error ratio; toll system; traffic flow;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Measuring Technology and Mechatronics Automation (ICMTMA), 2014 Sixth International Conference on
Conference_Location :
Zhangjiajie
Print_ISBN :
978-1-4799-3434-8
Type :
conf
DOI :
10.1109/ICMTMA.2014.159
Filename :
6802776
Link To Document :
بازگشت