Please use this identifier to cite or link to this item:
|Title:||Development of fuzzy neural technique for traffic intersection||Authors:||Liang, Tianchi||Keywords:||DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation::Control engineering||Issue Date:||2015||Abstract:||As the quick development economy, transport problems also become more and more serious in nowadays. In Singapore, the government is facing first class seriousness public transport problems. Congestion, overcrowding and crowd bottle neck is the main problem need to be treat with. To solve this problem, the a highly designed monitor and control system is required for city traffic network. The most common method is pre-timed controller installed in the signalized intersection. However, with the increasing of seriousness of traffic condition, these traditional method is no longer as efficient as before. In recent research area, intelligent techniques are employed in the system to predict the traffic condition and a human-liked analyze techniques are used in the control system to achieve the desired condition. In order to treat with this problem, this project introduce a fuzzy neural network system to design the highly intelligent controller for the traffic lights. The Generalised Dynamic Fuzzy Neural Network, that is GDFNN, is used to teach the learning algorithm which generated by matlab codes. As the online traffic simulation softwares are not suitable for various traffic condition, most of them can only simulate a very simple traffic condition. Therefore, a traffic data are collected and entered the system as input. Therefore with the data input to the system, fuzzy logic is used to set the signal time. These outputs will act as input of another traffic fuzzy logic to evaluate the performance of the whole system. The whole system is run on the base of former 4 intersections near the target intersection. In the system, the bigger number of cars travel through the opposite directions will be saved as one of the input. The second input will be the bigger number of the other direction. The sum of the two input will be the input to generate the total time one round of traffic lights. The time of in one round of the traffic lights will be divided due to the traffic heavy condition of each direction. The performance of the system is evaluated by another fuzzy system. Along the project, the GDFNN has generated a good outcome to the aim of the project, which means good prediction and low error rate.||URI:||http://hdl.handle.net/10356/64384||Rights:||Nanyang Technological University||Fulltext Permission:||restricted||Fulltext Availability:||With Fulltext|
|Appears in Collections:||EEE Student Reports (FYP/IA/PA/PI)|
Page view(s) 1095
checked on Oct 21, 2020
checked on Oct 21, 2020
Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.