17-05-2013, 01:06 PM
INTELLIGENT TRAFFIC CONGESTION CONTROL SCHEME
INTELLIGENT TRAFFIC.pptx (Size: 1.57 MB / Downloads: 44)
INTRODUCTION
Major plague in modern life
Characterized by slower speeds, longer trip times and increased queuing
Congestion is incurred when interaction of vehicles slows the speed of traffic stream
When vehicles are fully stopped for periods of time, this is colloquially known as traffic jam
Numerous methods are available for controlling traffic signal
No approach always gives better response for all system
PROBLEMS ASSOSCIATED WITH TRAFFIC CONGESTION
Traffic congestion has been causing many critical problems and challenges in most modern cities
Due to congestion problems people loss time, miss opportunities and get frustrated
Traffic congestion directly impacts the companies
Cause loss in productivity from workers, trade opportunities are lost, delivery gets delayed
Costs go on increasing
FUZZY EXPERT SYSTEM
Used to control traffic in many cities
Fuzzy Expert system consists of seven elements
a) RFID card reader
b) Active RFID Tag
c) A personal Digital Assistance
d) A Wireless Network
e) A Database
f) A Knowledge Base
g) A Backend Server
ARTIFICIAL NEURAL NETWORK APPROACH
The Adaptive Traffic Light Problem was modeled using ANN approach
It include predicting the traffic parameters for the next time frame
It also include computing cycle time adjustment values
This model consisted of nine inputs and three output nodes
One of each past and present traffic parameters
One hidden layers with 70 hidden nodes
INTELLIGENT DECISION MAKING SYSTEM FOR URBAN TRAFFIC CONTROL
Real time intelligent decision making system
Computes decisions within a dynamically changing application environment
The architecture consists of an artificial neural network and a fuzzy-rule-based expert system
Sensors are placed at the road to sense the different parameters of the traffic conditions
CONCLUSION
IDUTC system provides relievable decisions better than ANN approach
IDUTC impose a lower average wait time than the other two approaches
ANN requires more neural network nodes than ANN in IDUTC
ANN approach need more training time and high implementation cost
FES system leads to correct decision but didn’t reduce wait time as compared to IDUTC
IDUTC uses current and past values but FES uses only current traffic flows
IDUTC system is best for controlling traffic light in best and smart way