Genetic fuzzy logic approach to local ramp metering control using microscopic traffic simulation : a thesis presented in partial fulfillment of the requirements for the degree of Master of Engineering in Mechatronics at Massey University, Auckland, New Zealand
Loading...
Date
2009
DOI
Open Access Location
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Massey University
Rights
The Author
Abstract
Ramp metering, one of the most effective solutions for improving motorway traffic
flows, is playing increasingly important role in traffic management systems. Because of
its capability to handle nonlinear and non-stationary problems, fuzzy logic based ramp
metering algorithms have been always considered as an extremely suitable control
measures to handle a complex nonlinear traffic system. This thesis proposes a genetic
fuzzy approach to design a traffic-responsive ramp control algorithm for an isolated onramp.
For a local ramp meter algorithm, the problem could be described as the inflow
optimization of on-ramp, based on the evaluation of motorway traffic condition. If the
inflow of on-ramp is considered as the decision variable, the ramp control problem could
be treated as a nonlinear optimization problem of maximizing the evaluation function.
The adaptive genetic fuzzy approach is actually a control approach to maximize the
inflow of on-ramp under the restriction of evaluation function.
In this thesis, a well-known fuzzy logic based ramp metering algorithms developed by
Bogenberger is introduced and implemented with an on-ramp congestion model of
Constellation Drive Interchange in a stochastic microscopic traffic simulator, Aimsun. To
improve the performance of fuzzy control system, genetic algorithm is applied to tune the
parameterized membership function of each fuzzy input to maintain the flow density of
motorway blow the estimated congestion density. The performances of the genetic fuzzy
logic control ramp metering are compared with FLC (fuzzy logic control) ramp metering
by means of the percentage change of TTT (Total Travel Time) based on no control
condition in Aimsun. The simulation results show the genetic fuzzy ramp metering has a
more significant improvement on TTT and more strong stability to maintain system flow
density than FLC ramp metering.
Description
Keywords
Traffic flow, Ramp meter, Traffic control, Motorway ramps