IEEE-MI-2012-Poster

Developing a Traffic Network Based on Wireless Communication
to Reduce Vehicle Energy Consumption and Emission
Mohamad Abdul-Hak1, Malok Alamir Tamer1, Nizar Al-Holou1, Ph.D., Youssef Bazzi2, Ph.D.
1: Department of Electrical and Computer Engineering, University of Detroit Mercy
2: Lebanese University, Beirut, Lebanon
Overview
Objectives
Vehicle traffic adopts navigation rules
established by vehicle navigation
systems, such as shortest distance,
shortest time or Eco-friendly. Each
navigation system calculates path
using criteria selected by the driver,
and creates destination pathways
used to direct driver to the next road
segment. Existing approaches plan a
path based on historical traffic
information and then modify the route
as the vehicle receives current traffic
conditions thus sacrificing optimality.
In this poster, we propose a novel
genetic algorithm modeled as a Petri
Net (PN) for optimizing travel time and
vehicle emission in a connected
roadway network with minimal total
traffic capacity, to route vehicle in a
dynamically changing traffic
environment, utilizing a predictive
optimization approach. The novel
unfolded PN model presented in this
poster incorporates the essential
features in Dynamic Programming
(DP) to solve the stochastic traffic
routing problem. The effectiveness of
the proposed methodology is validated
by comparing the performance with
conventional routing methodologies.
1. Identify an Eco Friendly
Navigation algorithm, with a
focus on:
• Moving vehicle from a source to a
destination avoiding traffic
congestion thus optimizing
emissions and travel time
• Dynamic re-routing to
accommodate traffic condition
changes
2. Implement algorithm utilizing
microscopic traffic and emission
models to evaluate performance.
Figure 5. Total Waiting/Travel Time
Methodology
1. Transform Roadway network
into a mathematical model
using Petri Net (PN) as
illustrated in Figure 1.
Figure 6. Fuel Consumption and Emission
Figure 3. Emission Incidence Matrix
• The results in Figure 5 and Figure
6 demonstrate the benefits of the
proposed dynamic routing
algorithm in both emission as well
as travel time reduction. This was
achieved through dynamic rerouting however maintaining travel
time optimality.
Conclusion
Figure 4. Travel Time Incidence Matrix
Figure 2. Unfolded Petri Net Model
2. Determine reachability based on
cost objectives:
• emission (δui)
• Total travel time (φui)
Figure 1. Roadway Network
The optimal solution is path
connecting source to destination
with maximum number of emission
tokens arrived with at destination
provided travel time tokens are not
depleted.
3. The re-optimization of subnet uj is
executed on the condition that
higher arc weights are received
indicating worsen traffic condition.
The objective function for reoptimization remains the same
given re-optimization begins at
current marking.
Evaluation
A computer-based open source
simulation tool iTETRIS is selected to
integrate, evaluate and compare the
proposed Eco-route methodology to
the conventional static routing
approach.
• The Eco-friendly routing
methodology has been presented
based on Petri Nets modeling for
optimizing travel routes based on
vehicle emission and travel time.
• The proposed dynamic calculation
approach offers enhanced
performance than conventional
static methodology as it allows for
adapting to sudden changes in
traffic conditions.
This work has been partially supported by the U.S. Department of Transportation and MDOT through the University Transportation Center (UTC) program, MIOH, at the University of Detroit Mercy