EMBEDDED DISTRIBUTED SIMULATION FOR TRANSPORTATION SYSTEM MANAGEMENT

1
EMBEDDED DISTRIBUTED SIMULATION FOR
TRANSPORTATION SYSTEM MANAGEMENT

• Michael Hunter
• Richard M. Fujimoto
• Christos Alexopoulos
• Randall Guensler
• Frank Southworth
• Georgia Institute of Technology

2
Embedded Distributed Simulation for
Transportation System Management
Region Wide Traffic Network
IN-VEHILCE SIMULATION
IN-VEHILCE SIMULATION
Vehicle-to-Vehicle and Vehicle-to-Roadside
Communication Architecture to Central Server
IN-VEHILCE SIMULATION
Traffic Control Adjustments Data Requests
In-Vehicle Simulation Predictions
IN-VEHICLE SIMULATION
Traffic Sensors
REGIONAL SERVER
In-Vehicle Data Module
Historical Data
Physical Data
In-Vehicle Simulation
-
Area wide traffic projection
-
REGIONAL DISTRIBUTED SIMULATION
-
Wide area control adjustments
-
Control Updates
Traffic Control
-
Redirect mobile sensor
-
-
Adjust sensor polling
-
OBJECTIVE - Reconfigurable, resilient,
monitoring, forecasting control with ad hoc
distributed simulation
RESEARCH CHALLENGES - Ad hoc distributed
simulation, client/server and infrastructure-less
architectures, statistical assessment, client
simulation area, adaptive learning, aggregation
of sensor data and simulator predictions, and
reconfigurable system test bed
3
Expected Transformative Benefits

• Today widespread deployment of sensors,
  computers, and communications in vehicles and
  roadways is creating new challenges and
  opportunities to effectively exploit the wealth
  of newly realized real-time data and information
• ARES-CI - to include dynamic collections of
  interacting autonomous simulations continuously
  running in a real-time distributed environment
  to
• Mitigate recurring and non-recurring congestion
• Inform system operators and users through
  collaborative use of information
• Emergency response planning and visualization
  (situational awareness during emergencies)
• Enhance the overall economic efficiencies by
• Highlighting benefits of gathering system
  intelligence
• Providing feedback to autonomous agents within a
  system
• A broad applicability of the ad hoc distributed
  simulation concept to a wide variety of systems
• Participation of underrepresented groups
• Integrate Georgia Intern-Fellowships for Teachers
  Program
• Engage Metro-Atlanta high-school students
• Active participation of undergraduate students

4
EXPERIMENT AND MANAGEMENT PLAN
Phase 1 Development of ad hoc distributed
simulation V2R communication Phase 2
Deployment of infrastructure-based ad hoc
simulation Phase 3 Algorithm development of
infrastructure-less systems Phase 4 Deployment
of infrastructure-less ad hoc simulation
Dr. Hunter, Project Director - transportation
simulation and scenario development Dr.
Fujimoto - distributed simulation algorithms and
software Dr. Alexopoulos - data aggregation and
statistical analysis Dr. Guensler -
reconfigurable system test bed Dr. Southworth -
adaptive learning and client simulation area
definition