WTI

Ant Colony Optimization for Transportation Optimization Problems - UTC

Primary Investigator Contact Information

External Project Contact(s)

Report(s)

There are no reports associated with this project.

Project Objective

The overall objective of this project is to formalize, test and enhance propagation estimate techniques used to select the optimal placement of communications infrastructure using artificial intelligence.

Project Abstract

WTI leads numerous research projects to improve transportation in rural areas. A typical challenge in locations is the absence of seamless or reliable communication, due to the difficulty of effectively placing infrastructure in remote areas or rugged terrain. A WTI researcher has conducted preliminary research into the use of an optimization technique called “Ant Colony Optimization,” to investigate whether it can be applied to select optimal placement of communications infrastructure along a roadside. The initial research has included computational analysis, integration of digital elevation models, and development of a working algorithm. Ant Colony Optimization (ACO) is an artificial intelligence algorithm and is a form of “Swarm Intelligence.” Ant Colony Optimization algorithms mimic the behavior of ants searching for food. Ants deposit pheromones on the paths they follow when searching for food. The ants that find food survive and retrace their paths back to their homes, making the pheromone deposited along trails leading to food even stronger. Other ants follow pheromone-laden trails leading to food, and ultimately the shortest paths to food are found by the collective ant colony. Similarly, ACO explores potential paths to solutions of problems and increases the weights of paths leading to good solutions. Ultimately, ACO algorithms converge to near-optimal solutions for very complex problems, such as the infrastructure placement problem. This project will allow the researchers of the Systems Engineering Group to build on the previous research through enhancement of the estimation techniques and algorithm, generation of test cases based on real locations and equipment, and identification of other applications for the techniques used in the research.

Task Descriptions

1. Project Management
   1. This task will also include supervision of staff, budget monitoring, and general reports.
2. Algorithm Enhancement
   1. The algorithm will be enhanced to cover a hierarchy of possible scenarios for installing infrastructure for a roadside communications network: • Roadside backbone • Roadside backbone with RF coverage of entire roadway • Mountain-top backbone with Point-to-point Roadside Gateways and LANs (hybrids) Scenarios above with additional constraints and flexible objective functions to include prioritization of coverage and quality of service measures.
3. Final Report and Tech Transfer
   1. A final report will be generated to document project work and results. This report will serve as the basis for presentations and publications, as appropriate.
4. Formalize Propagation Models
   1. Researchers will formalize propagation estimation techniques to conform to known propagation models.
5. Test Case Generation and Evaluation
   1. The team will generate test cases to correspond to scenarios in the TMC-TMS Communications project for Caltrans. These will include real roadside equipment locations (TMS), mountain top tower locations, etc. The result will be the “optimal” configuration of communications equipment to provide sufficient coverage for each scenario.

Milestones, Dates, Schedule

Student Involvement

True

Relationship to Other Research Projects

True

Technology Transfer Activities

True

Transportation Research Board Keywords

Ant Colony optimization; communications networks; communications infrastructure placement

Partners

Research and Innovative Technologies Administration; Office of Research, Development and Technology (USDOT)