Methodology Framework for Developing Adaptive Traffic Control Strategy – A Novel Concept for Traffic Engineers,

Koltovska Nechoska, Daniela and Bombol, Kristi Methodology Framework for Developing Adaptive Traffic Control Strategy – A Novel Concept for Traffic Engineers,. In: ISEP 2013, 26 March, 2013, Ljubljana, Slovenia.

[thumbnail of Koltovska.Bombol.pdf] Text

Download (318kB)


The intelligent agent technology, sharing the characteristics of applicability in real time and adaptation, capability of self-analyzing by errors and success, learning and improvement (in the course of time) interacting with the environment, quick learning from a large amount of data, represents the new approach employed in the development and design of new adaptive control strategies. These are strategies that incorporate a higher level of intelligence and are capable of self-learning and experience-based decision making.
However, the process of developing such strategies, that are undoubtedly our future, is by no means easy, particularly as regards traffic engineers – researchers. Nowadays they are facing a double challenge: application of new unknown, narrowly specialized methods in the area of computer science, as well as accessibility to professional literature, predominantly meant for individuals possessing pre-conception in the area of artificial intelligence and machine learning.
Hence is our motive to write this paper – to present the methodology for designing and developing an adaptive control strategy for an isolated intersection, by applying the intelligent agent and the reinforcement learning method, refracted through the traffic engineer’s prism. The designed methodology comprises of three steps: development of a model, design and development of an intelligent agent, strategy testing and evaluation. Each step is explained in an easily understandable manner from the point of view of a traffic professional.
We assume that this clearly established methodology is going to be of invaluable assistance to all traffic engineering researchers dealing with the issue of artificial intelligence and machine learning for the first time.

Item Type: Conference or Workshop Item (Paper)
Subjects: Scientific Fields (Frascati) > Engineering and Technology > Other engineering and technologies
Divisions: Faculty of Technical Sciences
Depositing User: Prof. d-r. Daniela Koltovska Nechoska
Date Deposited: 15 Sep 2020 07:07
Last Modified: 13 Oct 2020 09:00

Actions (login required)

View Item View Item