Home // International Journal On Advances in Systems and Measurements, volume 8, numbers 3 and 4, 2015 // View article
Authors:
Leonard Petnga
Mark Austin
Keywords: Dilemma Zone; Metrics; Cyber-Physical Transportation Systems; Artificial Intelligence; Safety.
Abstract:
Our research is concerned with the modeling and design of cyber-physical transportation systems (CPTS), a class of applications where the tight integration of software with physical processes allows for the automated management of system functionality, superior levels of performance, and safety assurance. Part of the safety assurance problem is prevention of deadly accidents at traffic intersections and, in particular, finding ways for vehicles to traverse the dilemma zone (DZ), an area at a traffic intersection where drivers are indecisive on whether to stop or cross at the onset of a yellow light. State-of-the-art approaches to the dilemma zone problem treat the cars and stoplights separately, with the problem formulation being expressed exclusively in either spatial or temporal terms. In this paper, we formulate a methodology that accounts for two-way interactions between the cars and stoplights, and propose quantitative metrics and three-dimensional dilemma tubes as a means for compactly describing sets of conditions for which the vehicle-light system will be in an unsafe state. The proposed metrics enable simple and actionable decision capabilities to deal with unsafe configurations of the system. The second purpose of this paper is to describe a pathway toward the integration of dilemma metrics and dilemma tubes with an ontological framework. The associated platform infrastructure supports algorithmic implementations of simulation and reasoning for resolving unsafe configurations of CPTS, such as those created by the DZ problem.
Pages: 241 to 254
Copyright: Copyright (c) to authors, 2015. Used with permission.
Publication date: December 30, 2015
Published in: journal
ISSN: 1942-261x