Variants of genetic algorithms based on localized genetic mutations.
Published in Chapter:
Real-Time Smart Navigation and the Genetic Approach to Vehicle Routing
Cristina De Castro (IEIIT-CNR, National Research Council of Italy and WiLab, University of Bologna, Italy), Barbara Mavì Masini (IEIIT-CNR, National Research Council of Italy and WiLab, University of Bologna, Italy), Ibrahim Habib (Grove School of Engineering, The City College of New York, USA), and Oreste Andrisano (DEI and WiLab, School of Engineering and Architecture, University of Bologna, Italy)
Copyright: © 2015
|Pages: 12
DOI: 10.4018/978-1-4666-5888-2.ch192
Abstract
The ever increasing traffic requirements are boosting the evolution of Smart Navigation Systems, where innately time-variant information, such as road conditions, must be properly handled with. For instance, when the quickest route to a destination is requested, if significant traffic variations occur during the trip, the first path suggested can become no longer efficient. In this case, the user must be indicated an alternative in real-time. Three main interactive factors contribute to service quality: architectures, wireless communication methods and vehicle routing algorithms. An infrastructure and communication strategies are thus presented, for traffic monitoring, vehicle-system data exchange, and real-time transmission of path variations. Results about saved travel time are reported. A survey is then presented about the performance of Genetic Algorithms in real-time routing, with special attention to localized searches; a variant is consequently proposed of the adopted traffic control and prediction architecture.