Gerçek zamanlı trafik yol durumu bilgisi ile a* tabanlı rota planlaması: Çok erkinli bir benzetim
Loading...
Date
2010
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Open Access Color
OpenAIRE Downloads
OpenAIRE Views
Abstract
Araç trafiğinin önemli özelliklerinden birisi karmaşıklığıdır. Trafikteki araçlar tahmin edilemeyecek şekilde hareket ettikleri için, trafik ortamında düzenli bir araç akışı yoktur. Bu gibi karmaşık bir ortamda, iki nokta arasındaki en hızlı rotanın bulunması zor bir problemdir. Piyasada iki nokta arasındaki en kısa rotayı ve hız limitleri, geçmiş yol istatistikleri, fm bandı üzerinden yayın ve benzeri kaynaklardan alınan bilgiye dayanarak en hızlı rotayı hesaplayan navigasyon cihazları bulunmaktadır. Fakat hızlı rota hesabında kullanılan bu veriler yolların gerçek zamanlı hız durumlarını içermemektedir. Bu bilgiler kullanılarak yapılan hesaplamalarda, o anda kullanılabilecek en hızlı rota bulunamamaktadır. Bu çalışmada, gerçek zamanlı trafik bilgisi paylaşımı ve dağıtımı yapmak için tasarlanmış çok erkinli bir simülasyon yazılımı geliştirilmiştir. Yazılımda, araçların harita üzerindeki hareketlerinin gerçek zamanlı veriye dayanarak benzetimi yapılmıştır. Erkinlerin yaptığı rota planlamasının kalitesinin sınanması için, bir çok aracın hareket ettiği bir ortam oluşturulmuştur. Yapılan deneylerde, önerilen gerçek zamanlı veriye dayalı rota planlamasının, en hızlı rotayı bulmakta verimli ve etkili olduğu görülmüştür.
A major characteristic of real world traffic is the amount of complexity. The vehicles do not travel in organized clusters, but they move in unpredictable ways, with different speeds even when they are on the same route. It is a challenging problem to find the fastest routes in such complex environment. There are navigation devices and software on the market that are capable of calculating the shortest route from a given source location to a desired destination location using traffic information such as maximum speed limits of the roads, historical road database, broadcasting technology over FM band, or similar. However these techniques do not provide real-time information. Therefore, the shortest routes suggested from one point to another may not be the fastest one in real traffic condition at one time. In this study, a multi-agent simulation has been designed and developed to collect and disseminate real-time traffic information. Each vehicle in this environment has been modeled as an intelligent agent. This software simulates the movement of the vehicle on a road map based on real-time data. In the scope of this study, several tests have been conducted in order to analyze the effectiveness of routing based on real-time data. The experiments showed that the proposed real-time data based route planning is both efficient and effective at finding fastest routes.
A major characteristic of real world traffic is the amount of complexity. The vehicles do not travel in organized clusters, but they move in unpredictable ways, with different speeds even when they are on the same route. It is a challenging problem to find the fastest routes in such complex environment. There are navigation devices and software on the market that are capable of calculating the shortest route from a given source location to a desired destination location using traffic information such as maximum speed limits of the roads, historical road database, broadcasting technology over FM band, or similar. However these techniques do not provide real-time information. Therefore, the shortest routes suggested from one point to another may not be the fastest one in real traffic condition at one time. In this study, a multi-agent simulation has been designed and developed to collect and disseminate real-time traffic information. Each vehicle in this environment has been modeled as an intelligent agent. This software simulates the movement of the vehicle on a road map based on real-time data. In the scope of this study, several tests have been conducted in order to analyze the effectiveness of routing based on real-time data. The experiments showed that the proposed real-time data based route planning is both efficient and effective at finding fastest routes.
Description
Keywords
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrol, Gerçek zamanlı benzetim, Computer Engineering and Computer Science and Control, Rota belirleme, Real time simulation, Trafik planlama, Routing assignment, Traffic planning, Çok ajanlı sistemler, Multiagent systems
Turkish CoHE Thesis Center URL
Fields of Science
Citation
WoS Q
Scopus Q
Source
Volume
Issue
Start Page
0
End Page
74