Abstract
In many metropolitan areas, traffic congestion is an escalating problem which causes a significant waste of money and time. Nowadays, cars equipped with GPS devices become widespread. The location information of those cars is very useful for estimate traffic condition in the complex city road network. Using the accurate and real time traffic condition, we can provide dynamic route guidance to ease traffic congestion. In this paper, we proposed a speed pattern model, called two phase piecewise linear speed model (2PEED), to estimate traffic condition and represent speed pattern in a road network using GPS data collected vehicles. With the estimated traffic condition and speed pattern, a proposed classification-based route guidance approach using machine learning technique provides dynamic routing for drivers. Using both current traffic data and the experience learned from history data, our route guidance approach is able to accurately predict the future traffic condition and selects a best route. We give simulation results to show that the proposed approach is able to select and dynamically update a route to prove drivers a best (e.g., less traffic and shortest travel time) route to their destination.
Original language | English |
---|---|
Title of host publication | Proceedings - 2011 International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2011 |
Pages | 703-708 |
Number of pages | 6 |
DOIs | |
Publication status | Published - 2011 Sep 19 |
Event | 2011 International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2011 - Kaohsiung, Taiwan Duration: 2011 Jul 25 → 2011 Jul 27 |
Other
Other | 2011 International Conference on Advances in Social Networks Analysis and Mining, ASONAM 2011 |
---|---|
Country | Taiwan |
City | Kaohsiung |
Period | 2011/07/25 → 2011/07/27 |
Keywords
- And machine learning
- GPS
- Route guidance
- Speed pattern estimation
ASJC Scopus subject areas
- Computer Networks and Communications
- Electrical and Electronic Engineering