Multiple ACO-based method for solving dynamic MSMD traffic routing problem in connected vehicles
- PDF / 943,236 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 28 Downloads / 183 Views
(0123456789().,-volV)(0123456789(). ,- volV)
ORIGINAL ARTICLE
Multiple ACO-based method for solving dynamic MSMD traffic routing problem in connected vehicles Tri-Hai Nguyen1 • Jason J. Jung1 Received: 8 May 2020 / Accepted: 24 September 2020 Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract In this study, we focus on dynamic traffic routing of connected vehicles with various origins and destinations; this is referred to as a multi-source multi-destination traffic routing problem. Ant colony optimization (ACO)-based routing method, together with the idea of coloring ants, is proposed to solve the defined problem in a distributed manner. Using the concept of coloring ants, traffic flows of connected vehicles to different destinations can be distinguished. To evaluate the performance of the proposed method, we perform simulations on the multi-agent NetLogo platform. The simulation results indicate that the ACO-based routing method outperforms the shortest path-based routing method (i.e., given the same simulation period, the average travel time decreases by 8% on average and by 11% in the best case, whereas the total number of arrived vehicles increases by 13% on average and by 23% in the best case). Keywords Ant colony optimization Dynamic traffic routing IoV MSMD
1 Introduction With the rise in metropolitan traffic in recent years, traffic collisions and congestion have become more serious issues. With the help of the Internet of vehicles (IoV) as a particular case of the Internet of things (IoT) and advanced wireless communication technologies such as vehicle-toeverything (V2X) communication, an intelligent transportation system (ITS) can be developed and used to improve the traffic efficiency, enhance safety, and reduce the impacts of traffic congestion [16–18, 20]. One of the most critical problems in ITS research is dynamic traffic routing, which is defined as the process of dynamically selecting a sequence of road segments from the origin to the endpoint of a trip. Conventional traffic control systems have relied on centralized data collection and decision making; however, decentralized solutions have been increasingly studied in & Jason J. Jung [email protected] Tri-Hai Nguyen [email protected] 1
Department of Computer Engineering, Chung-Ang University, Seoul 156-756, Korea
recent years [1, 3, 4, 6, 14, 17, 20, 21], with multi-agent swarm intelligence (SI) techniques gradually becoming the most prominent solution [5, 12, 13, 16, 22]. SI effectively resolves optimization problems by exploiting the intelligent global behavior found in various animals (e.g., bird flocks and fish schools), which can be emulated with basic rules of interaction followed by a set of individual agents [7, 21]. Decision making is left to individual entities that exchange local decisions among themselves to optimize a measure of the welfare of the overall scenario. Decentralized systems and multi-agent systems are closely related to this distributed control scheme, thereby presenting a vast, untapped opport
Data Loading...
