A Multi-Agent Based Method for Large-Scale Route Planning Using Distributed Constraint Optimization

With the increasing volume of traffic on urban roads, the issue of congestion in urban areas has become more severe. Additionally, the rapid growth of smart devices and changes in driving habits have led to high-concurrency and high-volume routing requests. Traditional route planning algorithms only...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings of ... IEEE International Conference on Unmanned Systems (Online) S. 1213 - 1218
Hauptverfasser: Li, Yixuan, Che, Qian, Wang, Fengmin, Zhou, Yifeng, Li, Chuanyou, Wang, Wanyuan
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 13.10.2023
Schlagworte:
Constraint optimization
distributed constraint optimization
Heuristic algorithms
large-scale route planning
Machine learning algorithms
Max-sum algorithm
multi-agent systems
Roads
Routing
Stochastic processes
Urban areas
ISSN:2771-7372
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:With the increasing volume of traffic on urban roads, the issue of congestion in urban areas has become more severe. Additionally, the rapid growth of smart devices and changes in driving habits have led to high-concurrency and high-volume routing requests. Traditional route planning algorithms only consider the optimization of individual route query and often fall into local suboptimal solutions in large-scale routing scenarios. Some scholars have proposed dynamic large-scale route planning methods based on greedy and correction strategies for multi-route planning. However, these methods are unable to return a combination of routes in time-sensitive scenarios. In this paper, we model dynamic large-scale route planning using distributed constraint optimization and propose a max-sum-based route planning algorithm with pruning optimization. A parallel accelerated distributed stochastic search algorithm is also proposed to meet scenarios with computational efficiency requirements. Experimental results demonstrate that the proposed algorithms can effectively reduce the global travel time.
ISSN:2771-7372
DOI:10.1109/ICUS58632.2023.10318482