A Multiple Representation Entity-Based Approach to Hybrid Traffic Simulation Model
Weinan Huang, Jian Cheng Sun, Keping Li
Abstract
Weinan Huang, Jian Cheng Sun, Keping Li
Abstract
Traffic simulation is a widely used tool to analyze dynamic traffic system. In traffic simulation field, Microscopic simulation offers detailed representation of traffic system, but it is difficult to model large traffic network; Macroscopic and mesoscopic simulations can capture the dynamics of large network, but only on a low level of details. These limitations of mono-resolution traffic simulation emerge more clearly, as the development of its application. Multi-resolution traffic simulation, which uses hybrid traffic simulation model, can overcome those limitations by integrating traffic simulation models on different levels into a hybrid model, which combines advantages of all composing models. In this paper, we introduced UNIFY framework into traffic simulation and developed UNITRANS, a Multiple Representation Entity-based (MRE-based) framework, to integrate traffic simulation models on different resolution levels and maintain the consistency among their jointly-executing. A hybrid model was built using UNITRANS and a three-legged intersection was modeled with it. The results showed that UNITRANS was feasible and effective.
OpenAlex reports 5 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Traffic simulation is a widely used tool to analyze dynamic traffic system. In traffic simulation field, Microscopic simulation offers detailed representation of traffic system, but it is difficult to model large traffic network; Macroscopic and mesoscopic simulations can capture the dynamics of large network, but only on a low level of details. These limitations of mono-resolution traffic simulation emerge more clearly, as the development of its application. Multi-resolution traffic simulation, which uses hybrid traffic simulation model, can overcome those limitations by integrating traffic simulation models on different levels into a hybrid model, which combines advantages of all composing models. In this paper, we introduced UNIFY framework into traffic simulation and developed UNITRANS, a Multiple Representation Entity-based (MRE-based) framework, to integrate traffic simulation models on different resolution levels and maintain the consistency among their jointly-executing. A hybrid model was built using UNITRANS and a three-legged intersection was modeled with it. The results showed that UNITRANS was feasible and effective.
Key concepts: Network traffic simulation, Computer science, Traffic simulation, Traffic generation model, Representation (politics), Consistency (knowledge bases), Intersection (aeronautics), Field (mathematics)