Research on meta-model of driver behavior in agent-based traffic evacuation simulation
Shengcheng Yuan, Yi Liu, Gangqiao Wang, Yi Liu, Hui Zhang
Abstract
Shengcheng Yuan, Yi Liu, Gangqiao Wang, Yi Liu, Hui Zhang
Abstract
Driver behavior diversity affects traffic process greatly especially for the traffic evacuation process. Taking the driver behavior diversity into consideration, this paper developed a meta-model of driver behavior in microscopic traffic evacuation simulation, named DGIT frame, which are Decision, Game, Individual and Transform. DGIT frame integrates heterogeneous models in different levels into a unified workflow. By using this frame, driver behavior models can be coupled in various scenarios, between different agents, during multi-period and within big data resources. Based on integrating models with different structures or functions, DGIT may describe driver behavior diversity satisfactorily. A simulation system is developed in this paper with DGIT frame, and comparison simulation experiments were performed. Simulation results prove the capability of DGIT frame for describing diversity and evaluating plan and decision strategy.
OpenAlex reports 2 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.
Driver behavior diversity affects traffic process greatly especially for the traffic evacuation process. Taking the driver behavior diversity into consideration, this paper developed a meta-model of driver behavior in microscopic traffic evacuation simulation, named DGIT frame, which are Decision, Game, Individual and Transform. DGIT frame integrates heterogeneous models in different levels into a unified workflow. By using this frame, driver behavior models can be coupled in various scenarios, between different agents, during multi-period and within big data resources. Based on integrating models with different structures or functions, DGIT may describe driver behavior diversity satisfactorily. A simulation system is developed in this paper with DGIT frame, and comparison simulation experiments were performed. Simulation results prove the capability of DGIT frame for describing diversity and evaluating plan and decision strategy.
Key concepts: Computer science, Frame (networking), Process (computing), Workflow, Traffic simulation, Plan (archaeology), Diversity (politics), Simulation