Authors:
Hyeyoung Kim
and
Chulmin Jun
Affiliation:
University of Seoul, Korea, Republic of
Keyword(s):
3D data model, Spatial DBMS, Pedestrian simulation, CA.
Related
Ontology
Subjects/Areas/Topics:
Agents
;
Artificial Intelligence
;
Artificial Intelligence and Decision Support Systems
;
Bioinformatics
;
Biomedical Engineering
;
Distributed and Mobile Software Systems
;
Enterprise Information Systems
;
Information Systems Analysis and Specification
;
Knowledge Engineering and Ontology Development
;
Knowledge-Based Systems
;
Methodologies and Technologies
;
Multi-Agent Systems
;
Operational Research
;
Simulation
;
Software Engineering
;
Symbolic Systems
Abstract:
Most crowd simulation models for pedestrian dynamics are based on analytical approach using experimental settings without being related to real world data. In order for the models to be adapted to real world applications such as fire evacuation or warning systems, some technical aspects first must be resolved. First, the base data should represent the 3D indoor model which contains semantic information of each space. Second, in order to communicate with the indoor localization sensors to capture the real time pedestrians and to store the simulation results for later uses, the data should be in a DBMS instead of files. The purpose of this paper is two folds. One is to suggest a DBMS-based 3D modeling approach for pedestrian simulations. The other is to improve the existing floor field based pedestrian model by modifying the dynamic field. We illustrated the data construction processes and simulations using the proposed DBMS approach and the enhanced pedestrian model.