Agent-based Modeling of Spatial Configuration and Pedestrian Movement

• Main Authors: Jen-Der Chen, 陳正德
• Other Authors: Feng-Tyan Lin
• Format: Others
• Language: zh-TW
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/72307428054839792089

碩士 === 國立臺灣大學 === 建築與城鄉研究所 === 92 === This research aims to develop an agent-based system to simulate interaction of space and pedestrian. Based on theories of human cognition, the interaction model employs visual distance and area to guide pedestrian navigation. This research uses NetLogo as simulation platform where space and pedestrian are considered as patches and agents respectively. Agents move according to visual distance and adjust heading if their visual area above threshold. It is found that proportion of agents who are able to arrive their destinations and times they have spent varies with respect to thresholds of visual area. In aggregate level, five check points are set up to compare pedestrian flows observed and simulated. The simulation results show very good correlation where the maximum correlation coefficient is 0.99.