Document Type
Article
Abstract
This paper explores the use of agent-based modeling (ABM), enhanced by reinforcement learning techniques such as Q- learning, to optimize urban transportation systems. The focus is on modeling and improving aspects such as traffic design, vehicular flow, and pedestrian mobility. The central research question is how to effectively simulate realistic agent behavior in order to develop models that can inform and support policy-making for more efficient and adaptive urban planning. The paper presents and analyzes simulation-based case studies that demonstrate how learning agents can repro- duce realistic movement patterns and provide insights for larger-scale urban systems.
Recommended Citation
Kalvoda, Tristan
(2026)
"Improving Urban Planning Through Agent Based Modeling and Q-Learning,"
Scholarly Horizons: University of Minnesota, Morris Undergraduate Journal: Vol. 13:
Iss.
2, Article 7.
DOI: https://doi.org/10.61366/2576-2176.1187
Available at:
https://digitalcommons.morris.umn.edu/horizons/vol13/iss2/7
Primo Type
Article