A cognitive architecture for intelligent urban planning
OPEN ACCESS
Loading...
Author / Creator
Date
2025
Publication Type
Conference Paper
ETH Bibliography
yes
Citations
Altmetric
OPEN ACCESS
Data
Rights / License
Abstract
Urban planning requires complex reasoning across interconnected systems such as transportation, demographics, land use, and environmental factors. Artificial intelligence can integrate diverse urban data and models, but current tools often lack domain grounding, memory, or simulation integration needed for proper planning support.
We present Echo, a cognitive architecture embedded in digital twin model of a transport system, that combines an urban knowledge graph, multi-memory reasoning (semantic, episodic, working), and a multi-agent controller to answer natural-language planning queries grounded in explicit sources. Echo lets agents traverse an ontology, retrieve spatio-temporal relationships, and incorporate simulation feedback into explanations.
In a transport planning case study, Echo produced richer, more reliable answers than standard approaches (RAG, ReAct), achieving an 88% first-attempt accuracy when validated by domain experts, at the cost of higher per-query computation. These results advance practical, source-grounded decision support for urban planners.
Permanent link
Publication status
published
External links
Editor
Book title
Journal / series
Volume
Pages / Article No.
Publisher
ETH Zurich
Event
1st International Conference on Artificial Intelligence and Innovative Applications (AIIA'25)
Edition / version
Methods
Geographic location
Date collected
Date created
Subject
Urban planning; Artificial intelligence; Cognitive architecture; Urban knowledge graph; Transport planning; Accuracy; Reasoning quality; Computational Time; Urban scenario analysis; Multi-agent
Organisational unit
09611 - Corman, Francesco / Corman, Francesco
02655 - Netzwerk Stadt u. Landschaft ARCH u BAUG / Network City and Landscape ARCH and BAUG
Notes
Conference lecture held on December 3, 2025