Incorporating Planning Intelligence into Deep Learning: A Planning Support Tool for Street Network Design

ABSTRACT

Deep learning applications in shaping ad hoc planning proposals are limited by the difficulty of integrating professional knowledge about cities with artificial intelligence. We propose a novel, complementary use of deep neural networks and planning guidance to automate street network generation that can be context-aware, learning-based, and user-guided. The model tests suggest that the incorporation of planning knowledge (e.g., road junctions and neighborhood types) in the model training leads to a more realistic prediction of street configurations. Furthermore, the new tool provides both professional and lay users an opportunity to systematically and intuitively explore benchmark proposals for comparisons and further evaluations.

KEYWORDS:

• Planning support system (PSS)
• applied urban model
• street network
• deep learning
• urban design

Disclosure Statement

No potential conflict of interest was reported by the authors.

Notes

1 Color channels in RGB (Red, Green, and Blue) images, coded on 256 levels from 0 to 255.

2 Network topology with varied types of links and nodes.

3 Altitude information of terrain.

4 The direction a slope is facing.

5 Provided by a joint endeavor of the National Aeronautics and Space Administration, the National Geospatial-Intelligence Agency, and the German and Italian Space Agencies.

6 To quantify the total pixel-to-pixel difference between the generated street network and the real-world network within the generation region.

7 Adversarial loss is considered and included in the loss functions of most image completion algorithms. By taking adversarial loss into account, the standard pixel-to-pixel loss minimization process can be turned into a min–max optimization problem in which the discriminator is jointly updated with the generation network at each iteration. This is crucial for our approach, due to the existence of multiple possible solutions.

Additional information

Notes on contributors

Zhou Fang

Zhou Fang is a PhD candidate in the Martin Centre for Architectural and Urban Studies at the University of Cambridge.

Ying Jin

Ying Jin is a professor in the Martin Centre for Architectural and Urban Studies at the University of Cambridge.

Tianren Yang

Tianren Yang is an assistant professor in the Department of Urban Planning and Design at the University of Hong Kong.