| id |
cdrf2023_125 |
| authors |
Yubo Liu, Zhilan Zhang, Kai Hu, Qiaoming Deng |
| year |
2023 |
| title |
Graph Constrained Multiple Schemes Generation for Campus Layout |
| doi |
https://doi.org/https://doi.org/10.1007/978-981-99-8405-3_11
|
| source |
Proceedings of the 2023 DigitalFUTURES The 5st International Conference on Computational Design and Robotic Fabrication (CDRF 2023) |
| summary |
The campus layout is a major stage in the early stages of campus planning and design. When assessing the feasibility of a campus site in stage, we usually compare multiple campus layout schemes, which consumes a lot of time. The design process can be accelerated if multiple campus planning schemes can be generated quickly to meet the desired requirements. This study aims to explore the possibility of using graph neural networks (GNN) to generate multiple campus layouts. We use a step-by-step generation method. The first step is generating campus functional zonings based on user constraints. The second step is generating campus building layouts based on the functional zonings. Ultimately the machine is able to quickly generate multiple campus layout schemes by user input of graph constraints such as the number of functional zonings, the type of functions and their adjacency. In the experiment, we trained 200 campus layout samples and verified the validity and accuracy of the experiment after qualitative and quantitative analysis. |
| series |
cdrf |
| email |
|
| full text |
 file.pdf (5,559,433 bytes) |
| references |
Content-type: text/plain
|
| last changed |
2024/05/29 14:04 |
|
