| id |
ecaade2023_52 |
| authors |
Le, Thanh-Luan and Kim, Sung-Ah |
| year |
2023 |
| title |
Game-based Platform for Daylight Analysis using Deep Learning |
| source |
Dokonal, W, Hirschberg, U and Wurzer, G (eds.), Digital Design Reconsidered - Proceedings of the 41st Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2023) - Volume 2, Graz, 20-22 September 2023, pp. 481–490 |
| doi |
https://doi.org/10.52842/conf.ecaade.2023.2.481
|
| summary |
Daylight analysis is not easy and requires skills in specific software and techniques and significant computation time. These skills are necessary for architecture education, but some students may find them challenging. For this reason, a software-free and simulation-free approach that quickly calculates daylight performance may be a more effective way for students to learn and practice architecture design. From these ideas, a game environment, which is familiar to the young generation, may enhance the excitement and engagement of education in this field. The development of a cubic builder game platform that utilizes the Deep Learning Model (DLM) to help users learn about daylight analysis within the game environment is currently underway. This paper presents the preliminary results of this study that focused on exploring methods for implementing and using DLM to predict daylight performance in a game environment. Using a drawing canvas, users can give design inputs in this environment. A framework involving three steps has been developed to combine data from the design and gaming environments. First, small-scale building models with specific design contexts and simulation data were created in Rhino and Grasshopper using LadyBugs and HoneyBee. Second, a DLM was trained on these data to make predictions. Last, developing the game environment with the well-trained DLM in Unity3D. Through analysis, the DLM's performance in game environments confirmed the potential of this approach. A building system will fully implement the game environment in future research. The DLM's predictive performance will be enhanced using more extensive and diverse data sets. |
| keywords |
Daylight Simulation, Architecture Education, Game-based, Unity3D, Deep Learning |
| series |
eCAADe |
| email |
sakim@skku.edu |
| full text |
 file.pdf (1,806,515 bytes) |
| references |
Content-type: text/plain
|
| last changed |
2023/12/10 10:49 |
|
