Digital Twin Floor Layout Generation via Activity Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current floor plan design tools require expertise and are inefficient in capturing user movement data to generate customizable and realistic floor plans, especially for residential and non-residential facilities, which limits their usability and quality.
Innovation Solution
A digital twin-based floor layout generation apparatus using a convolutional message passing network analyzer, image synthesizer, and discriminator to generate vectorized floor plans from human activity maps, enabling interactive and customizable design without human intervention, and converting 2D plans to 3D models for construction guidance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional CAD tools are used for floor plan design, then design flexibility is maintained, but the process requires high expertise and is time-consuming
Solution Approach 1:
The system enables self-service floor plan generation by automatically creating designs based on user inputs and activity maps without requiring expert intervention. The AI model generates floor plans autonomously, allowing users with minimal expertise to obtain professional-quality designs quickly.
Solution Approach 2:
The patent replaces the manual mechanical process of CAD drawing with an AI-based automated system. Instead of requiring users to manually operate complex CAD tools, the system uses machine learning models to automatically generate floor plans from activity maps and requirements, significantly reducing both time and expertise requirements.
2Productivity
If expert architects manually design floor plans, then design quality is high, but productivity is low
Solution Approach 1:
The system creates digital copies of expert architectural knowledge embedded in the AI model. By training on extensive floor plan data, the model captures and replicates expert design patterns, allowing automated generation of high-quality floor plans that reflect professional standards without requiring actual experts to perform each design manually.
Solution Approach 2:
The system changes the parameters of the design process from manual expert judgment to automated algorithmic optimization. By transforming qualitative expert knowledge into quantifiable parameters and optimization functions, the system achieves both high productivity through automation and high quality through systematic optimization of design parameters.
3Adaptability or versatility
If traditional design tools are used, then customization is possible, but the ability to capture and utilize user activity data is limited
Solution Approach 1:
The system implements feedback loops where user activity data captured from digital twins continuously informs and refines floor plan designs. Activity maps derived from sensor data and user behavior patterns provide feedback to the AI model, enabling iterative optimization of layouts to better match actual usage patterns and user preferences.
Solution Approach 2:
The system performs preliminary analysis of user activity data and digital twin information before generating floor plans. By pre-processing and understanding user behavior patterns, movement trajectories, and activity distributions in advance, the system can customize designs proactively to optimize for predicted user needs rather than reacting to requirements after design completion.
Data Source
AI summary
In some examples, digital twin-based floor layout generation may include receiving, for a floor plan that is to be generated, an activity map that includes movement of at least one user within a digital twin of a specified area. Based on the activity map, embedding vectors may be generated for each room type of a plurality of room types in the specified area. An input boundary feature map may be received. The floor plan may be generated based on an analysis of the embedding vectors for each room type of the plurality of room types and based on an analysis of the input boundary feature map.