Page 725 - AI for Good Innovate for Impact
P. 725
AI for Good Innovate for Impact
Use Case - 11: AI-Powered Space Optimization for Low-Income
Housing cities 4.8: Smart home/
Organization: VIT University
Country: India
Contact Person(s):
Contact:
Mithun Balaji V – mithunbalaji.v2022@ vitstudent .ac .in mailto:mithunbalaji.v2022@
vitstudent .ac .in
Sankar Raja S - sankar.raja2022@ vitstudent .ac .in
Mentor: Dr. Geetha S - geetha.s@vit.ac.in
1 Use Case Summary Table
Item Details
Category Smart home/cities
Problem Inefficient use of space in government-built low-income housing
Addressed
Key Aspects of AI-driven layout optimization, RL-based adaptive planning, constraint-
Solution based filtering, and CV for layout analysis
Technology Deep Learning, Reinforcement Learning, Constraint Solvers, Floor Plan
Keywords Generation, Computer Vision, 2D/3D Spatial Analysis
Data Availability Public (ZInD, FloorPlanCAD, HouseExpo, SYNBUILD-3D, UN-Habitat,
HUD)
Metadata (Type of 2D CAD floor plans, 3D building wireframes, image data (PNG/JPEG),
Data) spatial graphs, furniture dimensions, user preferences
Model Training and CNNs for layout analysis, RL (PPO/DQN) for layout optimization,
Fine-Tuning constraint filtering with OR-Tools
Testbeds or Pilot Concept tests using FloorPlanCAD and HouseExpo layouts; pilot simu-
Deployments lations planned with Indian low-income floor plans
Code repositories [1]
2 Use Case Description
2�1 Description
Context and Background:
689
