POSE: Phase I: Open Source Ecosystem for Accelerating Artificial Intelligence in Construction Engineering

NSF Award Search · 01002627DB NSF RESEARCH & RELATED ACTIVIT · $300,000 · view on nsf.gov ↗

Abstract

The Advanced Construction Image Dataset suite is one of the most well-known artificial intelligence (AI) dataset resources in the construction engineering domain. It provides construction-specific AI datasets (i.e., object detection, image captioning, and instance segmentation), annotation guidelines, and benchmark analyses to over 700 active users. This project seeks to further support and grow the dataset, broaden the userbase, and increase its open-source contributors. The ecosystem will establish a decentralized, self-sustaining hub that provides datasets, models, algorithms, documentation, and learning materials for the construction domain. It will also foster collaboration among academia, industry, government, and professional societies to accelerate AI innovation in the construction engineering domain. The project conducts an in-depth systematic analysis of the current technological landscape of construction data, evaluates construction-specific gaps in capabilities, and broadly surveys the community to understand potential avenues of future growth for the Advanced Construction Image Dataset. It also evaluates the viability of a hierarchical consortium governance model for the open-source ecosystem, with the goal of ensuring long-term quality, ethical compliance, and contributor engagement. The open-source ecosystem develops a security-by-design framework where artifacts (e.g., datasets, models, applications, and documentation) can be validated and shared across distributed contributors/users under well-defined data quality guidance and privacy compliance. The project pursues community building plan through workshops, competitions, and international collaborations. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Key facts

NSF award ID
2550033
Awardee
Michigan Technological University (MI)
SAM.gov UEI
GKMSN3DA6P91
PI
Bo Xiao
Primary program
01002627DB NSF RESEARCH & RELATED ACTIVIT
All programs
Estimated total
$300,000
Funds obligated
$300,000
Transaction type
Standard Grant
Period
05/01/2026 → 04/30/2027