Construction progress verification is a core task in engineering management, but traditional manual inspection is inefficient, lagging, and subjective, and carries safety risks. This paper, through a method of literature review, explores how the integration of Building Information Modeling (BIM) and unmanned aerial vehicle (UAV) vision can automate construction progress verification. Based on 24 listed studies, the review organizes the field around three fundamental problems: how site data are acquired, how data are converted and aligned with BIM, and how progress is automatically identified. It finds that point-cloud registration identifies components with accuracy above 95% in a bridge case, a deep-learning study obtains mAP@0.5 of 0.989, and a wind-solar-storage project reports about 400% higher acquisition efficiency with early-warning response within one hour. The paper concludes that the technology has moved from manual assistance to automated monitoring, yet the transition from progress monitoring to compliance-based verification remains unresolved, and future work should move toward digital twin closure, multi-modal fusion, and edge intelligence.
No takes yet. Share an insight, caveat, or question.
Chenyi Lu (2026) studied this question.