文献类型：期刊文献

英文题名：Dynamic Collision Risk Warning in Construction Sites for Nonlinear Movements: A Visual-Spatial Fusion Multimodal Prediction Method

作者：Wang, Zeli[1];Li, Jue[2];Yan, Xuzhong[3];Li, Heng[4];Li, Jia[1]

机构：[1]East China Univ Sci & Technol, Dept Management Sci & Engn, Shanghai 200030, Peoples R China;[2]China Univ Geosci, Sch Econ & Management, Wuhan 430074, Peoples R China;[3]Zhejiang Univ Technol, Sch Management, Hangzhou 310014, Peoples R China;[4]Hong Kong Polytech Univ, Dept Bldg & Real Estate, Kowloon, Hong Kong 999077, Peoples R China

年份：2026

卷号：152

期号：10

外文期刊名：JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT

收录：;EI(收录号：20263121199641);Scopus(收录号：2-s2.0-105045813377);WOS:【SCI-EXPANDED(收录号:WOS:001850044000019)】；

基金：The authors are thankful for the financial support of the following grants: National Nature Science Foundation of China (No. 72304098), National Natural Science Foundation of China (No. 72201247), and Shanghai Pujiang Program (No. 22PJC030).

语种：英文

外文关键词：Collision prevention; Trajectory prediction; Transformer; Computer vision

摘要：Collisions between vehicles and workers at construction sites frequently occur at blind corners, where trajectories change abruptly, making it one of the major causes of severe safety incidents. Existing trajectory prediction methods struggle to accurately predict vehicle turning behavior, whereas distance-based collision warning systems frequently generate false alarms. These issues limit their effectiveness in safety management. To address these challenges, this study develops a visual-spatial fusion multimodal trajectory prediction method that can sense turning intent in advance and predict turning trajectories. An optimized computer vision detector enhanced with a high-frequency feature extraction branch enables robust small-target recognition for tower-crane viewpoints, ensuring stable long-term trajectory extraction. The proposed multimodal Transformer model incorporates dynamic localized environmental features that reflect road shape and obstacles, which strongly influence turning behavior on construction sites. Experiments using real construction video demonstrate that the method reduces mean trajectory prediction error by 12.2% relative to a trajectory-only Transformer baseline, with more than 14% improvement in turning scenarios. Furthermore, this study presents an end-to-end integration framework that maps predicted risk regions to physical warning devices in real time, enabling targeted, viewpoint-robust alarms without additional calibration even when camera viewpoints change. This study demonstrates that localized dynamic environmental information is a necessary predictor of nonlinear movement in field conditions and provides a deployable method for proactive collision prevention in dynamic construction sites, thus supporting the intelligent management of collision accidents at construction sites.