Automatic detection of hazardous scenarios during spatial interaction between tower cranes and workers
ZHANG Zhitian1, WANG Yuanyuan2, LUO Zhub1, GUO Ziyang1, GUO Hongling1
1. Department of Construction Management, Tsinghua University, Beijing 100084, China; 2. Construction Engineering Branch, The Third Highway Engineering Co., Ltd., Beijing 100020, China
Abstract：[Objective] Tower crane operations are characterized by long durations, extensive moving scopes, heavy loads, and complex spatial interactions with workers. These factors often contribute to construction accidents. Furthermore, construction workers standing under crane hooks and their lifting objects during the lifting process pose high safety risks, often encountering accidents such as collisions and object falling. Information technology plays a crucial role in enhancing tower crane monitoring and reducing workers' safety risks. Although existing studies on tower crane monitoring have made considerable advancements, they primarily focus on the operating state of cranes and overlook safety issues arising from interactions between cranes and workers. This study aims to employ the schedule information extracted from building information modeling (BIM) and computer vision and sensing technologies to propose an automatic hazard detection method for detecting dangerous scenarios during the lifting process in tower cranes. [Methods] This study develops an automatic detection framework for identifying hazardous scenarios involving spatial interaction between tower cranes and workers. This framework comprises four components. (1) Equipment installation and network environment establishment:cameras are installed at elevated positions to monitor the spatial locations of workers under the operating plane of a tower crane in real time. Furthermore, various sensors and cameras are fixed beneath the crane's trolley and cab to collect data regarding its operating status. A local area network is set up on the site to facilitate instantaneous data transmission. (2) Collection of tower crane operating data:the exact spatial location of the crane's hook is calculated using arm tracking and spatial trigonometric relations to determine its operating status. (3) Collection of workers' operational status data:advanced image recognition techniques are used to identify workers' positions, which are then converted into three-dimensional spatial coordinates through coordinate transformation. (4) Spatial relationship analysis and identification:precise spatial mapping of the tower crane's operating status and workers' positions is obtained using a unified BIM, followed by automatic detection according to predefined hazard assessment rules. [Results] The effectiveness and feasibility of the proposed method are validated by implanting it during a one-month real construction project. The analysis of data collected for 15 days reveals that the number of hazardous scenarios fluctuates considerably, peaking 523 times and plunging 35 times. These fluctuations correlate strongly with the number of workers on site, verifying the reliability of the proposed method and highlighting the need for intelligent hazardous scenario detection. Moreover, the results show that construction workers generally lack adequate awareness of the safety implications of tower crane trajectories. [Conclusions] This study successfully integrates BIM, sensing, and computer vision technologies to develop an automatic hazard detection method that focuses on the spatial interaction between tower cranes and workers. The proposed method enhances the timeliness and accuracy of hazard detection and provides innovative perspectives and technical support for construction site safety management. However, this study has certain limitations, such as data interferences caused by minor vibrations during tower crane operations to be further mitigated using noise reduction techniques in future research.
[1] 李春雷. 建筑塔吊监控系统远程终端的开发[D]. 成都:电子科技大学, 2010. LI C L. The development of remote terminal of tower crane monitoring system [D]. Chengdu:University of Electronic Science and Technology of China, 2010. (in Chinese) [2] 唐凯, 陈陆, 张洲境, 等. 我国建筑施工行业生产安全事故统计分析及对策[J]. 建筑安全, 2020, 35(9):40-43. TANG K, CHEN L, ZHANG Z J, et al. Statistical analysis and countermeasures of production safety accidents in construction industry in China [J]. Construction Safety, 2020, 35(9):40-43. (in Chinese) [3] 黄莺, 瑚珊, 姚思梦. 施工塔吊安全管理的系统动力学分析[J]. 安全与环境学报, 2020, 20(6):2060-2068. HUANG Y, HU S, YAO S M. System dynamics analysis and safety risk of building tower crane [J]. Journal of Safety and Environment, 2020, 20(6):2060-2068. (in Chinese) [4] Bureau of Labor Statistics, USA. Crane-related work deaths trended down from 1992 to 2017[R/OL]. (2019-06-26) [2023-05-01]. https://www.heavyequipmentforums.com/threads/crane-related-work-deaths-trended-down-from-1992-to-2017.77801/. [5] 周炜. 建筑工程施工塔吊安全风险分析与监控研究[D]. 武汉:华中科技大学, 2019. ZHOU W. Risk analysis and monitoring technologies of tower crane safety in construction engineering [D]. Wuhan:Huazhong University of Science and Technology, 2019. (in Chinese) [6] 胡振中, 张建平, 张新. 基于四维时空模型的施工现场物理碰撞检测[J]. 清华大学学报(自然科学版), 2010, 50(6):820-825. HU Z Z, ZHANG J P, ZHANG X. Construction collision detection for site entities based on 4-D space-time model [J]. Journal of Tsinghua University (Science & Technology), 2010, 50(6):820-825. (in Chinese) [7] 李英攀, 史明亮, 刘名强, 等. 基于Cloud-BIM和UWB的施工现场智能安全系统研究[J]. 中国安全生产科学技术, 2018, 14(9):151-157. LI Y P, SHI M L, LIU M Q, et al. Research on intelligent safety system of construction site based on Cloud-BIM and UWB [J]. Journal of Safety Science and Technology, 2018, 14(9):151-157. (in Chinese) [8] 周进. 塔吊安全监控保护系统及关键算法研究[D]. 哈尔滨:哈尔滨工业大学, 2010. ZHOU J. Monitoring and protection system for safety of tower crane and its key technologies [D]. Harbin:Harbin Institute of Technology, 2010. (in Chinese) [9] GHEISARI M, ESMAEILI B. Applications and requirements of unmanned aerial systems (UASs) for construction safety [J]. Safety Science, 2019, 118:230-240. [10] ZHONG D X, LV H Q, HAN J Q, et al. A practical application combining wireless sensor networks and internet of things:Safety management system for tower crane groups [J]. Sensors, 2014, 14(8):13794-13814. [11] WU H T, ZHONG B T, LI H, et al. On-site safety inspection of tower cranes:A blockchain-enabled conceptual framework [J]. Safety Science, 2022, 153:105815. [12] TEIZER J, VENUGOPAL M, WALIA A. Ultrawideband for automated real-time three-dimensional location sensing for workforce, equipment, and material positioning and tracking [J]. Transportation Research Record:Journal of the Transportation Research Board, 2008, 2081(1):56-64. [13] GUO H L, ZHANG Z T, YU R, et al. Action recognition based on 3D Skeleton and LSTM for the monitoring of construction workers' safety harness usage [J]. Journal of Construction Engineering and Management, 2023, 149(4):04023015. [14] 孙宏军, 赵作霖, 徐冠群. 塔式起重机机器视觉监控系统设计[J]. 传感器与微系统, 2016, 35(8):70-73. SUN H J, ZHAO Z L, XU G Q. Design of computer vision surveillance system for tower cranes [J]. Transducer and Microsystem Technologies, 2016, 35(8):70-73. (in Chinese) [15] 张锐. 基于机器视觉的塔式起重机控制策略研究[D]. 合肥:合肥工业大学, 2019. ZHANG R. Study on control strategy of tower crane based on machine vision [D]. Hefei:Hefei University of Technology, 2019. (in Chinese) [16] YANG Z, YUAN Y B, ZHANG M Y, et al. Safety distance identification for crane drivers based on Mask R-CNN [J]. Sensors, 2019, 19(12):2789. [17] 段锐, 邓晖, 邓逸川. ICT支持的塔吊安全管理框架:回顾与展望[J]. 图学学报, 2022, 43(1):11-20. DUAN R, DENG H, DENG Y C. Information communications technology assisted tower crane safety management:Review and prospect [J]. Journal of Graphics, 2022, 43(1):11-20. (in Chinese)
2024, Vol. 64 
