Research on Tower Crane Transportation Path Guidance Integrating YOLOv8 and TC-DWA Algorithm

Proper guidance and operation of tower crane transportation are crucial for the safety and production efficiency of projects. However, traditional tower crane transportation relies heavily on manual experience, lacking intelligent decision support. To mitigate the risk of falls from height for workers during tower crane transportation, this paper proposes a dynamic tower crane transportation path guidance method that integrates computer vision technology and path planning algorithms. From the perspectives of worker safety and construction efficiency, this method addresses the issue of dynamic path guidance for tower crane transportation. By using cameras mounted above the hook to monitor the transportation process, the YOLOv8 (You Only Look Once version 8) object detection framework is employed to detect the position of the hook and workers during transportation. The study develops dynamic path guidance through the Tower Crane-Dynamic Window Approach (TC-DWA). Experimental results demonstrate that the dynamic transportation path guidance for tower cranes, integrating YOLOv8 and TC-DWA algorithms, can effectively reduce the risk of falls during crane transportation with only a minor increase in the transportation path.