Feasibility of real-time graphical simulation for active monitoring of visibility-constrained construction processes

The lack of clear visibility and spatial awareness frequently results in construction accidents such as workers being struck by heavy equipment; and collisions between equipment and workers or between two pieces of equipment. In addition, certain processes such as excavation and drilling inherently pose constraints on equipment operators' abilities to clearly perceive and analyze their working environment. In this paper, the authors investigate the types of spatial interactions on construction sites and the need for graphical real-time monitoring. A computing framework is presented for monitoring interactions between mobile construction equipment and static job-site entities, workers, and other equipment. The framework is based on the use of sensor-based tracking, georeferenced models, and a resulting concurrent, evolving 3D graphical database. The developed framework enables a real-time 3D visualization scheme that provides equipment operators with graphical job-site views that are not possible through conventional on-site cameras. The two key parameters affecting a proximity monitoring framework's effectiveness are measurement error and latency. Measurement error refers to the error in proximity computation-with respect to ground truth or theoretically expected values. Latency is a difference in the time between when an event occurs in the real world and when a proximity monitoring framework provides output to warning systems that end users depend upon. Results from validation experiments conducted to analyze the achievable measurement error and latency of the monitoring framework using indoor GPS tracking as a ground truth system are also presented and discussed.