| description abstract | Falls, struck-bys, and caught-in/betweens are among the most common types of fatal accidents on construction sites. Despite their significance, the majority of today’s accident prevention programs react passively to situations in which workers or equipment enter predefined unsafe zones. To support systems that proactively prevent these accidents, this paper presents a path prediction model for workers and equipment. The model leverages the extracted video frames to predict upcoming worker and equipment motion trajectories on construction sites. Specifically, the model takes two-dimensional (2D) tracks of workers and equipment from visual data—based on computer vision methods for detection and tracking—and uses a long short-term memory (LSTM) encoder-decoder followed by a mixture density network (MDN) to predict their locations. A multihead prediction module is introduced to predict locations at different future times. The method is validated on an existing dataset, TrajNet, and a new dataset of 105 high-definition videos recorded over 30 days from a real-world construction site. On the TrajNet dataset, the proposed model significantly outperforms Social LSTM. On the new dataset, the presented model outperforms conventional time-series models and achieves average localization errors of 7.30, 12.71, and 24.22 pixels for 10, 20, and 40 future steps, respectively. The benefits and limitations of the method to worker and equipment path prediction are discussed. | |
