Search for: All records

This paper introduces the Intelligent Learning Platform for Robotics Operations (IL-PRO), a Virtual Reality (VR) system designed to enhance robotics training in the Architecture, Engineering, and Construction (AEC) industry. IL-PRO addresses the growing need for effective training methods as the AEC sector adopts robotic automation. The system integrates VR technology with game-assisted learning, combining online multimedia lessons for theory with immersive VR tasks for practical skills. Developed iteratively using Design-Based Research principles, IL-PRO incorporates realistic robot simulations and progressive task complexity. The VR environment, built in Unity, aims to enhance engagement, motor coordination, and spatial awareness in robotics training. While future goals include AI-driven personalized instruction, this work-in-progress focuses on VR curriculum development and implementation. The paper concludes by discussing future directions, including curriculum expansion and cross-institutional adoption, to establish new benchmarks in innovative robotics education for the AEC industry. 

This paper proposes a data-driven framework for quantifying disaster vulnerability using social media analytics, repurposing a previously collected Twitter dataset originally intended for evacuation behavior analysis. After refining the dataset to isolate signals of distress and need, a category based classification strategy is introduced in which thematic dictionaries guide the grouping of Tweets based on the semantic similarity of their embeddings. Focusing on Hurricane Dorian, a compound disaster during the COVID-19 pandemic characterized by high distress and negative sentiment, a weighted amplification factor is incorporated that prioritizes Tweet categories based on the immediacy of impact on human life, while normalizing by Tweet volume and population density. The resulting Media Impact Index (MII) is calculated at the Census Block Group (CBG) level for the United States. To demonstrate the cross-cultural flexibility of the pipeline, the same methodology is applied to Typhoon Hagibis in Japan, with a comparable vulnerability index generated at the district level. The findings suggest that the proposed framework can provide emergency management agencies with a scalable and adaptable tool for identifying and prioritizing vulnerable regions in diverse types of disasters and sociocultural contexts. 

Timely and precise forecasting of ground-level ozone (O3) remains a critical challenge in atmospheric science due to the compound influences of spatial heterogeneity and temporal variability. This paper introduces STCN-Ozone, a novel graph-based spatiotemporal framework specifically designed to model ozone dynamics by jointly learning spatial interactions and temporal evolution from real-world pollutant data. The model constructs geospatial graphs based on monitoring station proximity and applies spatial encoding through a Graph Convolutional Network (GCN). These spatial representations are temporally unfolded and modeled using a dilated Temporal Convolutional Network (TCN) to capture long-range temporal dependencies with high resolution. STCN-Ozone is trained on five years of ozone concentration data (1999–2003) across 451 unique locations, using 30-day sequences to predict the subsequent day’s ozone level. We use the model to estimate 95% confidence intervals to support uncertainty-aware forecasting. STCN-Ozone has been shown to be a scalable and domain-adaptive solution for operational atmospheric prediction systems.