Research Area Description

Our research group is dedicated to pioneering intelligent infrastructure management systems that enhance the resilience, safety, and operational efficiency of urban transportation networks. By bridging the gap between advanced artificial intelligence and civil engineering, we develop robust diagnostic frameworks for the life-cycle management of pavement and pedestrian infrastructure, addressing the evolving demands of smart city mobility.

Scope of Research

Pavement Condition & Structural Assessment

We specialize in non-destructive testing (NDT), particularly leveraging Ground Penetrating Radar (GPR) and advanced signal processing to estimate pavement layer thickness and dielectric constants. Our work encompasses the structural health monitoring of asphalt and concrete pavements, including automated rutting measurement and surface condition modeling.

Infrastructure Diagnostic Technologies

We utilize machine learning and computer vision to automate the detection and characterization of infrastructure distresses. This includes automated crack segmentation, rebar recognition in concrete bridge decks, and the use of vibroacoustic systems for subsurface defect identification.

Pedestrian Path Mobility & Sidewalk Quality

We prioritize the development of inclusive urban environments through the automated monitoring of sidewalk infrastructure. Our frameworks utilize instance segmentation and object detection to identify sidewalk defects, directly contributing to improved pedestrian safety and accessible urban mobility.

Key Contributions

Generative AI & Synthetic Data

We address the challenge of data scarcity by developing synthetic datasets (e.g., AI500) and leveraging generative AI to improve the accuracy and data-efficiency of deep learning paradigms in pavement crack segmentation.

Multimodal Large Language Models (LLMs)

We are pioneering the use of LLMs as "synthetic experts" to create validation frameworks for pavement performance assessment, modernizing traditional inspection workflows.

Data-Driven Optimization

Our group provides integrated decision-making approaches that combine multi-criteria decision-making (MCDM), game theory, and optimization algorithms to enhance project scheduling, cash-flow management, and maintenance prioritization for transportation infrastructure portfolios.