Research
Advancing safer, smarter, and more efficient transportation through artificial intelligence, sensing, data science, edge computing, and intelligent infrastructure.
Research Vision
Transportation systems are being transformed by rapid advances in artificial intelligence, connected and automated vehicles, multimodal sensing, edge computing, and vehicle-to-everything (V2X) communication. Roadside sensors, cameras, radar, connected vehicles, mobile devices, and other emerging technologies are generating unprecedented amounts of high-resolution transportation data.
The key challenge is no longer simply collecting data, but transforming these data into reliable, timely, and actionable intelligence that can improve transportation safety, mobility, efficiency, and infrastructure management.
Prof. Wang's research focuses on developing and deploying intelligent transportation technologies that integrate advanced sensing, artificial intelligence, edge computing, and transportation data science. The work spans fundamental research, technology development, field deployment, and close collaboration with public agencies and industry.
Research Areas
Developing AI and machine-learning methods for transportation perception, prediction, reasoning, decision-making, and control. Current interests include deep learning, foundation models, large language models, vision-language models, generative AI, reinforcement learning, and agentic AI.
Moving artificial intelligence from centralized computing to the transportation infrastructure itself. Edge AI enables roadside systems to sense, understand, communicate, and respond to traffic conditions in real time while reducing latency and communication requirements.
Developing advanced sensing systems that integrate cameras, radar, LiDAR, wireless sensing, connected-vehicle data, and other sources to detect, classify, track, and understand vehicles, pedestrians, bicyclists, roadway assets, and environmental conditions.
Developing methods for transportation data quality control, integration, management, analytics, forecasting, and knowledge extraction from large-scale and heterogeneous transportation data sources.
Using advanced sensing, connected-vehicle data, computer vision, machine learning, surrogate safety measures, and real-time analytics to understand crash risk and move transportation safety from reactive analysis toward proactive prevention.
Studying connected and automated vehicles, cooperative perception, vehicle trajectory prediction, human-machine interaction, V2X/I2X communication, and infrastructure support for increasingly automated transportation systems.
Developing data-driven and AI-enabled methods for traffic-state estimation, network-wide forecasting, traffic signal control, corridor management, truck parking, freeway operations, and transportation-system performance assessment.
Applying intelligent transportation technologies to urban, rural, and tribal transportation challenges, with emphasis on resilient infrastructure, equitable mobility, active transportation, and practical deployment in real-world environments.
Selected Projects
Developing deployable edge-AI methods for automated traffic-sign detection, recognition, and infrastructure asset monitoring.
Exploring low-cost intelligent roadway communication technologies that connect infrastructure, vehicles, and emergency-response systems through V2X communication.
Using AI and transportation data to forecast system performance and support infrastructure investment and management decisions.
Advancing sensing, prediction, information sharing, and intelligent management technologies to improve truck-parking availability and freight operations along the I-5 corridor.
Investigating connected and automated transportation technologies and their safe and effective deployment in tribal and rural communities.
Developing cooperative multitask sensing and I2X technologies for detecting vulnerable road users and improving pedestrian and bicyclist safety.
Innovation & Translation
A central goal of the STAR Lab is to move transportation research beyond algorithms and laboratory demonstrations into deployable technologies. Research systems are developed and tested with transportation agencies, communities, and industry partners in real operating environments.
Examples include intelligent roadside sensing, edge-AI traffic monitoring, cooperative perception, automated roadway-asset assessment, truck-parking information systems, and real-time safety applications.
Intellectual Property
Research conducted through the STAR Lab has contributed to patented technologies in roadway sensing, computer vision, transportation data collection, and mobile transportation systems.
U.S. Patent US12462674 B2, awarded in 2025.
U.S. Patent 12,608,907 B2, awarded in 2026.
U.S. Patent 8,358,808 B2.
Research Ecosystem
The Smart Transportation Applications and Research Laboratory (STAR Lab) brings together researchers from diverse disciplinary backgrounds, including transportation engineering, computer science, mechanical engineering, automation, electrical engineering, and related fields. This interdisciplinary environment enables the lab to approach transportation problems from multiple perspectives and to integrate advances in sensing, artificial intelligence, edge computing, automation, and data science. STAR Lab research is strongly motivated by real-world transportation challenges. Working closely with transportation agencies, nonprofit organizations, industry partners, and other stakeholders, the lab develops and deploys cutting-edge technologies and cost-effective solutions that can be implemented in practice. These partnerships help translate fundamental research into field-tested systems and measurable improvements in transportation safety, mobility, efficiency, and infrastructure management.