Research
Research
Our research spans cutting-edge areas in transportation engineering, data science, and artificial intelligence, with a focus on creating innovative solutions for real-world challenges.
Research
Our research spans cutting-edge areas in transportation engineering, data science, and artificial intelligence, with a focus on creating innovative solutions for real-world challenges.
Active Research
Innovative research projects and software development initiatives
Exploring cooperative perception for autonomous driving using the V2X-Real dataset. This project investigates vehicle-to-vehicle (V2V), vehicle-to-infrastructure (V2I), and infrastructure-to-infrastructure (I2I) communication strategies, training V2X-ViT baseline models on multi-agent 3D detection tasks with LiDAR and camera fusion.
Automating Pavement Condition Index (PCI) estimation using the Gemini large language model. The system identifies pavement type and distress categories, measures severity levels, computes deduct values from standard curves, and calculates corrected deduct values (CDV) to produce final PCI scores — streamlining what has traditionally been a manual inspection process.
Developing RCSNet (Road-Conditioned Spatiotemporal Network) for multi-city traffic flow forecasting. The architecture features four modules — Static Road Encoder, Temporal Traffic Encoder, Road-Guided Cross-Attention Fusion, and Road-Weighted Loss — to predict traffic patterns on high-resolution grids across cities including Moscow, Barcelona, Bangkok, and Chicago.
Finalizing a multi-class multi-object tracking framework designed to simultaneously detect and track diverse road users — vehicles, pedestrians, and cyclists — across complex traffic scenes. This project advances robust association and re-identification methods for real-time transportation monitoring.
Developing autonomous robotic systems to safely monitor and guide bison herds using computer vision and adaptive AI algorithms that respect natural behaviors.
Completed
A showcase of completed research and development initiatives
Enhancing construction zone safety with real-time monitoring systems that alert workers and drivers to potential hazards using IoT sensors and predictive analytics.
Implementing machine learning algorithms to predict and optimize electric grid performance for EV charging infrastructures, focusing on load balancing and efficiency.
Creating an integrated digital twin for urban planning and mobility optimization, connecting transportation systems, energy networks, and public services through a unified data platform for sustainability and efficiency.