intelligent-transportation

🏆 Official implementation of LangCoop: Collaborative Driving with Natural Language

[ICLR'25] Official Implementation of STAMP: Scalable Task And Model-agnostic Collaborative Perception

Mobiletrans Plus Core

Official code for "Expression is enough: Improving traffic signal control with advanced traffic state representation ".

Official code for "Efficient Pressure: Improving efficiency for signalized intersections ".

Vehicle detection and traffic analysis system leveraging YOLOv11L, with data logging and dynamic visualization of traffic metrics

🚦 Next-generation AI Traffic Management System with real-time computer vision, reinforcement learning optimization, emergency vehicle detection, and immersive 3D visualization

Repo for the Bikestream project (aka Databike Pilot Project)

Multi-Vehicle V2V simulation platform with UDP mesh networking, custom 28-byte BSM protocol, collision prediction, and real-time dashboard telemetry.

This repository addresses popular content routing design for public transportation as discussed in my Ph.D. thesis titled as: "Popular Content Distribution in Public Transportation Using Artificial Intelligence Techniques". The code used for the entire content routing design is provided twice using two programming languages, namely: Python and M…

AI-powered Smart Traffic Management System built with Kotlin Multiplatform. Real-time traffic monitoring, adaptive signal control, and emergency vehicle prioritization across Android, iOS, Desktop, and Server platforms.

AI-powered adaptive traffic management and signal-control platform for mixed urban traffic

A high-precision real-time vehicle speed detection and monitoring system built with YOLOv8 and OpenCV. Features include dual-line accuracy, multi-class classification (Car, Bus, Truck), automated CSV data logging, and overspeed alerts

Physics-injected Graph Attention Networks for trajectory prediction in heterogeneous Indian urban traffic. M.Tech thesis, IIT Guwahati.

DeepTrafficQ is a reinforcement learning-based traffic signal control system that uses Deep Q-Networks (DQN) to minimize vehicle waiting times at a 4-way intersection. By leveraging Q-learning with experience replay and a convolutional neural network (CNN), the agent dynamically adjusts traffic light phases to optimize traffic flow.

Adaptive traffic signal management with YOLOv8 + SORT: real‑time vehicle counting, line crossing, and congestion‑aware timing

Digital twin framework for 6G ITS using mmWave sensing and beamforming

Developed an end-to-end machine learning solution to predict urban traffic congestion levels using data preprocessing, exploratory data analysis, feature engineering, and predictive modeling. The project enables intelligent traffic management, reduces congestion, and supports data-driven smart city transportation planning.

AI-powered real-time traffic monitoring and emergency priority system using FastAPI, YOLOv8, WebSockets, and React with live congestion, intersection intelligence, alerts, solar, and environmental monitoring.

A production-grade, AI-driven route optimization platform built for Jordan's terrain, leveraging Gemini 1.5 & Google Cloud to minimize fuel costs (JOD) and CO2 emissions.