This article presents FollowGen, a conditional diffusion model for vehicle trajectory prediction in car-following scenarios. Unlike existing diffusion-based approaches that introduce conditions only during the denoising stage, FollowGen incorporates a scaled noise conditioning mechanism in the forward process to embed historical motion features, and employs a cross-attention transformer in the reverse process to explicitly model interactions between leading and following vehicles. Experiments demonstrate that FollowGen consistently outperforms state-of-the-art baselines, achieving higher accuracy and robustness in diverse car-following environments. 

To address the limited extensibility of standardized message format this study proposes a modular, edge-intelligent framework — The mobility Operating System (mOS) — integrated with a mixed-reality testbed for realistic validation of infrastructure-guided autonomous vehicle coordination. We analyzed to verify the feasibility of the C-V2X Framework for autonomous vehicle guidance in real-time at the physical testbed. The dataset was collected from the testbed experiment to analyze framework performance (speed profiles, post-encroachment time, and numerical error) and service performance (latency, jitter, and packet loss).

This repository contains all necessary assets to replicate the study "Multimodal traffic assignment from privacy-protected OD data."

📁 Data

Location: /data
The dataset includes the following components:

• Network Link Data: Parameters (cost, capacity, travel time) for each network link.

• Route Sets: Pre-defined sequences of links (routes) analyzed in the case studies.

💻 Code

Implementation: The full PPTA model, privacy mechanism, and evaluation framework.

• Language: Python 3.8+

• Dependencies: Install required libraries:

  pip install numpy pandas matplotlib scipy cvxpy

  Note: The Mosek solver requires a separate license and installation.

• Execution: Navigate to the PrivacyPreservingTrafficAssignment directory and run:

  python main_ppta_new.py

  This will execute the main script and reproduce the key results from the paper.

📄 Documentation

For a detailed explanation of the methodology, file descriptions, and step-by-step instructions, refer to:
replication-explanatory-file-commtr-PPTA.docx

‘final_df.csv’ includes travel time index (TTI) and all potential  influential factors for pre-lockdown, lockdown, and post-lockdown periods.

This package contains R codes and python code as well as model input data to conduct replication. An explanatory file has also been provided. 

This replication package accompanies the manuscript “What patterns contribute to autonomous vehicle crashes?  A study of Levels 2 and 4 automation using association rule analysis.”
It provides: (i) code to harmonize California DMV AV crash reports and NHTSA SGO crash reports into a unified, analysis-ready table; and (ii) scripts to run Apriori association rule mining with support/confidence/lift threshold

This is the code and datasets for the potential publication "Scalable and Reliable Multi-agent Reinforcement Learning for Traffic Assignment".

This replication package contains the source code, configuration files, input data, and result samples for the paper “Efficient and stable ride-pooling through a multi-level coalition formation game”. The package allows users to reproduce all experiments reported in the article. Instructions for environment setup and execution are provided in the included README and configuration files.

This is the code and datasets for the paper "Quantitative assessment of mid-air collision probability in urban air mobility: A safety barrier-based framework for integrated operations". This research proposes a method to systematically quantify the mid-air collision (MAC) risk for different operation types in urban air mobility (UAM).  

This study proposes a lane-changing decision and trajectory planning algorithm for intelligent vehicles on highways that takes driver behavior into consideration. A co-simulation model based on Prescan and Simulink was built to validate the designed trajectory planning algorithm. The code and data related to the proposed algorithm and its verification are also provided.

The is the code for paper "Towards fair lights: A multi-agent masked deep reinforcement learning for efficient corridor-level traffic signal control". Run train.py for deep reinforcement learning model training. Run test.py for testing.

The code is used in the paper:"Beyond Conventional Vision: RGB-Event Fusion for Robust Object Detection in Dynamic Traffic Scenarios"

This replication package includes the complete code and dataset used in our study titled “Can Combined Virtual-Real Testing Speed Up Autonomous Vehicle Testing? Findings from AEB Field Experiments." The materials provided are essential for researchers and practitioners interested in replicating our experiments and validating the findings.

该数据集包括 2023 年 10 月城市十字路口四个方向的监控视频。 这些视频的分辨率为 3840×2160，帧速率为每秒 30 帧 （fps），场景理解标注涵盖交通拥堵程度、行人和车辆数量、行为意图等正常场景。

This repository includes the replication code for utilizing the PFD method for traffic state estimation applications.

The replication package contains the matlab code and data to replicate the performed experiments described in the manuscript.

This dataset is the original data used in the paper Privacy-preserving Personalized Pricing and Matching for Ride-hailing Platforms. Based on the real demand information contained in this dataset, we generated the dataset used in our study. The generation process is detailed in the numerical experiments section of the paper.

Safety-critical scenario test for intelligent vehicles via hybrid participations of natural and adversarial agents

This repository contains the official code and datasets for A Knowledge-Informed Deep Learning Paradigm for Generalizable and Stability-Optimized Car-Following Models.