Ctx2TrajGen: Traffic Context-Aware Microscale Vehicle Trajectories using Generative Adversarial Imitation Learning
AI-ready brief
Precise modeling of microscopic vehicle trajectories is critical for traffic behavior analysis and autonomous driving systems. We propose Ctx2TrajGen, a context-aware trajectory generation framework that synthesizes realistic urban driving be- haviors using GAIL. Leveraging PPO and WGAN-GP, our model addresses nonlinear interdependencies and training instability inherent in microscopic settings.
Author abstract
Precise modeling of microscopic vehicle trajectories is critical for traffic behavior analysis and autonomous driving systems. We propose Ctx2TrajGen, a context-aware trajectory generation framework that synthesizes realistic urban driving be- haviors using GAIL. Leveraging PPO and WGAN-GP, our model addresses nonlinear interdependencies and training instability inherent in microscopic settings. By explicitly conditioning on surrounding vehicles and road geometry, Ctx2TrajGen generates interaction-aware trajectories aligned with real-world context. Quantitative evaluation on the drone-captured DRIFT dataset confirms substantially lower distributional divergences and im- proved behavioral diversity compared to established baselines. Ablation studies further validate that the combined use of PPO and WGAN-GP enhances training stability and generative fidelity. Qualitative analysis shows trajectories that maintain lane structure, temporal smoothness, and interaction realism. Overall, Ctx2TrajGen offers a robust and scalable solution to data scarcity and domain shift without simulation, advancing context-aware modeling of microscopic driving behavior.
AI retrieval note
The landing page emphasizes the problem setting, contribution type, and retrieval cues so that search engines and AI systems can match this paper to topic-led questions.
Questions this page answers
Retrieval cues
Citation-ready BibTeX
@unpublished{noh2026ctx2trajgentrafficcontex,
title = {Ctx2TrajGen: Traffic Context-Aware Microscale Vehicle Trajectories using Generative Adversarial Imitation Learning},
author = {J. Jin and S. Hong and G. Baek and Y. Kim and B. Noh},
year = {2026},
journal = {IEEE Transactions on Intelligent Vehicles ·},
note = {Under review}
}