Zero-shot trajectory generation with self-supervised diffusion model

Abstract

Vehicle trajectory data play a crucial role in modern intelligent transportation systems. Traditional generative models, such as Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), often struggle to accurately capture the distributional characteristics of trajectories in real-world scenarios. Moreover, conventional traffic trajectory generation methods face challenges due to their reliance on real-world road network data that are difficult to acquire and have limited availability. To address these issues, this study leverages the inherent correlation between surveillance camera data and vehicle trajectories, developing a zero-shot trajectory generation method (defined as generating valid trajectories without any complete single-vehicle trajectory annotations, only using aggregated traffic flow statistics and road network topology) based on a diffusion model. We treat surveillance traffic flow distributions as conditioning information to guide the denoising process continuously, employing an adaptive optimization framework to generate compliant and high-quality trajectories. All experiments are validated on real-world traffic datasets from Dadukou District, Chongqing. Notably, the entire model training process does not use any complete single-vehicle trajectory annotations, relying only on aggregated traffic flow statistics and public road network topology. Using surveillance traffic flow data as input, the generated synthetic trajectories not only align closely with real trajectories but also demonstrate comparable effectiveness. Extensive experiments validate the efficacy of the proposed method in zero-shot trajectory generation tasks, showing excellent performance in both trajectory similarity and distributional similarity.

Keywords

Vehicle trajectory generation / Diffusion model / Adaptive optimization framework / High-dimensional space sampling

References

[1] Shahraki, A., Abbasi, M., Taherkordi, A., & Jurcut, AD. (2022). A Comparative Study on Online Machine Learning Techniques for Network Traffic Streams Analysis. Computer Networks, 207, 108836.

[2] Chen, T., Kornblith, S., Norouzi, M., & Hinton, G. (2020). A Simple Framework for Contrastive Learning of Visual Representations. In Proceedings of the 37th InternationalConference onMachineLearning.PMLR.

[3] Choi, S., Kim, J., & Trajgail, Y. H. (2021). Generating Urban Vehicle Trajectories Using Generative Adversarial Imitation Learning. Transportation Research Part C-emerging Technologies, 128,103091

[4] El annas, M., Ouzineb, M., & Benyacoub, B. (2022). Hidden Markov Models Training Using Hybrid Baum Welch - Variable Neighborhood Search Algorithm. Statistics, Optimization and Information Computing, 10(1), 160–170.

[5] Grill, J.-B., Strub, F., Altché, F., Tallec, C., Richemond, P., Buchatskaya, E., Doersch, C., Avila Pires, B., Guo, Z., Gheshlaghi Azar, M., & et al. (2020). Bootstrap Your Own Latent: A New Approach to Self-Supervised Learning. In Conference on Neural Information Processing Systems.Curran Associates,Inc.

[6] He, K., Chen, X., Xie, S., Li, Y., Dollár, P., & Girshick, R. (2022). Masked Autoencoders Are Scalable Vision Learners. Computing Research Repository, 15979-15988.

[7] Ho, J., Jain, A., & Abbeel, P. (2020). Denoising Diffusion Probabilistic Models. Computing Research Repository. abs/2006.11239.

[8] Jepsen, T. S., Jensen, C. S., & Nielsen, T. D. (2020). Relational Fusion Networks: Graph Convolutional Networks for Road Networks. 1908(11567), 418-429.

[9] Jiang, Y., Chang, S., & Wang, Z. (2021). TransGAN: Two Pure Transformers Can Make One Strong GAN, and That Can Scale Up. NEURIPS,34,18082-18095.

[10] Jones GL, Qin Q. Markov Chain Monte Carlo in Practice. Annual review of statistics and its application, 2021, 9(1): 557-578

[11] Liang, X., Du, X., Wang, G., & Han, Z. (2019). A Deep Reinforcement Learning Network for Traffic Light Cycle Control. IEEE Transactions on Vehicular Technology, 68(2), 1243-1253.

[12] Liao, Z., Xiao, H., Liu, S., Liu, Y., & Yi, A. (2021). Impact Assessing of Traffic Lights Via GPS Vehicle Trajectories. ISPRS International Journal of Geo-Information, 10(11).769

[13] Malarz, D., Kasymov, A., Zięba, M., Tabor, J., & Spurek, P. (2025). Classifier-free Guidance with Adaptive Scaling. Computing Research Repository. abs/2502.10574.

[14] Niu, Z., Yu, K., & Wu, X. (2020). LSTM-Based VAE-GAN for Time-Series Anomaly Detection. Sensors, 20(13), 3738.

[15] Wang, W., Xia, F., Nie, H., Chen, Z., Gong, Z., Kong, X., & Wei, W. (2021). Vehicle Trajectory Clustering Based on Dynamic Representation Learning of Internet of Vehicles. 22(6), 3567-3576.

[16] Xinyu C., Jiajie X., Rui Z., Wei C., Junhua F., Chengfei L., & et al. (2021). TrajVAE: A Variational AutoEncoder Model for Trajectory Generation. Neurocomputing, 428, 332-339.

[17] Yang, L., Zhang, Z., Song, Y., Hong, S., Xu, R., Zhao, Y., Zhang, W., Cui, B., & Yang, M.-H. (2024). Diffusion Models: A Comprehensive Survey of Methods and Applications. ACM Computing Surveys, 56(4),1-39.

[18] Zeng, J., & Xiao, F. (2023). A Fractal Belief KL Divergence for Decision Fusion. Engineering Applications of Artificial Intelligence, 121,106027.

[19] Zhang, X., & Mahadevan, S. (2021). Bayesian Network Modeling of Accident Investigation Reports for Aviation Safety Assessment. Reliability Engineering & System Safety, 209,107321.

[20] Zhou, K., Huang, X., Song, Q., Chen, R., & Hu, Xia. (2022). Auto-GNN: Neural Architecture Search of Graph Neural Networks. Frontiers in BIG DATA, 5,1029307.

[21] Zhu, Y., Ye, Y., Zhang, S., Zhao, X., & Yu, J. (2023). DiffTraj: Generating GPS Trajectory with Diffusion Probabilistic Model. Computing Research Repository,36,65168-65188.

Funding

Natural Science Foundation of Sichuan Province(2025ZNSFSC0488)

RIGHTS & PERMISSIONS

The Author(s)