Vehicle Trajectory Prediction Using Residual Diffusion Model Based on Image Information
Source abstract
Abstract
In automatic driving, accurate prediction of vehicle trajectory is the key to achieve automatic driving, and multi-vehicle joint trajectory prediction has become an important part of modern human-computer interaction systems such as automatic driving. In order to better predict vehicle trajectories, we propose a new residual diffusion model to infer the joint distribution of future multi-vehicle trajectories. This approach has several major advantages. First, the model is able to learn multiple probability distributions from trajectory data to obtain potential outcomes for vehicles to multiple future trajectories. Secondly, in order to integrate the motion characteristics of multiple vehicles in the same scene, we use the method of combining the reference denoising and multiple residual denoising to improve the model performance and prediction speed. Finally, on this basis, a general trajectory constraint function is introduced, so that the generated trajectories of multiple vehicles will not collide with each other. We perform a rich experimental comparison of various existing methods on the NGSIM dataset and demonstrate that the proposed algorithm achieves a 26% improvement on mAP.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Autonomous Vehicle Technology and Safety · Traffic and Road Safety · Traffic control and management
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Shanghai Institute of Microsystem and Information Technology, Shanghai 201800, China
Department of Electronic Information, University of Chinese Academy of Sciences, Beijing 100000, China
Department of Electronic Information, University of Science and Technology of China, Hefei 230031, China
Department of Electronic Information, ShanghaiTech University, Shanghai 201800, China
References (39)
Show source-provided references
1. Mercat, J., Zoghby, N.E., Sandou, G., Beauvois, D., and Gil, G.P. (2019). Kinematic single vehicle trajectory prediction baselines and applications with the NGSIM dataset. arXiv.
2. Probabilistic analysis of dynamic scenes and collision risks assessment to improve driving safety10.1109/MITS.2011.942779 ↗
3. Alahi, A., Goel, K., Ramanathan, V., Robicquet, A., Fei-Fei, L., and Savarese, S. (2016, January 27–30). Social lstm: Human trajectory prediction in crowded spaces. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.10.1109/CVPR.2016.110 ↗
4. Gupta, A., Johnson, J., Fei-Fei, L., Savarese, S., and Alahi, A. (2018, January 18–23). Social gan: Socially acceptable trajectories with generative adversarial networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.10.1109/CVPR.2018.00240 ↗
5. Multimodal deep generative models for trajectory prediction: A conditional variational autoencoder approach10.1109/LRA.2020.3043163 ↗
6. Oh, G., and Peng, H. (2022). Cvae-h: Conditionalizing variational autoencoders via hypernetworks and trajectory forecasting for autonomous driving. arXiv.
7. Deo, N., and Trivedi, M.M. (2018, January 18–23). Convolutional social pooling for vehicle trajectory prediction. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, Salt Lake City, UT, USA.10.1109/CVPRW.2018.00196 ↗
8. Shalev-Shwartz, S., Shammah, S., and Shashua, A. (2016). Safe, multi-agent, reinforcement learning for autonomous driving. arXiv.
9. Palanisamy, P. (2020, January 19–24). Multi-agent connected autonomous driving using deep reinforcement learning. Proceedings of the 2020 International Joint Conference on Neural Networks (IJCNN), Glasgow, UK.10.1109/IJCNN48605.2020.9207663 ↗
10. Song, Y., Sohl-Dickstein, J., Kingma, D.P., Kumar, A., Ermon, S., and Poole, B. (2020). Score-based generative modeling through stochastic differential equations. arXiv.
11. Denoising diffusion probabilistic models
12. Ho, J., Chan, W., Saharia, C., Whang, J., Gao, R., Gritsenko, A., Kingma, D.P., Poole, B., Norouzi, M., and Fleet, D.J. (2022). Imagen Video: High Definition Video Generation with Diffusion Models. arXiv.
13. Ishiguro, K., Ujihara, K., Sawada, R., Akita, H., and Kotera, M. (2020). Data transfer approaches to improve seq-to-seq retrosynthesis. arXiv.
14. Song, J., Meng, C., and Ermon, S. (2020). Denoising diffusion implicit models. arXiv.
15. Salimans, T., and Ho, J. (2022). Progressive distillation for fast sampling of diffusion models. arXiv.
16. Gu, T., Chen, G., Li, J., Lin, C., Rao, Y., Zhou, J., and Lu, J. (2022, January 18–24). Stochastic trajectory prediction via motion indeterminacy diffusion. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.10.1109/CVPR52688.2022.01660 ↗
17. Mao, W., Xu, C., Zhu, Q., Chen, S., and Wang, Y. (2023, January 17–24). Leapfrog diffusion model for stochastic trajectory prediction. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada.10.1109/CVPR52729.2023.00534 ↗
18. Nachiket, D., Wolff, E., and Beijbom, O. (2021, January 8–11). Multimodal trajectory prediction conditioned on lane-graph traversals. Proceedings of the 5th Conference on Robot Learning, PMLR, London, UK.
19. Amirian, J., Hayet, J.B., and Pettré, J. (2019, January 15–20). Social ways: Learning multi-modal distributions of pedestrian trajectories with gans. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, Long Beach, CA, USA.10.1109/CVPRW.2019.00359 ↗
20. Forecasting: Theory and practice10.1016/j.ijforecast.2021.11.001 ↗
21. Nichol, A.Q., and Dhariwal, P. (2021, January 18–24). Improved denoising diffusion probabilistic models. Proceedings of the International Conference on Machine Learning, PMLR, Virtual.
22. Zhong, Z., Rempe, D., Xu, D., Chen, Y., Veer, S., Che, T., Ray, B., and Pavone, M. (June, January 29). Guided conditional diffusion for controllable traffic simulation. Proceedings of the 2023 IEEE International Conference on Robotics and Automation (ICRA), London, UK.10.1109/ICRA48891.2023.10161463 ↗
23. Chen, J., Wang, W., Chen, J., and Cai, M. (2023, January 4–10). Dynamic Vehicle Graph Interaction for Trajectory Prediction Based on Video Signals. Proceedings of the ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Island, Greece.10.1109/ICASSP49357.2023.10095089 ↗
24. Wang, M., Zhu, X., Yu, C., Li, W., Ma, Y., Jin, R., Ren, X., Ren, D., Wang, M., and Yang, W. (June, January 29). Ganet: Goal area network for motion forecasting. Proceedings of the 2023 IEEE International Conference on Robotics and Automation (ICRA), London, UK.10.1109/ICRA48891.2023.10160468 ↗
25. Chung, H., Sim, B., and Ye, J.C. (2022, January 18–24). Come-closer-diffuse-faster: Accelerating conditional diffusion models for inverse problems through stochastic contraction. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.10.1109/CVPR52688.2022.01209 ↗
26. Kawar, B., Elad, M., Ermon, S., and Song, J. (2022). Denoising diffusion restoration models. arXiv.
27. Zhao, H., Gao, J., Lan, T., Sun, C., Sapp, B., Varadarajan, B., Shen, Y., Shen, Y., Chai, Y., and Schmid, C. (2020, January 16–18). Tnt: Target-driven trajectory prediction. Proceedings of the Conference on Robot Learning, PMLR, Virtual.
28. Wu, J., Qiao, S., Li, H., Sun, B., Gao, F., Hu, H., and Zhao, R. (2024). Goal-Guided Graph Attention Network with Interactive State Refinement for Multi-Agent Trajectory Prediction. Sensors, 24.10.3390/s24072065 ↗
29. Nayakanti, N., Al-Rfou, R., Zhou, A., Goel, K., Refaat, K.S., and Sapp, B. (2023). Wayformer: Motion forecasting via simple & efficient attention networks. Proceedings of the 2023 IEEE International Conference on Robotics and Automation (ICRA), London, UK, 29 May–2 June 2023, IEEE.10.1109/ICRA48891.2023.10160609 ↗
30. On the assessment of vehicle trajectory data accuracy and application to the Next Generation SIMulation (NGSIM) program data10.1016/j.trc.2010.12.007 ↗
31. Sta: Spatial-temporal attention for large-scale video-based person re-identification
32. Salzmann, T., Ivanovic, B., Chakravarty, P., and Pavone, M. (2020, January 23–28). Trajectron++: Dynamically-feasible trajectory forecasting with heterogeneous data. Proceedings of the Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK. Proceedings, Part XVIII 16.10.1007/978-3-030-58523-5_40 ↗
33. Gao, J., Sun, C., Zhao, H., Shen, Y., Anguelov, D., Li, C., and Schmid, C. (2020, January 13–19). Vectornet: Encoding hd maps and agent dynamics from vectorized representation. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.10.1109/CVPR42600.2020.01154 ↗
34. Liang, M., Yang, B., Hu, R., Chen, Y., Liao, R., Feng, S., and Urtasun, R. (2020, January 23–28). Learning lane graph representations for motion forecasting. Proceedings of the Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK. Proceedings, Part II 16.
35. Varadarajan, B., Hefny, A., Srivastava, A., Refaat, K.S., Nayakanti, N., Cornman, A., Chen, K., Douillard, B., Lam, C.P., and Anguelov, D. (2022, January 23–27). Multipath++: Efficient information fusion and trajectory aggregation for behavior prediction. Proceedings of the 2022 International Conference on Robotics and Automation (ICRA), Philadelphia, PA, USA.10.1109/ICRA46639.2022.9812107 ↗
36. Gu, J., Sun, C., and Zhao, H. (2021, January 11–17). Densetnt: End-to-end trajectory prediction from dense goal sets. Proceedings of the IEEE/CVF International Conference on Computer Vision, Montreal, BC, Canada.10.1109/ICCV48922.2021.01502 ↗
37. Luo, W., Park, C., Cornman, A., Sapp, B., and Anguelov, D. (2022, January 14–18). Jfp: Joint future prediction with interactive multi-agent modeling for autonomous driving. Proceedings of the Conference on Robot Learning, PMLR, Auckland, New Zealand.
38. Liu, Y., Zhang, J., Fang, L., Jiang, Q., and Zhou, B. (2021, January 20–25). Multimodal motion prediction with stacked transformers. Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA.10.1109/CVPR46437.2021.00749 ↗
39. Ramesh, A., Dhariwal, P., Nichol, A., Chu, C., and Chen, M. (2022). Hierarchical text-conditional image generation with clip latents. arXiv.