Citation :

Brigitte Paola Mamani Concha, Enrique Aurelio Amu Jimenez, Jesús Talavera Suarez, "Optimal Trajectory Prediction System for Automated Driving using Deep Learning," International Journal of Electronics and Communication Engineering, vol. 13, no. 9, pp. 147-159, 2026. Crossref, https://doi.org/10.14445/23488549/IJECE-V13I9P109

Abstract

Research in the field of autonomous driving has revealed that one of the main focuses is on predicting the future trajectories of vehicles. These accurate forecasts enable safe navigation and efficient route planning. A hybrid Convolutional Neural Network (CNN)-Long Short-Term Memory (LSTM) framework integrated with the Robot Operating System (ROS) is presented in this work. The aim of the framework is to estimate future vehicle trajectories in real time. The network was trained using vehicle motion data from the Argoverse Motion Forecasting dataset. CNN layers extract local temporal and kinematic patterns from input sequences, and LSTM layers model their temporal evolution. Once trained, the model is incorporated into an ROS-based pipeline where nodes dedicated to processing incoming trajectory information publish predicted paths for real-time visualisation in RViz. Performance of the model was evaluated using Mean Squared Error (MSE), Average Displacement Error (ADE) and Final Displacement Error (FDE), and was then compared to that of classical motion prediction methods. Research results are promising, as they enable the computational efficiency required for real-time operation alongside predictive accuracy. Additionally, robotic middleware, deep learning and a publicly available benchmark dataset were combined to develop a reproducible experimental framework that will facilitate the replication and evaluation of autonomous driving systems.

Keywords

Autonomous driving, Trajectory prediction, Deep Learning, CNN-LSTM, Simulation.

References

1. Ming-Fang Chang et al., “Argoverse: 3D Tracking and Forecasting with Rich Maps,” 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA, pp. 8748-8757, 2019.
   [CrossRef] [Google Scholar] [Publisher Link]
2. Matteo Lisotto, Pasquale Coscia, and Lamberto Ballan, “Social and Scene-Aware Trajectory Prediction in Crowded Spaces,” 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW), Seoul, Korea (South), pp. 2567-2574, 2019.
   [CrossRef] [Google Scholar] [Publisher Link]
3. Jing-Tai Chang, and Feng-Cheng Lin, “An Enhanced Time-Attention-Based Social GAN for Pedestrian Trajectory Prediction and Risk Alert System Utilizing Edge Computing Device,” 2025 Seventh International Symposium on Computer, Consumer and Control (IS3C), Taichung, Taiwan, pp. 1-4, 2025. [CrossRef] [Google Scholar] [Publisher Link]
4. Jianwei Tang et al., “Stochastic Human Motion Prediction with Memory of Action Transition and Action Characteristic,” 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA, pp. 1883-1893, 2025.
   [CrossRef] [Google Scholar] [Publisher Link]
5. Tim Salzmann et al., “Trajectron++: Dynamically-Feasible Trajectory Forecasting with Heterogeneous Data,” 16^th European Conference Computer Vision, Glasgow, UK, pp. 683-700, 2020.
   [CrossRef] [Google Scholar] [Publisher Link]
6. Sidra Rashid et al., “Optimizing Multi-Layer LSTM Based on Non-Local Spatiotemporal Interactions for Vehicle Trajectory Prediction,” IEEE Access, vol. 13, pp. 147690-147702, 2025.
   [CrossRef] [Google Scholar] [Publisher Link]
7. Alexey Dosovitskiy et al., “CARLA: An Open Urban Driving Simulator,” Proceedings of the 1^st Annual Conference on Robot Learning, vol. 78, pp. 1-16, 2017.
   [CrossRef] [Google Scholar] [Publisher Link]
8. Yongchao Zhang, and Pengzhan Chen, “Path Planning of a Mobile Robot for a Dynamic Indoor Environment Based on an SAC-LSTM Algorithm,” Sensors, vol. 23, no. 24, pp. 1-28, 2023.
   [CrossRef] [Google Scholar] [Publisher Link]
9. Mayank Bansal, Alex Krizhevsky, and Abhijit Ogale, “ChauffeurNet: Learning to Drive by Imitating the Best and Synthesizing the Worst,” arXiv preprint, pp. 1-20, 2018.
   [CrossRef] [Google Scholar] [Publisher Link]
10. Balakrishnan Varadarajan et al., “MultiPath++: Efficient Information Fusion and Trajectory Aggregation for Behavior Prediction,” 2022 International Conference on Robotics and Automation (ICRA), Philadelphia, PA, USA, pp. 7814-7821, 2022.
    [CrossRef] [Google Scholar] [Publisher Link]
11. Anis Koubaa, Robot Operating System (ROS): The Complete Reference, Springer International Publishing, 2016.
    [CrossRef] [Google Scholar] [Publisher Link]
12. Morgan Quigley, Brian Gerkey, and William D. Smart, Programming Robots with ROS: A Practical Introduction to the Robot Operating System, O’Reilly Media, pp. 1-448, 2015.
    [Google Scholar] [Publisher Link]
13. Alexandre Alahi et al., “Social LSTM: Human Trajectory Prediction in Crowded Spaces,” 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 961-971, 2016.
    [CrossRef] [Google Scholar] [Publisher Link]
14. Agrim Gupta et al., “Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks,” 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, pp. 2255-2264, 2018.
    [CrossRef] [Google Scholar] [Publisher Link]
15. Nachiket Deo, and Mohan M. Trivedi, “Convolutional Social Pooling for Vehicle Trajectory Prediction,” 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Salt Lake City, UT, USA, pp. 1549-15498, 2018.
    [CrossRef] [Google Scholar] [Publisher Link]
16. Yuning Chai et al., “MultiPath: Multiple Probabilistic Anchor Trajectory Hypotheses for Behavior Prediction,” arXiv preprint, pp. 1-14, 2019.
    [CrossRef] [Google Scholar] [Publisher Link]
17. Jiyang Gao et al., “VectorNet: Encoding HD Maps and Agent Dynamics from Vectorized Representation,” 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, pp. 11522-11530, 2020.
    [CrossRef] [Google Scholar] [Publisher Link]
18. Tung Phan-Minh et al., “CoverNet: Multimodal Behavior Prediction Using Trajectory Sets,” 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, pp. 14062-14071, 2020.
    [CrossRef] [Google Scholar] [Publisher Link]
19. Karttikeya Mangalam et al., “It Is Not the Journey but the Destination: Endpoint Conditioned Trajectory Prediction,” 16^th European Conference Computer Vision, Glasgow, UK, pp. 759-776, 2020.
    [CrossRef] [Google Scholar] [Publisher Link]
20. Ye Yuan et al., “AgentFormer: Agent-Aware Transformers for Socio-Temporal Multi-Agent Forecasting,” 2021 IEEE/CVF International Conference on Computer Vision (ICCV), Montreal, QC, Canada, pp. 9793-9803, 2021.
    [CrossRef] [Google Scholar] [Publisher Link]
21. Benjamin Coifman, and Lizhe Li, “A Critical Evaluation of the Next Generation Simulation (NGSIM) Vehicle Trajectory Dataset,” Transportation Research Part B: Methodological, vol. 105, pp. 362-377, 2017.
    [CrossRef] [Google Scholar] [Publisher Link]
22. Hao Cheng et al., “GATraj: A Graph- and Attention-Based Multi-Agent Trajectory Prediction Model,” ISPRS Journal of Photogrammetry and Remote Sensing, vol. 205, pp. 163-175, 2023.
    [CrossRef] [Google Scholar] [Publisher Link]