A Deep Concept Graph Network for Interaction-Aware Trajectory Prediction
Publication
, Conference
Ban, Y; Li, X; Rosman, G; Gilitschenski, I; Meireles, O; Karaman, S; Rus, D
Published in: Proceedings - IEEE International Conference on Robotics and Automation
January 1, 2022
Temporal patterns (how vehicles behave in our observed past) underline our reasoning of how people drive on the road, and can explain why we make certain predictions about interactions among road agents. In this paper we propose the ConceptNet trajectory predictor - a novel prediction framework that is able to incorporate agent interactions as explicit edges in a temporal knowledge graph. We demonstrate the sample efficiency and the overall accuracy of the proposed approach, and show that using the graphical structure to explicitly model interactions enables better detection of agent interactions and improved trajectory predictions on a large real-world driving dataset.
Duke Scholars
Published In
Proceedings - IEEE International Conference on Robotics and Automation
DOI
ISSN
1050-4729
Publication Date
January 1, 2022
Start / End Page
8992 / 8998
Citation
APA
Chicago
ICMJE
MLA
NLM
Ban, Y., Li, X., Rosman, G., Gilitschenski, I., Meireles, O., Karaman, S., & Rus, D. (2022). A Deep Concept Graph Network for Interaction-Aware Trajectory Prediction. In Proceedings - IEEE International Conference on Robotics and Automation (pp. 8992–8998). https://doi.org/10.1109/ICRA46639.2022.9811567
Ban, Y., X. Li, G. Rosman, I. Gilitschenski, O. Meireles, S. Karaman, and D. Rus. “A Deep Concept Graph Network for Interaction-Aware Trajectory Prediction.” In Proceedings - IEEE International Conference on Robotics and Automation, 8992–98, 2022. https://doi.org/10.1109/ICRA46639.2022.9811567.
Ban Y, Li X, Rosman G, Gilitschenski I, Meireles O, Karaman S, et al. A Deep Concept Graph Network for Interaction-Aware Trajectory Prediction. In: Proceedings - IEEE International Conference on Robotics and Automation. 2022. p. 8992–8.
Ban, Y., et al. “A Deep Concept Graph Network for Interaction-Aware Trajectory Prediction.” Proceedings - IEEE International Conference on Robotics and Automation, 2022, pp. 8992–98. Scopus, doi:10.1109/ICRA46639.2022.9811567.
Ban Y, Li X, Rosman G, Gilitschenski I, Meireles O, Karaman S, Rus D. A Deep Concept Graph Network for Interaction-Aware Trajectory Prediction. Proceedings - IEEE International Conference on Robotics and Automation. 2022. p. 8992–8998.
Published In
Proceedings - IEEE International Conference on Robotics and Automation
DOI
ISSN
1050-4729
Publication Date
January 1, 2022
Start / End Page
8992 / 8998