Towards more accurate and robust vehicle trajectory prediction
Event details
| Date | 14.09.2023 |
| Hour | 15:00 › 17:00 |
| Speaker | Ahmad Rahimi |
| Location | |
| Category | Conferences - Seminars |
EDIC candidacy exam
Exam president: Prof. Robert West
Thesis advisor: Prof. Alexandre Alahi
Co-examiner: Prof. Antoine Bosselut
Abstract
This thesis focuses on enhancing the precision and robustness of vehicle trajectory prediction. Leveraging cutting-edge algorithms and data-driven techniques, the research delves deep into understanding the complexities of vehicular movements in diverse environments. By accounting for various factors like traffic conditions, roadway geometry, and driver behaviors, we aim to predict future trajectories with reduced error margins, paving the way for safer and more efficient transportation systems.
Background papers
1. Latent Variable Sequential Set Transformers For Joint Multi-Agent Motion Prediction
Roger Girgis, Florian Golemo, Felipe Codevilla, Martin Weiss, Jim A. D’Souza, Samira E. Kahou, Felix Heide, and Christopher Pal
https://arxiv.org/abs/2104.00563
2. Multimodal Trajectory Prediction Conditioned on Lane-Graph Traversals
Nachiket Deo, Eric M. Wolff, and Oscar Beijbom
https://arxiv.org/abs/2106.15004
3. Learning to summarize from human feedback
Nisan Stiennon, Long Ouyang, Jeff Wu, Daniel M. Ziegler, Ryan Lowe, Chelsea Voss, Alec Radford, Dario Amodei, and Paul Christiano
https://arxiv.org/abs/2009.01325
Exam president: Prof. Robert West
Thesis advisor: Prof. Alexandre Alahi
Co-examiner: Prof. Antoine Bosselut
Abstract
This thesis focuses on enhancing the precision and robustness of vehicle trajectory prediction. Leveraging cutting-edge algorithms and data-driven techniques, the research delves deep into understanding the complexities of vehicular movements in diverse environments. By accounting for various factors like traffic conditions, roadway geometry, and driver behaviors, we aim to predict future trajectories with reduced error margins, paving the way for safer and more efficient transportation systems.
Background papers
1. Latent Variable Sequential Set Transformers For Joint Multi-Agent Motion Prediction
Roger Girgis, Florian Golemo, Felipe Codevilla, Martin Weiss, Jim A. D’Souza, Samira E. Kahou, Felix Heide, and Christopher Pal
https://arxiv.org/abs/2104.00563
2. Multimodal Trajectory Prediction Conditioned on Lane-Graph Traversals
Nachiket Deo, Eric M. Wolff, and Oscar Beijbom
https://arxiv.org/abs/2106.15004
3. Learning to summarize from human feedback
Nisan Stiennon, Long Ouyang, Jeff Wu, Daniel M. Ziegler, Ryan Lowe, Chelsea Voss, Alec Radford, Dario Amodei, and Paul Christiano
https://arxiv.org/abs/2009.01325
Practical information
- General public
- Free