METDrive: Multi-modal End-to-end Autonomous Driving with Temporal Guidance
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916736270336000 |
|---|---|
| author | Guo, Ziang Lin, Xinhao Yagudin, Zakhar Lykov, Artem Wang, Yong Li, Yanqiang Tsetserukou, Dzmitry |
| author_facet | Guo, Ziang Lin, Xinhao Yagudin, Zakhar Lykov, Artem Wang, Yong Li, Yanqiang Tsetserukou, Dzmitry |
| contents | Multi-modal end-to-end autonomous driving has shown promising advancements in recent work. By embedding more modalities into end-to-end networks, the system's understanding of both static and dynamic aspects of the driving environment is enhanced, thereby improving the safety of autonomous driving. In this paper, we introduce METDrive, an end-to-end system that leverages temporal guidance from the embedded time series features of ego states, including rotation angles, steering, throttle signals, and waypoint vectors. The geometric features derived from perception sensor data and the time series features of ego state data jointly guide the waypoint prediction with the proposed temporal guidance loss function. We evaluated METDrive on the CARLA leaderboard benchmarks, achieving a driving score of 70%, a route completion score of 94%, and an infraction score of 0.78. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_12667 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | METDrive: Multi-modal End-to-end Autonomous Driving with Temporal Guidance Guo, Ziang Lin, Xinhao Yagudin, Zakhar Lykov, Artem Wang, Yong Li, Yanqiang Tsetserukou, Dzmitry Robotics Computer Vision and Pattern Recognition Multi-modal end-to-end autonomous driving has shown promising advancements in recent work. By embedding more modalities into end-to-end networks, the system's understanding of both static and dynamic aspects of the driving environment is enhanced, thereby improving the safety of autonomous driving. In this paper, we introduce METDrive, an end-to-end system that leverages temporal guidance from the embedded time series features of ego states, including rotation angles, steering, throttle signals, and waypoint vectors. The geometric features derived from perception sensor data and the time series features of ego state data jointly guide the waypoint prediction with the proposed temporal guidance loss function. We evaluated METDrive on the CARLA leaderboard benchmarks, achieving a driving score of 70%, a route completion score of 94%, and an infraction score of 0.78. |
| title | METDrive: Multi-modal End-to-end Autonomous Driving with Temporal Guidance |
| topic | Robotics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2409.12667 |