Closed-loop Planning on NAVSIM v1
100NCHuman
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| HumanVenue=-, Sensors=-2026.02 | 100 | 100 | 100 | 99.9 | 87.5 | 94.8 | |
| DriveWorld-VLAVenue=-, Sensors=3×C2026.02 | 99.1 | 98.2 | 96.1 | 100 | 85.9 | 91.3 | |
| DrivingGPTVenue=ICCV 2025, Sensors=1×C2026.02 | 98.9 | 90.7 | 94.9 | 95.6 | 79.7 | 82.4 | |
| DriveVLA-W0Venue=ICLR 2026, Sensors=1×C2026.02 | 98.7 | 99.1 | 95.3 | 99.3 | 83.3 | 90.2 | |
| WoTEVenue=ICCV 2025, Sensors=3×C + L2026.02 | 98.5 | 96.8 | 94.4 | 99.9 | 81.9 | 88.3 | |
| AutoVLAVenue=NeurIPS 2025, Sensors=3×C2026.02 | 98.4 | 95.6 | 98 | 99.9 | 81.9 | 89.1 | |
| RoCA-Orionsource_dataset=Bench2Drive, target_dataset=NAVSIM v1, evaluation_mode=Cross-domain closed-loop, base_planner=Orion2025.06 | 98.4 | — | — | — | 82.6 | 87.9 | |
| Hydra-MDPVenue=arXiv 2024, Sensors=3×C + L2026.02 | 98.3 | 96 | 94.6 | 100 | 78.7 | 86.5 | |
| DiffusionDriveVenue=CVPR 2025, Sensors=3×C + L2026.02 | 98.2 | 96.2 | 94.7 | 100 | 82.2 | 88.1 | |
| RecogdriveVenue=ICLR 2026, Sensors=3×C2026.02 | 98.2 | 97.8 | 95.2 | 99.8 | 83.5 | 89.6 | |
| RoCA-SparseDrivesource_dataset=Bench2Drive, target_dataset=NAVSIM v1, evaluation_mode=Cross-domain closed-loop, base_planner=SparseDrive2025.06 | 98 | — | — | — | 77.5 | 85.8 | |
| PARA-DriveVenue=CVPR 2024, Sensors=6×C2026.02 | 97.9 | 92.4 | 93 | 99.8 | 79.3 | 84 | |
| EponaVenue=ICCV 2025, Sensors=3×C2026.02 | 97.9 | 95.1 | 93.8 | 99.9 | 80.4 | 86.2 | |
| Orionsource_dataset=Bench2Drive, target_dataset=NAVSIM v1, evaluation_mode=Cross-domain closed-loop2025.06 | 97.9 | — | — | — | 78.7 | 84.4 | |
| TransFuserVenue=TPAMI 2023, Sensors=3×C + L2026.02 | 97.7 | 92.8 | 92.8 | 100 | 79.2 | 84 | |
| SparseDrivesource_dataset=Bench2Drive, target_dataset=NAVSIM v1, evaluation_mode=Cross-domain closed-loop2025.06 | 97.1 | — | — | — | 74.2 | 82.1 | |
| LAWVenue=ICLR 2024, Sensors=1×C2026.02 | 96.4 | 95.4 | 88.7 | 99.9 | 81.7 | 84.6 |