A perception-free MLP reduces average L2 trajectory error by ~20% versus perception-based methods on nuScenes, suggesting current open-loop evaluation may reward trajectory mimicry over safe planning.
Pnpnet: End-to-end per- ception and prediction with tracking in the loop
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2023 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
Rethinking the Open-Loop Evaluation of End-to-End Autonomous Driving in nuScenes
A perception-free MLP reduces average L2 trajectory error by ~20% versus perception-based methods on nuScenes, suggesting current open-loop evaluation may reward trajectory mimicry over safe planning.