The authors identify a previously underappreciated factor in the open-loop/closed-loop gap for end-to-end driving: standard waypoint-based displacement metrics fail to detect physically incoherent intermediate trajectories, even when the endpoint is accurate.
Trajectory optimization that preserves endpoints improves closed-loop driving
A lightweight post-processing layer repairs intermediate waypoints while keeping the predicted endpoint fixed, boosting scores for diverse end-to-end policies.
Big Tech
University of Toronto · Vector Institute · NVIDIA Research · ELLIS Institute Tübingen · KE:SAI
Research Digest··3 min read
Zhang et al.
Why this paper
From NVIDIA Research and 4 others
In one line
Endpoint-constrained optimization repairs intermediate waypoints while preserving predicted endpoints, substantially improving closed-loop performance across six end-to-end driving policies in two simulators.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ✓Reports numbers on named benchmarks (3 benchmarks)
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§