AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Risk-Aware Self-Consistent Imitation Learning for Trajectory Planning in Autonomous Driving

Yixuan Fan, Ya-Li Li, Shengjin Wang

ECCV 2024 · Probabilistic and scientific modeling › Molecules and proteins · approach: Multimodal trajectory prediction · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add risk-awareness TTC imitation (ablation)RaSc baseline · nuPlan Val14NR-CL score: 79 -> 83Improved
add self-consistency constraint (ablation)RaSc · nuPlan Val14NR-CL score: 83 -> 85Improved
add self-consistency-based OHEM (ablation)RaSc · nuPlan Val14NR-CL score: 85 -> 86Improved
use trajectory-loss-based OHEM instead of none (ablation)RaSc · nuPlan Val14NR-CL score: 85 -> 83Hurt
use cross-attention TTC decoder instead of summing agent tokens (ablation)RaSc · nuPlan Val14NR-CL score: 82 -> 86Improved
predict TTC at 1.0s rather than 2.0s (ablation)RaSc · nuPlan Val14NR-CL score: 82 -> 86Improved
predict TTC at 0.5s instead of 1.0s (ablation)RaSc · nuPlan Val14NR-CL score: 86 -> 84Hurt
add post-processing safety checksRaSc · nuPlan Test14-hard (reactive closed-loop)R-CL score: NR-CL 75 -> 77; OL 86 -> 82Mixed
RaSc versus PlanTF in reactive closed-loopRaSc · nuPlan Test14-random (reactive closed-loop)R-CL score: 81 -> 76Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy