端到端自动驾驶新框架！PrismAD把规划器拆成三位专家，转弯碰撞率降低83.3%！
Ranking
Overall
58
Content
70
Popularity
31
Observed public metrics from 1 member.
Merged summary
TL;DR - PrismAD is an end-to-end driving framework that separates planning into interaction, road-geometry, and navigation-intent experts, then dynamically routes their outputs. It substantially reduced collisions in turning scenarios while producing modest overall trajectory-error gains.
- Each expert is a complete motion-prediction and planning branch with independent parameters, rather than a conventional MoE feed-forward layer.
- Separate routing weights are learned for predicting other agents and planning the ego vehicle’s trajectory.
- On Turning-nuScenes, collision rates fell from 0.40% to 0.07% with SparseDrive and from 0.06% to 0.01% with DiffusionDrive.
- Ablations indicate geometry and intent chiefly improve safety, while learned noisy routing outperforms uniform expert averaging.
Sources (1)
端到端自动驾驶新框架！PrismAD把规划器拆成三位专家，转弯碰撞率降低83.3%！
Public signals
Semantic Scholar citations 0 · Semantic Scholar influential citations 0
TL;DR - PrismAD is an end-to-end driving framework that separates planning into interaction, road-geometry, and navigation-intent experts, then dynamically routes their outputs. It substantially reduced collisions in turning scenarios while producing modest overall trajectory-error gains.
- Each expert is a complete motion-prediction and planning branch with independent parameters, rather than a conventional MoE feed-forward layer.
- Separate routing weights are learned for predicting other agents and planning the ego vehicle’s trajectory.
- On Turning-nuScenes, collision rates fell from 0.40% to 0.07% with SparseDrive and from 0.06% to 0.01% with DiffusionDrive.
- Ablations indicate geometry and intent chiefly improve safety, while learned noisy routing outperforms uniform expert averaging.