Diffusion models trained separately on static and dynamic scenes can be composed at test time to plan collision-free, kinematically feasible trajectories in unseen scenes, with real-time performance on an F1TENTH vehicle.
Optimization-based collision avoidance,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
baseline 1
citation-polarity summary
fields
cs.RO 1years
2025 1verdicts
CONDITIONAL 1roles
baseline 1polarities
baseline 1representative citing papers
citing papers explorer
-
Rapid and Safe Trajectory Planning over Diverse Scenes through Diffusion Composition
Diffusion models trained separately on static and dynamic scenes can be composed at test time to plan collision-free, kinematically feasible trajectories in unseen scenes, with real-time performance on an F1TENTH vehicle.