Event
Proposing Shape-Aware Neural Time Fields for Rigid Body Motion Planning
1 reports · 1 sources
Yanming Zhou et al. propose Shape-Aware Neural Time Fields, a physics-informed neural motion planning framework that learns reusable travel-time representations for rigid body motion, explicitly distinguishes the different geometric constraints of translation and rotation, and works with sampling-based planners for rigid body motion and non-prehensile pushing manipulation planning. The paper states that in closed-loop pushing experiments under execution deviation and external disturbances, fast replanning makes task execution reactive and adaptive.
Report timeline
- Shape-Aware Neural Time Fields: Physics-Informed Neural Time Fields for Rigid-Body Motion and Manipulation Planning
We propose Shape-Aware Neural Time Fields, a physics-informed neural motion planning framework that learns a reusable travel-time (cost-to-go) representation for rigid-body motion, explicitly distinguishing the different geometric constraints on translation and rotation, and pairs it with a sampling-based planner.
arXiv cs.RO ·