robotics
The Collision Model Learned Where False Safety Hurts
A differentiable distance field puts extra training weight on sign errors near the boundary between clearance and contact.
Summary
A differentiable distance field puts extra training weight on sign errors near the boundary between clearance and contact.
PI-UDF combines analytical forward kinematics, learned link-geometry embeddings and a shared residual network to predict distances between articulated robot bodies. Its training procedure mines samples near the collision boundary and penalizes false-safe sign errors asymmetrically. The resulting field becomes a differentiable clearance term inside nonlinear model-predictive control. Tests included a real dual-Franka platform performing close-proximity 14-degree-of-freedom arm swaps and dynamic evasion, plus offline replay against established geometry tools. The paper presents evidence that the estimator is useful for closed-loop planning; it does not replace certified collision checking or prove safety outside the tested configurations.
Why it matters
A differentiable distance field puts extra training weight on sign errors near the boundary between clearance and contact.
Limits and context
- The paper presents evidence that the estimator is useful for closed-loop planning; it does not replace certified collision checking or prove safety outside the tested configurations.
Key claims
A differentiable distance field puts extra training weight on sign errors near the boundary between clearance and contact.
Qualification: The paper presents evidence that the estimator is useful for closed-loop planning; it does not replace certified collision checking or prove safety outside the tested configurations.
Evidence: source-2026-10-11-006
Sources
- arXiv preprint 2610.12404, original v1arXiv / submitting research authors · primary research
Corrections
No corrections have been recorded for this story.