KEPT: Conformal Keepouts Kept in Planning for Enclosed-Space Navigation in Degraded Sensing
KEPT: Conformal Keepouts Kept in Planning for Enclosed-Space Navigation in Degraded Sensing
This paper introduces KEPT, a system that uses conformal prediction to calibrate safety margins around mapped walls in enclosed spaces like tunnels and mines where smoke, dust, or fog degrade sensing.
The authors propose a planning and control method that conditions conformal prediction margins on label-free severity estimates to maintain coverage targets in degraded sensing environments like smoke and fog, validated on ground-robot recordings and simulated quadrotor flights.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- Control
- Reported autonomousOpen source S2
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
arXiv:2610.37217v2 Announce Type: replace Abstract: Autonomous vehicles in enclosed spaces such as tunnels and mines rely on range maps that smoke, dust and fog degrade. Conformal prediction calibrates a keepout around the mapped walls that covers the true unsafe set at a chosen probability. In narrow passages this margin and the planner's buffer close the route, and existing remedies recover reach by relaxing the co
Open source S2
on ground-robot smoke recordings and the closed loop on simulated quadrotor flights in a foggy tunnel. The proposed system substantially reduces entries into the calibrated keepout and reaches numerically more goals than the deployment pattern in both calibration regimes. The per-class keepout holds its coverage near the target on held-out recordings.
Open source S4
Source：arXiv Robotics — research abstracts · arxiv.org