Incorporating World Models in Decision Making for Next Best View Selection in Active Sensing
Incorporating World Models in Decision Making for Next Best View Selection in Active Sensing
This research presents a deployment adaptation methodology for next-best-view selection in robotic 3D reconstruction. By learning a sensor-return adapter from limited local calibration data while keeping a pretrained predictor fixed, the method estimates measurement dropout over predicted surfaces. In an experiment with a structured-light sensor, the adapted approach achieved higher measured surface coverage using 60% fewer captures compared to nominal selection and successfully recovered surface regions that the nominal method failed to measure.
The study demonstrates a method for adapting pretrained world models to real-world sensor constraints, achieving higher surface coverage with fewer captures in a structured-light sensor experiment.
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.
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
that learns a sensor-return adapter from limited local calibration data while keeping the pretrained predictor fixed. The adapter estimates measurement dropout over each predicted surface, allowing the selection method to favor views likely to fill gaps in the measured reconstruction. In a real-robot experiment with a structured-light sensor, the adapted method achieves higher measured surface coverage with 60% fewe
Open source S3
r captures than nominal selection. The adapted method recovers a surface region that the nominal method repeatedly targets but fails to measure. These results suggest that robots can reuse pretrained predictive models while adapting their decisions to local operating conditions in data-scarce environments.
Open source S4
arXiv:2611.13229v1 Announce Type: new Abstract: Efficient robotic 3D reconstruction requires selecting views that contribute new surface measurements with few physical captures. Pretrained world models can guide this choice, but a predicted surface may not produce valid depth measurements under the operating conditions of the deployed sensor. We present a deployment adaptation methodology for next-best-view selection
Open source S2
Incorporating World Models in Decision Making for Next Best View Selection in Active Sensing
Open source S1
Source：arXiv Robotics — research abstracts · arxiv.org