Risk-Constrained Belief-Space Optimization for Safe Control under Latent Uncertainty
Clinton Enwerem, John S. Baras, Calin Belta · 2026
Read Risk-Constrained Belief-Space Optimization for Safe Control under Latent Uncertainty by Clinton Enwerem, John S. Baras, Calin Belta (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-03868.txt.
Description
Clinton Enwerem and colleagues built a risk-aware controller for high-stakes systems whose key traits stay hidden from sensors. Their belief-space Model Predictive Path Integral controller plans under a belief about those traits, weighs risk with Conditional Value-at-Risk, and caps risk on the planned path. Theory ties the cap to a safety promise that holds with high odds, and stretches it across repeat runs. In MuJoCo stowing tests, the method hit 82 percent success with zero contact faults, beating risk-blind and chance-capped rivals.
Catalog record
| CREATOR | Clinton Enwerem, John S. Baras, Calin Belta |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43315 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.03868