Research paper
Barrier-certified learning for a brushbot
Adaptive model learning and barrier certificates constrain policies under changing dynamics. The abstract separates quadrotor simulation from a physical brushbot test and discusses conditional recovery after safety violations.
Reviewed 2026-11-05 · Source and bibliography checked
- Original title
- Barrier-Certified Adaptive Reinforcement Learning With Applications to Brushbot Navigation
- Publisher
- Institute of Electrical and Electronics Engineers (IEEE)
- Authors
- Motoya Ohnishi, Li Wang, Gennaro Notomista, Magnus Egerstedt
- DOI
- 10.1109/tro.2019.2920206
- Journal issue date
- 2019-10
- Source language
- en
- Source country
- US
- Edition
- IEEE TRO35(5),2019; related arXiv1801.09627v3
- Access
- Publisher metadata and open author manuscript
- Publication status
- Published
Source overview
Adaptive model learning and barrier certificates constrain policies under changing dynamics. The abstract separates quadrotor simulation from a physical brushbot test and discusses conditional recovery after safety violations.
Use and review
Empirical task/constraint results are not a no-injury guarantee. Data coverage, recovery reliability, sensing and execution limits need application-specific validation.
Sources & editions
Barrier-Certified Adaptive Reinforcement Learning With Applications to Brushbot Navigation ↗
Institute of Electrical and Electronics Engineers (IEEE) · IEEE TRO35(5),2019; related arXiv1801.09627v3
Archive abstract/authors/history reviewed; v3dated2019-08-06 explicitly links publisherDOI. Recovery after violation must not be read as prevention of every violation.