Research paper
Bayesian navigation · Learning collision likelihood beyond the map
A POMDP-inspired planner learns collision probabilities beyond mapped space. A prior encodes safety constraints to support cautious fallback in unfamiliar environments; the authors report high-speed navigation experiments in unknown indoor spaces.
Reviewed 2026-11-05 · Source and bibliography checked
- Original title
- Bayesian Learning for Safe High-Speed Navigation in Unknown Environments
- Publisher
- Springer International Publishing
- Authors
- Charles Richter, William Vega-Brown, Nicholas Roy
- DOI
- 10.1007/978-3-319-60916-4_19
- First online publication
- 2017-07-25
- Source language
- en
- Edition
- Springer Proceedings in Advanced Robotics; 2017-07-25; DOI 10.1007/978-3-319-60916-4_19
- Access
- Publisher abstract · restricted full text
- Publication status
- Published
Related resources and engineering notes
Source overview
A POMDP-inspired planner learns collision probabilities beyond mapped space. A prior encodes safety constraints to support cautious fallback in unfamiliar environments; the authors report high-speed navigation experiments in unknown indoor spaces.
Use and review
Original abstract and bibliography checked. Learned collision probabilities and prior assumptions require validation when the environment changes.
Sources & editions
Bayesian Learning for Safe High-Speed Navigation in Unknown Environments ↗
Springer International Publishing · Springer Proceedings in Advanced Robotics; 2017-07-25; DOI 10.1007/978-3-319-60916-4_19
Springer original chapter abstract, https://link.springer.com/10.1007/978-3-319-60916-4_19; online2017-07-25, book edition2018.