Onboard evolution of understandable swarm behaviors
- Submitting institution
-
University of Bristol
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 200013068
- Type
- D - Journal article
- DOI
-
10.1002/aisy.201900031
- Title of journal
- Advanced Intelligent Systems
- Article number
- 1900031
- First page
- -
- Volume
- 1
- Issue
- 6
- ISSN
- 2640-4567
- Open access status
- Compliant
- Month of publication
- July
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
3
- Research group(s)
-
A - Artificial Intelligence and Autonomy
- Citation count
- -
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- First paper demonstrating onboard artificial evolution of human understandable robot swarm controllers. This enables swarms to discover their behaviour in the wild. In 15 minutes, the robots go from behaving randomly, to completing a swarm task. Evolved behaviour trees are human readable, an important requirement for their deployment. This work was invited for presentation at top robotics conference (IROS) after publication in Advanced Intelligent Systems. Our swarm is the most computationally powerful available with 2 teraflops of computation. This led to two EPSRC IAA Commercialisation Awards (worth £200K with £100k from ToshibaTREL), one ICASE, and TAS Node award (EP/V026518/1, £3M).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -