English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Robot swarms often exhibit emergent behaviors that are fascinating toobserve; however, it is often difficult to predict what swarm behaviors canemerge under a given set of agent capabilities. We seek to efficiently leveragehuman input to automatically discover a taxonomy of collective behaviors thatcan emerge from a particular multi-agent system, without requiring the human toknow beforehand what behaviors are interesting or even possible. Our proposedapproach adapts to user preferences by learning a similarity space over swarmcollective behaviors using self-supervised learning and human-in-the-loopqueries. We combine our learned similarity metric with novelty search andclustering to explore and categorize the space of possible swarm behaviors. Wealso propose several general-purpose heuristics that improve the efficiency ofour novelty search by prioritizing robot controllers that are likely to lead tointeresting emergent behaviors. We test our approach in simulation on two robotcapability models and show that our methods consistently discover a richer setof emergent behaviors than prior work. Code, videos, and datasets are availableat https://sites.google.com/view/evolving-novel-swarms.

Leveraging Human Feedback to Evolve and Discover Novel Emergent Behaviors in Robot Swarms