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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Scene transfer for vision-based mobile robotics applications is a highlyrelevant and challenging problem. The utility of a robot greatly depends on itsability to perform a task in the real world, outside of a well-controlled labenvironment. Existing scene transfer end-to-end policy learning approachesoften suffer from poor sample efficiency or limited generalizationcapabilities, making them unsuitable for mobile robotics applications. Thiswork proposes an adaptive multi-pair contrastive learning strategy for visualrepresentation learning that enables zero-shot scene transfer and real-worlddeployment. Control policies relying on the embedding are able to operate inunseen environments without the need for finetuning in the deploymentenvironment. We demonstrate the performance of our approach on the task ofagile, vision-based quadrotor flight. Extensive simulation and real-worldexperiments demonstrate that our approach successfully generalizes beyond thetraining domain and outperforms all baselines.

Contrastive Learning for Enhancing Robust Scene Transfer in Vision-based Agile Flight