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

The grand aim of having a single robot that can manipulate arbitrary objectsin diverse settings is at odds with the paucity of robotics datasets. Acquiringand growing such datasets is strenuous due to manual efforts, operationalcosts, and safety challenges. A path toward such an universal agent wouldrequire a structured framework capable of wide generalization but trainedwithin a reasonable data budget. In this paper, we develop an efficient system(RoboAgent) for training universal agents capable of multi-task manipulationskills using (a) semantic augmentations that can rapidly multiply existingdatasets and (b) action representations that can extract performant policieswith small yet diverse multi-modal datasets without overfitting. In addition,reliable task conditioning and an expressive policy architecture enable ouragent to exhibit a diverse repertoire of skills in novel situations specifiedusing language commands. Using merely 7500 demonstrations, we are able to traina single agent capable of 12 unique skills, and demonstrate its generalizationover 38 tasks spread across common daily activities in diverse kitchen scenes.On average, RoboAgent outperforms prior methods by over 40% in unseensituations while being more sample efficient and being amenable to capabilityimprovements and extensions through fine-tuning. Videos athttps://robopen.github.io/

RoboAgent: Generalization and Efficiency in Robot Manipulation via Semantic Augmentations and Action Chunking