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

Instructing a robot to complete an everyday task within our homes has been along-standing challenge for robotics. While recent progress inlanguage-conditioned imitation learning and offline reinforcement learning hasdemonstrated impressive performance across a wide range of tasks, they aretypically limited to short-horizon tasks -- not reflective of those a homerobot would be expected to complete. While existing architectures have thepotential to learn these desired behaviours, the lack of the necessarylong-horizon, multi-step datasets for real robotic systems poses a significantchallenge. To this end, we present the Long-Horizon Manipulation (LHManip)dataset comprising 200 episodes, demonstrating 20 different manipulation tasksvia real robot teleoperation. The tasks entail multiple sub-tasks, includinggrasping, pushing, stacking and throwing objects in highly clutteredenvironments. Each task is paired with a natural language instruction andmulti-camera viewpoints for point-cloud or NeRF reconstruction. In total, thedataset comprises 176,278 observation-action pairs which form part of the OpenX-Embodiment dataset. The full LHManip dataset is made publicly available athttps://github.com/fedeceola/LHManip.

LHManip: A Dataset for Long-Horizon Language-Grounded Manipulation Tasks in Cluttered Tabletop Environments