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

In this paper, we propose LAN-grasp, a novel approach towards moreappropriate semantic grasping. We use foundation models to provide the robotwith a deeper understanding of the objects, the right place to grasp an object,or even the parts to avoid. This allows our robot to grasp and utilize objectsin a more meaningful and safe manner. We leverage the combination of a LargeLanguage Model, a Vision Language Model, and a traditional grasp planner togenerate grasps demonstrating a deeper semantic understanding of the objects.We first prompt the Large Language Model about which object part is appropriatefor grasping. Next, the Vision Language Model identifies the corresponding partin the object image. Finally, we generate grasp proposals in the regionproposed by the Vision Language Model. Building on foundation models providesus with a zero-shot grasp method that can handle a wide range of objectswithout the need for further training or fine-tuning. We evaluated our methodin real-world experiments on a custom object data set. We present the resultsof a survey that asks the participants to choose an object part appropriate forgrasping. The results show that the grasps generated by our method areconsistently ranked higher by the participants than those generated by aconventional grasping planner and a recent semantic grasping approach.

LAN-grasp: Using Large Language Models for Semantic Object Grasping