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

Deformable object manipulation stands as one of the most captivating yetformidable challenges in robotics. While previous techniques have predominantlyrelied on learning latent dynamics through demonstrations, typicallyrepresented as either particles or images, there exists a pertinent limitation:acquiring suitable demonstrations, especially for long-horizon tasks, can beelusive. Moreover, basing learning entirely on demonstrations can hamper themodel's ability to generalize beyond the demonstrated tasks. In this work, weintroduce a demonstration-free hierarchical planning approach capable oftackling intricate long-horizon tasks without necessitating any training. Weemploy large language models (LLMs) to articulate a high-level, stage-by-stageplan corresponding to a specified task. For every individual stage, the LLMprovides both the tool's name and the Python code to craft intermediate subgoalpoint clouds. With the tool and subgoal for a particular stage at our disposal,we present a granular closed-loop model predictive control strategy. Thisleverages Differentiable Physics with Point-to-Point correspondence(DiffPhysics-P2P) loss in the earth mover distance (EMD) space, appliediteratively. Experimental findings affirm that our technique surpasses multiplebenchmarks in dough manipulation, spanning both short and long horizons.Remarkably, our model demonstrates robust generalization capabilities to noveland previously unencountered complex tasks without any preliminarydemonstrations. We further substantiate our approach with experimental trialson real-world robotic platforms.

Make a Donut: Language-Guided Hierarchical EMD-Space Planning for Zero-shot Deformable Object Manipulation