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

We present a learning-based dynamics model for granular materialmanipulation. Inspired by the Eulerian approach commonly used in fluiddynamics, our method adopts a fully convolutional neural network that operateson a density field-based representation of object piles and pushers, allowingit to exploit the spatial locality of inter-object interactions as well as thetranslation equivariance through convolution operations. Furthermore, ourdifferentiable action rendering module makes the model fully differentiable andcan be directly integrated with a gradient-based trajectory optimizationalgorithm. We evaluate our model with a wide array of piles manipulation tasksboth in simulation and real-world experiments and demonstrate that itsignificantly exceeds existing latent or particle-based methods in bothaccuracy and computation efficiency, and exhibits zero-shot generalizationcapabilities across various environments and tasks.

Neural Field Dynamics Model for Granular Object Piles Manipulation