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

Many previous works approach vision-based robotic grasping by training avalue network that evaluates grasp proposals. These approaches require anoptimization process at run-time to infer the best action from the valuenetwork. As a result, the inference time grows exponentially as the dimensionof action space increases. We propose an alternative method, by directlytraining a neural density model to approximate the conditional distribution ofsuccessful grasp poses from the input images. We construct a neural networkthat combines Gaussian mixture and normalizing flows, which is able torepresent multi-modal, complex probability distributions. We demonstrate onboth simulation and real robot that the proposed actor model achieves similarperformance compared to the value network using the Cross-Entropy Method (CEM)for inference, on top-down grasping with a 4 dimensional action space. Ouractor model reduces the inference time by 3 times compared to thestate-of-the-art CEM method. We believe that actor models will play animportant role when scaling up these approaches to higher dimensional actionspaces.

Learning Probabilistic Multi-Modal Actor Models for Vision-Based Robotic Grasping