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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In industrial part kitting, 3D objects are inserted into cavities fortransportation or subsequent assembly. Kitting is a critical step as it candecrease downstream processing and handling times and enable lower storage andshipping costs. We present Kit-Net, a framework for kitting previously unseen3D objects into cavities given depth images of both the target cavity and anobject held by a gripper in an unknown initial orientation. Kit-Net usesself-supervised deep learning and data augmentation to train a convolutionalneural network (CNN) to robustly estimate 3D rotations between objects andmatching concave or convex cavities using a large training dataset of simulateddepth images pairs. Kit-Net then uses the trained CNN to implement a controllerto orient and position novel objects for insertion into novel prismatic andconformal 3D cavities. Experiments in simulation suggest that Kit-Net canorient objects to have a 98.9% average intersection volume between the objectmesh and that of the target cavity. Physical experiments with industrialobjects succeed in 18% of trials using a baseline method and in 63% of trialswith Kit-Net. Video, code, and data are available athttps://github.com/BerkeleyAutomation/Kit-Net.

Kit-Net: Self-Supervised Learning to Kit Novel 3D Objects into Novel 3D Cavities