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

Open containers, i.e., containers without covers, are an important andubiquitous class of objects in human life. In this letter, we propose a novelmethod for robots to "imagine" the open containability affordance of apreviously unseen object via physical simulations. We implement our imaginationmethod on a UR5 manipulator. The robot autonomously scans the object with anRGB-D camera. The scanned 3D model is used for open containability imaginationwhich quantifies the open containability affordance by physically simulatingdropping particles onto the object and counting how many particles are retainedin it. This quantification is used for open-container vs. non-open-containerbinary classification (hereafter referred to as open container classification).If the object is classified as an open container, the robot further imaginespouring into the object, again using physical simulations, to obtain thepouring position and orientation for real robot autonomous pouring. We evaluateour method on open container classification and autonomous pouring of granularmaterial on a dataset containing 130 previously unseen objects with 57 objectcategories. Although our proposed method uses only 11 objects for simulationcalibration (training), its open container classification aligns well withhuman judgements. In addition, our method endows the robot with the capabilityto autonomously pour into the 55 containers in the dataset with a very highsuccess rate. We also compare to a deep learning method. Results show that ourmethod achieves the same performance as the deep learning method on opencontainer classification and outperforms it on autonomous pouring. Moreover,our method is fully explainable.

Can I Pour into It? Robot Imagining Open Containability Affordance of Previously Unseen Objects via Physical Simulations