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/

Zendy Plus

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

Zendy Tools

Zendy Open

Imitation learning for robot dexterous manipulation, especially with a realrobot setup, typically requires a large number of demonstrations. In thispaper, we present a data-efficient learning from demonstration framework whichexploits the use of rich tactile sensing data and achieves fine bimanual pinchgrasping. Specifically, we employ a convolutional autoencoder network that caneffectively extract and encode high-dimensional tactile information. Further,We develop a framework that achieves efficient multi-sensor fusion forimitation learning, allowing the robot to learn contact-aware sensorimotorskills from demonstrations. Our comparision study against the framework withoutusing encoded tactile features highlighted the effectiveness of incorporatingrich contact information, which enabled dexterous bimanual grasping with activecontact searching. Extensive experiments demonstrated the robustness of thefine pinch grasp policy directly learned from few-shot demonstration, includinggrasping of the same object with different initial poses, generalizing to tenunseen new objects, robust and firm grasping against external pushes, as wellas contact-aware and reactive re-grasping in case of dropping objects undervery large perturbations. Furthermore, the saliency map analysis method is usedto describe weight distribution across various modalities during pinchgrasping, confirming the effectiveness of our framework at leveragingmultimodal information.

Learning Fine Pinch-Grasp Skills using Tactile Sensing from A Few Real-world Demonstrations