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

Existing robotic systems have a clear tension between generality andprecision. Deployed solutions for robotic manipulation tend to fall into theparadigm of one robot solving a single task, lacking precise generalization,i.e., the ability to solve many tasks without compromising on precision. Thispaper explores solutions for precise and general pick-and-place. In precisepick-and-place, i.e. kitting, the robot transforms an unstructured arrangementof objects into an organized arrangement, which can facilitate furthermanipulation. We propose simPLE (simulation to Pick Localize and PLacE) as asolution to precise pick-and-place. simPLE learns to pick, regrasp and placeobjects precisely, given only the object CAD model and no prior experience. Wedevelop three main components: task-aware grasping, visuotactile perception,and regrasp planning. Task-aware grasping computes affordances of grasps thatare stable, observable, and favorable to placing. The visuotactile perceptionmodel relies on matching real observations against a set of simulated onesthrough supervised learning. Finally, we compute the desired robot motion bysolving a shortest path problem on a graph of hand-to-hand regrasps. On adual-arm robot equipped with visuotactile sensing, we demonstratepick-and-place of 15 diverse objects with simPLE. The objects span a wide rangeof shapes and simPLE achieves successful placements into structuredarrangements with 1mm clearance over 90% of the time for 6 objects, and over80% of the time for 11 objects. Videos are available athttp://mcube.mit.edu/research/simPLE.html .

simPLE: a visuotactile method learned in simulation to precisely pick, localize, regrasp, and place objects