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

Billions of people use chopsticks, a simple yet versatile tool, for finemanipulation of everyday objects. The small, curved, and slippery tips ofchopsticks pose a challenge for picking up small objects, making them asuitably complex test case. This paper leverages human demonstrations todevelop an autonomous chopsticks-equipped robotic manipulator. Due to the lackof accurate models for fine manipulation, we explore model-free imitationlearning, which traditionally suffers from the covariate shift phenomenon thatcauses poor generalization. We propose two approaches to reduce covariateshift, neither of which requires access to an interactive expert or a model,unlike previous approaches. First, we alleviate single-step prediction errorsby applying an invariant operator to increase the data support at criticalsteps for grasping. Second, we generate synthetic corrective labels by addingbounded noise and combining parametric and non-parametric methods to preventerror accumulation. We demonstrate our methods on a real chopstick-equippedrobot that we built, and observe the agent's success rate increase from 37.3%to 80%, which is comparable to the human expert performance of 82.6%.

Grasping with Chopsticks: Combating Covariate Shift in Model-free Imitation Learning for Fine Manipulation