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

Operating under real world conditions is challenging due to the possibilityof a wide range of failures induced by execution errors and state uncertainty.In relatively benign settings, such failures can be overcome by retrying orexecuting one of a small number of hand-engineered recovery strategies. Bycontrast, contact-rich sequential manipulation tasks, like opening doors andassembling furniture, are not amenable to exhaustive hand-engineering. Toaddress this issue, we present a general approach for robustifying manipulationstrategies in a sample-efficient manner. Our approach incrementally improvesrobustness by first discovering the failure modes of the current strategy viaexploration in simulation and then learning additional recovery skills tohandle these failures. To ensure efficient learning, we propose an onlinealgorithm called Meta-Reasoning for Skill Learning (MetaReSkill) that monitorsthe progress of all recovery policies during training and allocates trainingresources to recoveries that are likely to improve the task performance themost. We use our approach to learn recovery skills for door-opening andevaluate them both in simulation and on a real robot with little fine-tuning.Compared to open-loop execution, our experiments show that even a limitedamount of recovery learning improves task success substantially from 71% to92.4% in simulation and from 75% to 90% on a real robot.

Efficient Recovery Learning using Model Predictive Meta-Reasoning