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

Insertion is a challenging haptic and visual control problem with significantpractical value for manufacturing. Existing approaches in the model-basedrobotics community can be highly effective when task geometry is known, but arecomplex and cumbersome to implement, and must be tailored to each individualproblem by a qualified engineer. Within the learning community there is a longhistory of insertion research, but existing approaches are typically either toosample-inefficient to run on real robots, or assume access to high-level objectfeatures, e.g. socket pose. In this paper we show that relatively minormodifications to an off-the-shelf Deep-RL algorithm (DDPG), combined with asmall number of human demonstrations, allows the robot to quickly learn tosolve these tasks efficiently and robustly. Our approach requires no modelingor simulation, no parameterized search or alignment behaviors, no vision systemaside from raw images, and no reward shaping. We evaluate our approach on anarrow-clearance peg-insertion task and a deformable clip-insertion task, bothof which include variability in the socket position. Our results show thatthese tasks can be solved reliably on the real robot in less than 10 minutes ofinteraction time, and that the resulting policies are robust to variance in thesocket position and orientation.

A Practical Approach to Insertion with Variable Socket Position Using Deep Reinforcement Learning