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

Many manipulation tasks require robots to interact with unknown environments.In such applications, the ability to adapt the impedance according to differenttask phases and environment constraints is crucial for safety and performance.Although many approaches based on deep reinforcement learning (RL) and learningfrom demonstration (LfD) have been proposed to obtain variable impedance skillson contact-rich manipulation tasks, these skills are typically task-specificand could be sensitive to changes in task settings. This paper proposes aninverse reinforcement learning (IRL) based approach to recover both thevariable impedance policy and reward function from expert demonstrations. Weexplore different action space of the reward functions to achieve a moregeneral representation of expert variable impedance skills. Experiments on twovariable impedance tasks (Peg-in-Hole and Cup-on-Plate) were conducted in bothsimulations and on a real FANUC LR Mate 200iD/7L industrial robot. Thecomparison results with behavior cloning and force-based IRL proved that thelearned reward function in the gain action space has better transferabilitythan in the force space. Experiment videos are available athttps://msc.berkeley.edu/research/impedance-irl.html.

Learning Variable Impedance Control via Inverse Reinforcement Learning for Force-Related Tasks