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

Tensegrity robots, composed of rigid rods and flexible cables, exhibit highstrength-to-weight ratios and significant deformations, which enable them tonavigate unstructured terrains and survive harsh impacts. They are hard tocontrol, however, due to high dimensionality, complex dynamics, and a coupledarchitecture. Physics-based simulation is a promising avenue for developinglocomotion policies that can be transferred to real robots. Nevertheless,modeling tensegrity robots is a complex task due to a substantial sim2real gap.To address this issue, this paper describes a Real2Sim2Real (R2S2R) strategyfor tensegrity robots. This strategy is based on a differentiable physicsengine that can be trained given limited data from a real robot. These datainclude offline measurements of physical properties, such as mass and geometryfor various robot components, and the observation of a trajectory using arandom control policy. With the data from the real robot, the engine can beiteratively refined and used to discover locomotion policies that are directlytransferable to the real robot. Beyond the R2S2R pipeline, key contributions ofthis work include computing non-zero gradients at contact points, a lossfunction for matching tensegrity locomotion gaits, and a trajectorysegmentation technique that avoids conflicts in gradient evaluation duringtraining. Multiple iterations of the R2S2R process are demonstrated andevaluated on a real 3-bar tensegrity robot.

Real2Sim2Real Transfer for Control of Cable-driven Robots via a Differentiable Physics Engine