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

Reinforcement Learning (RL) has witnessed great strides for quadrupedlocomotion, with continued progress in the reliable sim-to-real transfer ofpolicies. However, it remains a challenge to reuse a policy on another robot,which could save time for retraining. In this work, we present a framework forzero-shot policy retargeting wherein diverse motor skills can be transferredbetween robots of different shapes and sizes. The new framework centers on aplanning-and-control pipeline that systematically integrates RL and ModelPredictive Control (MPC). The planning stage employs RL to generate adynamically plausible trajectory as well as the contact schedule, avoiding thecombinatorial complexity of contact sequence optimization. This information isthen used to seed the MPC to stabilize and robustify the policy roll-out via anew Hybrid Kinodynamic (HKD) model that implicitly optimizes the footholdlocations. Hardware results show an ability to transfer policies from both theA1 and Laikago robots to the MIT Mini Cheetah robot without requiring anypolicy re-tuning.

Zero-Shot Retargeting of Learned Quadruped Locomotion Policies Using Hybrid Kinodynamic Model Predictive Control