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

Control of wheeled humanoid locomotion is a challenging problem due to thenonlinear dynamics and under-actuated characteristics of these robots.Traditionally, feedback controllers have been utilized for stabilization andlocomotion. However, these methods are often limited by the fidelity of theunderlying model used, choice of controller, and environmental variablesconsidered (surface type, ground inclination, etc). Recent advances inreinforcement learning (RL) offer promising methods to tackle some of theseconventional feedback controller issues, but require large amounts ofinteraction data to learn. Here, we propose a hybrid learning and model-basedcontroller Hybrid LMC that combines the strengths of a classical linearquadratic regulator (LQR) and ensemble deep reinforcement learning. Ensembledeep reinforcement learning is composed of multiple Soft Actor-Critic (SAC) andis utilized in reducing the variance of RL networks. By using a feedbackcontroller in tandem the network exhibits stable performance in the earlystages of training. As a preliminary step, we explore the viability of HybridLMC in controlling wheeled locomotion of a humanoid robot over a set ofdifferent physical parameters in MuJoCo simulator. Our results show that HybridLMC achieves better performance compared to other existing techniques and hasincreased sample efficiency

Hybrid LMC: Hybrid Learning and Model-based Control for Wheeled Humanoid Robot via Ensemble Deep Reinforcement Learning