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

This paper presents multi-agent reinforcement learning frameworks for thelow-level control of a quadrotor UAV. While single-agent reinforcement learninghas been successfully applied to quadrotors, training a single monolithicnetwork is often data-intensive and time-consuming. To address this, wedecompose the quadrotor dynamics into the translational dynamics and the yawingdynamics, and assign a reinforcement learning agent to each part for efficienttraining and performance improvements. The proposed multi-agent framework forquadrotor low-level control that leverages the underlying structures of thequadrotor dynamics is a unique contribution. Further, we introduceregularization terms to mitigate steady-state errors and to avoid aggressivecontrol inputs. Through benchmark studies with sim-to-sim transfer, it isillustrated that the proposed multi-agent reinforcement learning substantiallyimproves the convergence rate of the training and the stability of thecontrolled dynamics.

Multi-Agent Reinforcement Learning for the Low-Level Control of a Quadrotor UAV