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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Controlled gliding is one of the most energetically efficient modes oftransportation for natural and human powered fliers. Here we demonstrate thatgliding and landing strategies with different optimality criteria can beidentified through deep reinforcement learning without explicit knowledge ofthe underlying physics. We combine a two dimensional model of a controlledelliptical body with deep reinforcement learning (D-RL) to achieve gliding witheither minimum energy expenditure, or fastest time of arrival, at apredetermined location. In both cases the gliding trajectories are smooth,although energy/time optimal strategies are distinguished by small/highfrequency actuations. We examine the effects of the ellipse's shape and weighton the optimal policies for controlled gliding. Surprisingly, we find that themodel-free reinforcement learning leads to more robust gliding than model-basedoptimal control strategies with a modest additional computational cost. We alsodemonstrate that the gliders with D-RL can generalize their strategies to reachthe target location from previously unseen starting positions. The model-freecharacter and robustness of D-RL suggests a promising framework for developingmechanical devices capable of exploiting complex flow environments.

Deep-Reinforcement-Learning for Gliding and Perching Bodies