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

Unmanned aerial vehicles (UAVs) have been widely used in military warfare. Inthis paper, we formulate the autonomous motion control (AMC) problem as aMarkov decision process (MDP) and propose an advanced deep reinforcementlearning (DRL) method that allows UAVs to execute complex tasks in large-scaledynamic three-dimensional (3D) environments. To overcome the limitations of theprioritized experience replay (PER) algorithm and improve performance, theproposed asynchronous curriculum experience replay (ACER) uses multithreads toasynchronously update the priorities, assigns the true priorities and applies atemporary experience pool to make available experiences of higher quality forlearning. A first-in-useless-out (FIUO) experience pool is also introduced toensure the higher use value of the stored experiences. In addition, combinedwith curriculum learning (CL), a more reasonable training paradigm of samplingexperiences from simple to difficult is designed for training UAVs. By trainingin a complex unknown environment constructed based on the parameters of a realUAV, the proposed ACER improves the convergence speed by 24.66\% and theconvergence result by 5.59\% compared to the state-of-the-art twin delayed deepdeterministic policy gradient (TD3) algorithm. The testing experiments carriedout in environments with different complexities demonstrate the strongrobustness and generalization ability of the ACER agent.

Asynchronous Curriculum Experience Replay: A Deep Reinforcement Learning Approach for UAV Autonomous Motion Control in Unknown Dynamic Environments