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

Autonomous navigation in unknown complex environment is still a hard problem,especially for small Unmanned Aerial Vehicles (UAVs) with limited computationresources. In this paper, a neural network-based reactive controller isproposed for a quadrotor to fly autonomously in unknown outdoor environment.The navigation controller makes use of only current sensor data to generate thecontrol signal without any optimization or configuration space searching, whichreduces both memory and computation requirement. The navigation problem ismodelled as a Markov Decision Process (MDP) and solved using deep reinforcementlearning (DRL) method. Specifically, to get better understanding of the trainednetwork, some model explanation methods are proposed. Based on the featureattribution, each decision making result during flight is explained using bothvisual and texture explanation. Moreover, some global analysis are alsoprovided for experts to evaluate and improve the trained neural network. Thesimulation results illustrated the proposed method can make useful andreasonable explanation for the trained model, which is beneficial for bothnon-expert users and controller designer. Finally, the real world tests shownthe proposed controller can navigate the quadrotor to goal positionsuccessfully and the reactive controller performs much faster than someconventional approach under the same computation resource.

Explainable Deep Reinforcement Learning for UAV Autonomous Navigation