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

The dangers of adversarial attacks on Uncrewed Aerial Vehicle (UAV) agentsoperating in public are increasing. Adopting AI-based techniques and, morespecifically, Deep Learning (DL) approaches to control and guide these UAVs canbe beneficial in terms of performance but can add concerns regarding the safetyof those techniques and their vulnerability against adversarial attacks.Confusion in the agent's decision-making process caused by these attacks canseriously affect the safety of the UAV. This paper proposes an innovativeapproach based on the explainability of DL methods to build an efficientdetector that will protect these DL schemes and the UAVs adopting them fromattacks. The agent adopts a Deep Reinforcement Learning (DRL) scheme forguidance and planning. The agent is trained with a Deep Deterministic PolicyGradient (DDPG) with Prioritised Experience Replay (PER) DRL scheme thatutilises Artificial Potential Field (APF) to improve training times andobstacle avoidance performance. A simulated environment for UAV explainableDRL-based planning and guidance, including obstacles and adversarial attacks,is built. The adversarial attacks are generated by the Basic Iterative Method(BIM) algorithm and reduced obstacle course completion rates from 97\% to 35\%.Two adversarial attack detectors are proposed to counter this reduction. Thefirst one is a Convolutional Neural Network Adversarial Detector (CNN-AD),which achieves accuracy in the detection of 80\%. The second detector utilisesa Long Short Term Memory (LSTM) network. It achieves an accuracy of 91\% withfaster computing times compared to the CNN-AD, allowing for real-timeadversarial detection.

Robust Adversarial Attacks Detection based on Explainable Deep Reinforcement Learning For UAV Guidance and Planning