Skip to main navigation Skip to search Skip to main content

Broad-Learning-Based Adaptive Secure Control for AAV Swarm Under Deception Attacks

  • Jiacheng Li
  • , Chenjun Liu
  • , Jason J.R. Liu
  • , Dengxiu Yu
  • , Yan Jun Liu
  • , C. L. Philip Chen
  • University of Macau
  • Liaoning University of Technology
  • South China University of Technology
  • Guangdong Artificial Intelligence and Digital Economy Laboratory - Guangzhou

Research output: Contribution to journal › Article › peer-review

Abstract

In this paper, the swarm adaptive secure control (SASC) problem for multiple autonomous aerial vehicles (AAVs) is investigated, with a focus on identifying nonlinear elements within the system and addressing challenges arising from malicious attacks and unknown interference. In order to enhance the practical applicability of engineering while maintaining its general nature, the focus is on a disturbed nonlinear quadcopter swarm under cyber-attack. Subsequently, a dynamically adjusted SASC law is proposed to counter both additive and multiplicative deception attacks, while a broad learning approximator is introduced to handle nonlinear and unknown interference for SASC. Notably, an item for swarm aggregation–dispersion has been added to the SASC, allowing the swarm to maintain a fixed configuration or operate in a time-varying configuration. Following this, simulation validation is conducted to demonstrate the correctness and effectiveness of the proposed algorithm, highlighting the superiority through a comparison to common neural network approximators. Overall, the stability, security, and robustness of AAV swarms can be enhanced through our research, offering theoretical support for advanced control in autonomous systems.

Keywords

  • AAV swarm
  • Adaptive secure control
  • approximator
  • broad-learning
  • deception attacks

Fingerprint

Dive into the research topics of 'Broad-Learning-Based Adaptive Secure Control for AAV Swarm Under Deception Attacks'. Together they form a unique fingerprint.

Cite this