Sign in Subscribe
Sensors

Data-Driven Secure Control for Autonomous Vehicles

Strategic Insights

— 1 min read

Research conducted by Qufu Normal University has successfully developed a data-driven model identification and adaptive event-triggered secure control scheme for autonomous vehicles subject to sensor attacks. The proposed method exploits the dynamic mode decomposition (DMD) approach to identify the lateral dynamical model of autonomous vehicles, while an adaptive event-triggered scheme balances communication efficiency and control performance. The stability analysis and stabilization design are derived using Lyapunov theory and linear matrix inequalities technique.

Key Takeaways:

  • The research proposes a data-driven modeling approach using DMD to overcome modeling difficulties in autonomous vehicles.
  • The adaptive event-triggered scheme regulates the communication threshold based on feedback measurement to balance efficiency and performance.
  • The proposed control scheme actively mitigates sensor attacks using a sliding-mode-like control scheme.
  • The research concludes that the proposed scheme is effective in securing autonomous vehicles from sensor attacks, as demonstrated through comparison examples.
  • The proposed method has three key advantages: DMD overcomes modeling difficulties, event-triggered threshold is adaptively regulated, and sensor attacks can be actively mitigated.
  • The research was conducted by a team of authors from Qufu Normal University, led by Hong-Tao Sun.

Statistics:

  • The proposed data-driven modeling approach requires no a priori knowledge of the system dynamics.
  • The adaptive event-triggered scheme can regulate the communication threshold to achieve a balance between efficiency and performance.
  • The sliding-mode-like control scheme used in the research has been shown to be effective in countering sensor attacks.
  • The research demonstrates the effectiveness of the proposed control scheme through comparison examples.
  • The proposed method has been peer-reviewed and published in the journal ISA Transactions.

Sources:

  • "Data-driven modeling and adaptive event-triggered secure control for autonomous vehicles subject to sensor attacks." ISA Transactions, 2025.
  • Hong-Tao Sun, College of Engineering, Qufu Normal University
  • Xinyu Xie, Miao Rong, Zongying Feng, and Chen Peng, Qufu Normal University
  • ISA Transactions, Elsevier Science Inc, Ste 800, 230 Park Ave, New York, NY 10169, USA
  • Elsevier, www.elsevier.com
  • ISA Transactions, www.journals.elsevier.com/isa-transactions/

Read more