Data-driven security control for unknown nonlinear MASs with hybrid faults: A hierarchical control approach

Area of Science:

• Control Systems Engineering
• Artificial Intelligence
• Robotics

Background:

• Multi-agent systems (MASs) often face challenges with unknown dynamics and hybrid faults.
• Conventional model-free adaptive control (MFAC) methods can be limited by reliance on consensus errors.
• Scalability and robustness are critical for practical MAS applications.

Purpose of the Study:

• To develop a novel hierarchical data-driven consensus control strategy for unknown nonlinear MASs with hybrid faults.
• To enhance the robustness and scalability of control systems in the presence of system uncertainties and faults.
• To provide a fault-tolerant control solution that does not require prior system model knowledge.

Main Methods:

• A hierarchical framework is proposed, decoupling the control process for improved performance.
• A fully distributed observer estimates leader dynamics using only local input-output data.
• A distributed MFAC controller incorporates online actuator fault estimation for real-time fault handling.

Main Results:

• The hierarchical design allows independent agent operation and reduces inter-agent interference.
• Theoretical analysis confirms uniform boundedness of estimation and tracking errors under hybrid faults.
• Simulations on MAS examples and multi-manipulator platforms validate the method's effectiveness.

Conclusions:

• The proposed data-driven hierarchical control strategy offers an effective solution for unknown nonlinear MASs with hybrid faults.
• The method demonstrates practical applicability and enhanced robustness compared to conventional approaches.
• This research contributes to advancing fault-tolerant control in complex multi-agent systems.