Neural-Adaptive Finite-Time Consensus Tracking of Singular Multiagent Systems With Actuator Attack.

Area of Science:

• Control Theory
• Networked Systems
• Artificial Intelligence

Background:

• Actuator attacks pose significant threats to complex networked systems, compromising their stability and performance.
• Singular multiagent systems with time-varying delays present unique challenges in achieving coordinated control.
• Finite-time consensus tracking (FTCT) is crucial for ensuring rapid and stable agreement among agents.

Purpose of the Study:

• To investigate the finite-time consensus tracking (FTCT) control problem for singular multiagent systems under actuator attacks.
• To develop a robust control strategy that can handle time-varying delays and external disturbances.
• To ensure the stability and reliable performance of networked systems despite adversarial actions.

Main Methods:

• A state observer is proposed to approximate follower agent states and design sliding mode surfaces under actuator attack.
• A novel neural-adaptive distributed FTCT protocol and actuator attack estimation protocol are designed using radial basis function neural networks.
• Sliding mode control with a partitioning strategy is employed for analyzing FTCT in singular multiagent systems.

Main Results:

• Sufficient conditions and solving strategies for FTCT in singular multiagent systems over finite time intervals are provided.
• The proposed neural-adaptive protocol effectively estimates actuator attacks and ensures consensus tracking.
• Simulation examples validate the effectiveness of the proposed FTCT control strategy.

Conclusions:

• The developed FTCT control strategy is effective for singular multiagent systems susceptible to actuator attacks.
• The combination of neural networks and sliding mode control offers a robust solution for complex networked systems.
• The study provides a theoretical framework and practical validation for secure and efficient multiagent system control.