Adaptive Optimal Consensus Control for Nonlinear Uncertain Multiagent Systems Under DoS Attacks.

Area of Science:

• Control Systems Engineering
• Artificial Intelligence
• Network Security

Background:

• Nonlinear multiagent systems (MASs) face challenges with leader uncertainty and intermittent Denial-of-Service (DoS) attacks.
• Estimating leader dynamics is difficult due to unknown nonlinearities and disrupted communication.

Purpose of the Study:

• To develop a robust control strategy for leader-follower consensus in uncertain MASs under DoS attacks.
• To overcome the limitations of traditional identifier networks in complex MAS environments.

Main Methods:

• A learning-based adaptive distributed observer was designed to estimate the leader's unknown dynamics.
• A single-critic optimal consensus tracking control scheme was proposed, eliminating the need for an identifier network.

Main Results:

• The proposed observer effectively estimates leader dynamics despite intermittent DoS attacks and system uncertainties.
• The control scheme guarantees uniform ultimate boundedness (UUB) for all system signals.
• Successful consensus tracking was achieved in uncertain MASs.

Conclusions:

• The developed method provides a robust solution for optimal control in MASs under adversarial conditions.
• This approach enhances the reliability and performance of distributed systems facing cyber-attacks and uncertainties.