Decentralized Secure Tracking Control for Nonlinear Interconnected Systems: A Synergetic Learning-Based Strategy

Decentralized secure control faces significant challenges in handling unknown mismatched interconnections and reducing fault-tolerant delays. To address these issues, this paper proposes a synergetic learning-based decentralized secure tracking control scheme for nonlinear interconnected systems with multiple actuator faults. Replacing actual states with desired ones in the coupled system relaxes the assumption of requiring a known upper bound for interconnections, and a neural network observer is designed to estimate the replaced interconnections. To reduce fault-tolerant delays, the secure tracking control problem is reformulated as an adversarial evolution problem between fault signals and control inputs, eliminating the need for fault compensation. To achieve optimal tracking control, an augmented subsystem is constructed by integrating the dynamics of tracking error and the reference trajectory. A modified cost function is designed for the augmented subsystem, and a critic network with two cooperative updating laws is developed to solve the Hamilton–Jacobi–Isaacs equation, providing a synergetic approximate solution for the control input and fault assistance signal. It is proven that the tracking error converges to a small neighborhood of the equilibrium. Simulation results demonstrate the effectiveness of the proposed approach.