Decentralized adaptive neural safe tracking control for nonlinear systems with conflicted output constraints

Area of Science:

• Control Theory
• Nonlinear Systems
• Systems Engineering

Background:

• Existing control methods for interconnected large-scale nonlinear systems (ILSNSs) fail when output constraints conflict with reference signals.
• Barrier Lyapunov functions (BLF) and nonlinear transformation functions (NTF) are inadequate for handling such conflicted constraints.

Purpose of the Study:

• To develop a decentralized adaptive safe tracking control strategy for ILSNSs with conflicted output constraints.
• To address the practical scenario where system constraints conflict with pre-defined reference signals.

Main Methods:

• Design of a novel safety reference signal (SRS) using a boundary protection approach to ensure it stays within constraint ranges.
• Incorporation of a prescribed performance function to define convergence time and tracking accuracy.
• Development of a controller using backstepping technique and radial basis function neural networks (RBFNN).
• Integration of a dynamic event-triggering mechanism (DETM) to reduce communication load.

Main Results:

• The proposed controller ensures desired tracking performance while maintaining safety under conflicted output constraints.
• The safety reference signal (SRS) effectively resolves conflicts between output constraints and reference signals.
• The dynamic event-triggering mechanism (DETM) significantly reduces communication burden between the actuator and the plant.
• Simulation results validate the effectiveness of the developed control scheme.

Conclusions:

• The novel approach provides a robust solution for safe tracking control in ILSNSs with conflicted output constraints.
• The method enhances system safety and performance while optimizing communication efficiency.
• This work offers a practical framework for real-world applications involving complex nonlinear systems.