Fault-Tolerant Control Redesign for Noisy High-Order Fully Actuated Systems.

Area of Science:

• Control Systems Engineering
• Fault-Tolerant Control
• System Dynamics

Background:

• High-order fully actuated systems (HOFASs) are susceptible to actuator faults, sensor faults, and measurement noise.
• Existing fault-tolerant control (FTC) methods may have limitations in noise suppression and fault compensation accuracy.

Purpose of the Study:

• To develop and evaluate two novel FTC frameworks for HOFASs.
• To address actuator faults, sensor faults, and measurement noise simultaneously.
• To improve the performance and robustness of control systems.

Main Methods:

• Analysis of observable architectures for fault detection and isolation.
• Compensation of actuator faults using fusion observers.
• Rejection of sensor faults via redundant observability.
• Application of a dead-zone fusion observation strategy for noise suppression.

Main Results:

• The first FTC framework achieves ultimately uniformly bounded (UUB) error systems.
• The redesigned FTC framework with dead-zone fusion observation demonstrates superior noise suppression, particularly in linear HOFAS models.
• Experimental validation confirms improved trajectory tracking and noise suppression capabilities.

Conclusions:

• The proposed FTC frameworks offer effective solutions for fault tolerance in HOFASs.
• The novel FTC redesign significantly enhances measurement noise suppression.
• The developed methods provide robust control performance in the presence of various system uncertainties.