Event-Triggered Resilient Filtering With the Interval Type Uncertainty for Markov Jump Systems

Area of Science:

• Control Systems Engineering
• Stochastic Systems Analysis
• Robotics

Background:

• Investigates event-triggered resilient filtering for Markov jump systems with asynchronous constraints.
• Employs a hidden Markov model to represent filter-system asynchronicity.
• Addresses gain uncertainties using interval type, offering higher accuracy than norm-bounded types.

Purpose of the Study:

• To develop an event-triggered resilient filter for Markov jump systems.
• To reduce network resource consumption and computational complexity.
• To ensure stochastic stability and extended dissipation performance of the filtering error system.

Main Methods:

• Utilizes a hidden Markov model for asynchronous constraints.
• Models filter gain uncertainties as interval types and separates vertices to simplify linear matrix inequalities.
• Applies an event-triggered scheme with a diagonal matrix threshold parameter.
• Employs convex optimization to derive sufficient conditions.

Main Results:

• Significantly reduces the number of linear matrix inequality constraints, decreasing calculation difficulty and time.
• Successfully balances network bandwidth reduction with improved system performance.
• Guarantees stochastic stability and extended dissipation performance for the filtering error systems.

Conclusions:

• The proposed event-triggered resilient filtering method is effective for Markov jump systems.
• The approach offers advantages in terms of accuracy, computational efficiency, and resource management.
• Validated through a single-link robot arm system simulation.