A new method for control of networked systems with an experimental verification

Area of Science:

• Control Systems Engineering
• Networked Systems
• Robust Control

Background:

• Networked systems are susceptible to performance degradation due to transfer delays, packet loss, and disordering.
• Existing control methodologies often struggle to maintain stability and performance under such challenging network conditions.
• Robust control strategies are essential for reliable operation of distributed systems.

Purpose of the Study:

• To develop a variable selective control methodology for networked systems.
• To address challenges posed by transfer delay, packet loss, and packet disordering.
• To ensure closed-loop asymptotic stability in the presence of network uncertainties.

Main Methods:

• Extension of variable sampling period techniques with a packet-based control methodology.
• Development of step-invariant discrete-time plant models for a range of expected time delays.
• Design of a switching observer-based controller using linear matrix inequalities (LMIs).

Main Results:

• A sufficient condition for closed-loop asymptotic stability was derived.
• The proposed method effectively handles transfer delay, packet loss, and packet disordering.
• Simulation and experimental results validated the controller's effectiveness.

Conclusions:

• The variable selective control methodology offers a robust solution for networked systems.
• The approach ensures system stability despite network-induced uncertainties.
• The method is proven effective through both simulations and real-world experiments.