Tube-Based Output Feedback Robust MPC for LPV Systems With Scaled Terminal Constraint Sets.

Area of Science:

• Control Systems Engineering
• Robotics and Automation
• Systems Theory

Background:

• Linear Parameter Varying (LPV) systems are common in control applications but challenging due to parameter uncertainty.
• Robust Model Predictive Control (RMPC) offers a framework for handling uncertainties, but output feedback and computational complexity remain issues.

Purpose of the Study:

• To develop a tube-based output feedback RMPC for discrete-time LPV systems with bounded disturbances.
• To reduce the online computational burden of RMPC for LPV systems.

Main Methods:

• Synthesizing an offline optimization problem to create a look-up table of robust positively invariant (RPI) and robust control invariant (RCI) sets.
• Implementing an online tube-based RMPC using the look-up table, tightened constraints, and scaled terminal sets for LPV systems.
• Employing one-step nominal system prediction to handle uncertain scheduling parameters.

Main Results:

• The proposed method effectively designs a look-up table and an online RMPC with tightened constraints and scaled terminal sets.
• The online optimization problem is simplified, leading to a lower computational burden.
• Recursive feasibility and robust stability of the controlled LPV system are guaranteed.

Conclusions:

• The developed tube-based output feedback RMPC provides a computationally efficient and robust solution for discrete-time LPV systems.
• The approach ensures stability and feasibility even with bounded disturbances and parameter variations.
• Numerical verification confirms the efficacy of the proposed control strategy.