Robust optimal design of FOPID controller for five bar linkage robot in a Cyber-Physical System: A new...

Area of Science:

• Control Systems Engineering
• Robotics
• Optimization Theory

Background:

• Cyber-Physical Systems (CPS) require reliable control under real-world variability.
• Fractional-Order PID (FOPID) controllers offer advanced control but face challenges in uncertain environments.
• Existing methods struggle with optimizing FOPID controllers under multiple sources of variability.

Purpose of the Study:

• To enhance the reliability of optimal control results for FOPID controllers in a CPS context.
• To develop a robust multi-objective optimization approach for FOPID controllers under uncertainty.
• To minimize computational cost while ensuring robustness against environmental noise and sensor anomalies.

Main Methods:

• A hybrid surrogate-metaheuristic approach combining Particle Swarm Optimization (PSO) and Gaussian Process (GP) surrogate modeling.
• Sequential optimization for robust results using GP surrogate updates.
• Efficient global optimization with a jackknife leave-one-out estimator to reduce computational cost.

Main Results:

• The proposed method effectively optimizes two objective functions: signal energy control and response error control.
• Demonstrated robustness against multiple sources of variability, including setpoint changes and sensor feedback anomalies.
• Achieved reliable and robust FOPID control for a five-bar linkage robot manipulator with reduced computational effort.

Conclusions:

• The developed hybrid approach significantly improves the robustness and reliability of FOPID controllers in CPS.
• The method successfully tackles computational challenges in robust multi-objective optimization.
• This work provides an effective strategy for real-world CPS control applications.