Arkadiy Turevskiy, MathWorks
Automatically tune controllers to maximize performance over a range of parameter values using the Control System Tuner app from Robust Control Toolbox™. In the example featured in this video, the plant is a mass-spring-damper system. The nominal values of mass, spring stiffness, and damping are known, but the actual parameter values can vary up to 40% from the nominal values. The goal is to design a PID controller that will be robust to the uncertainty of the in-plant model parameters.
The video shows how to use the Control System Tuner app to design a robust controller. The plant and the controller are modeled in Simulink®. Parameter uncertainty is defined using Robust Control Toolbox. The video shows how you can use the block substitution capability of Simulink Control Design™ to linearize a plant model, while taking parameter uncertainty into account. It illustrates how to create an uncertain state-space system that represents variations in plant dynamics due to the uncertainty of the in-plant parameter values. Next, the Control System Tuner app is used to tune PID controller gains to meet design requirements for all possible combinations of plant parameters. You then see how to compute the worst-case step response of the closed-loop system and compare it with the worst-case step response of the PID controller tuned for the nominal plant parameter values. The comparison shows that a robustly tuned PID controller has superior worst-case performance.