The main disadvantage of gain scheduling is that the controller parameters are pre-computed offline. This leads to a fixed controller when placed into an actual system. Modeling errors can cause the actual performance of the controller to degrade as there is no adaptive capability. It is also effortful to create a reliable gain schedule for each operating point. You also want to make sure that your gain matrix doesn’t get too large. Due to the limited set of operating points to be scheduled, the extreme behavior in the corner cases must be accounted for.
[[JDAM Dynamic Inversion CLAW]] – used dynamic inversion to eliminate gain scheduling requirements.
Sources
- “Chapter 5.1.3 – Gain Scheduling | GlobalSpec.” Accessed: May 21, 2023. [Online]. Available: https://www.globalspec.com/reference/21382/160210/chapter-5-1-3-gain-scheduling
- What Is Gain Scheduling? | Control Systems in Practice, (Sep. 18, 2018). Accessed: Nov. 12, 2022. [Online Video]. Available: https://www.youtube.com/watch?v=YiUjAV1bhKs
Backlinks
Gain Scheduling
[[Modeling Errors]]