Model‐free online tuning of controller parameters using a globalized local search algorithm
Bibliographic record
Abstract
Summary In an earlier work, the authors proposed a globalized bounded Nelder‐Mead algorithm with deterministic restarts and a linearly growing memory vector. It was shown that the algorithm was a favorable option for solving multimodal optimization problems like controller tuning because of the greater probability of finding the global minimum and lesser numerical cost. Therefore, the algorithm was successfully used for model‐based offline tuning of sliding mode controller parameters for a servo‐pneumatic position control application. However, such offline tuning requires a sufficiently adequate system model, which, in some applications, is difficult to attain. Moreover, it is not generally appreciated as an essential requirement for controller tuning by the end user like the industry. An improvement in performance of optimization algorithm for tuning is expected if it relies on measurements coming directly from an actual physical system and not just its mathematical model. Therefore, in this paper, we apply the aforementioned algorithm for model‐free online optimization of controller parameters. The application involves the programmatic control of a real‐time interface of a physical system by the algorithm for data flow and logical decisions for optimization. For comparison with the results of the model‐based offline tuning suggested in earlier work, the sliding mode controller parameters are tuned online for the same position control application. The experimental results reveal that the system performance with controller parameters tuned online using the algorithm compares favorably to the one with model‐based offline tuning especially at higher priority level for accuracy. The improvement in system performance amounts to 21%.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".