This dataset identifies 1,116 patent families published between 2015 and the 2026-07-31 data cut-off that combine PID controller, process control or model predictive control language with tuning-specific claim terms such as overshoot, anti-windup or setpoint tracking, within the G05B11, G05B13 and G05B19 IPC classes. That figure counts families, which is the fairer unit than raw document counts because it removes duplication from continuations and multi-jurisdiction filing of the same invention. The true figure is likely somewhat higher once the most recent one to two years finish publishing, since filing-to-publication lag runs around eighteen months.

The full assignee ranking spans all 1,116 families and includes both multinational instrumentation vendors and smaller regional filers, so no single company holds a runaway majority. Notably, several historically prominent names — including major automation and instrumentation vendors — show zero filings in the most recent year captured, which suggests either a pause in patenting or a shift in filing strategy rather than exit from the field. Anyone assessing competitive position should look at the momentum data alongside the raw ranking, not the ranking alone.

Growth is uneven rather than steadily rising. Filings moved from 7 in 2017 to a midpoint of 29 in 2022 and up to a peak of 62 in 2025, but the trajectory between those points was not a smooth climb, and the 2026 count sits below the 2025 peak. Because 2026 is only partially published, some of that apparent slowdown is a lag artefact — but the flat stretch around the midpoint year is a genuine feature of the data, not an artefact of the cut-off.

US20150323912A1, filed by King Fahd University of Petroleum and Minerals and published in November 2015, claims a closed-loop setpoint-step method for tuning PI/PID controllers on stable and integrating processes with time delay. It improves on the classical Ziegler-Nichols continuous-cycling method by deriving controller settings directly from three measured quantities — overshoot, time to first peak, and relative steady-state output change — taken from a single setpoint experiment run with a P-controller. Anyone building a closed-loop, setpoint-experiment-based tuning method that extracts settings from those same three data points should review this claim scope closely before finalizing an approach.

The clearest under-claimed areas sit at the boundaries of the core G05B classification: power-conversion contexts (H02J and H02M, 19 and 18 records respectively) and catalytic or chemical-process contexts (B01J, 17 records) show much thinner filing density than the 1,115-record G05B core. Testing and balance instrumentation tied to tuning claims (G01M, 15 records) is similarly light. These are not proof of open claim space on their own, but they are reasonable starting points for a targeted freedom-to-operate search before committing to a filing strategy.