One assignee leads with 152 of the 876 records in this dataset, well ahead of the rest of the field — fifth place holds 29 records and tenth place holds 19. The top five combined account for 33.6% of all records in scope, and the top ten for 45.8%, after which filing counts drop off into a long tail of companies with only a handful of filings each. This shape means competitive monitoring should weight the leader heavily rather than spreading attention evenly across the ranked list.

Filing activity rose from 45 records in 2017 to a peak of 84 in 2022, then eased to 75 by 2024 — a 9% decline over that three-year span, not a collapse. Figures for 2025 and 2026 look lower still, but publication typically lags actual filing by around 18 months, so those most recent years are undercounted rather than genuinely quiet. A fair read is that the field has cooled modestly from its 2022 high rather than that it is shrinking.

Backtracking and control logic fall mainly under G05D (control of non-electric variables), which covers 9.2% of the 876 records, with a smaller AI-specific slice under G06N at 3.2%. Most records instead carry core photovoltaic classification under H02S (58.4%) or solar heat collector structure classes F24S and F24J (49.4% and 23.2%), since backtracking claims are usually filed alongside a structural or PV-specific claim rather than standing alone. Anyone drafting a pure control-algorithm claim should expect comparatively less crowded prior art in G05D and G06N than in the structural classes.

US12393208B2, assigned to OMCO Solar and published in August 2025, claims a backtracking method that calculates table inclination angles separately for defined morning and evening backtracking periods, specifically to reduce adjacent-row shading compared with normal tracking mode, and it ties that calculation to imaging the ends of each assembly's torque tube beam. It is a refinement of an established backtracking approach rather than a foundational tracker design, so it is most relevant to anyone implementing time-segmented backtracking with beam-end imaging rather than to broad single-axis tracker claims generally. Design-around discussions should focus on the specific imaging step and the period-segmentation logic, not on backtracking as a general concept.

The clearest gaps sit in classes with real but thin filing activity relative to the core tracker problem: G05D control logic (9.2% of 876 records), G06N AI-based methods (3.2%), G01S positioning (5.6%) and E02D foundation engineering (5.5%). Sub-areas like torsional galloping mitigation, AI-driven backtracking optimisation, and pile-foundation slope tolerance show up in the search criteria but are not heavily claimed compared with structural and photovoltaic classes. A first claim in these areas would likely combine a specific sensor or control input with a defined structural constraint, rather than claiming a general tracking or stow method broadly.