Adaptive Soft Sensor Patents: Who Leads, Where the Gaps Are 2026
- No single filer dominates. the ranked leader holds just 276 records, and the top 5 combined account for only 4.6% of all 18,298 records in scope.
- Filing activity peaked in 2017 at 1,404 records and has since declined toward a 2022 midpoint of 327, with the most recent year understated by publication lag.
- The technology reads across into unexpected classes, with meaningful shares in conveying/material handling (B65G) and containers/packaging (B65D) alongside the expected data-processing classes.
Filing growth compares 2021 (279 records) with 2024 (183) — a three-year span. 2024 is the most recent year we treat as complete: publication lags filing by roughly 18 months, so 2025 onwards are still filling in and any growth rate that ends there would understate the field. Top-5 share is the combined record count of the five largest assignees divided by all 18,298 records in scope (CR5), not by the ranked leaders only.
What this landscape covers
Adaptive and just-in-time soft sensors use statistical or machine-learning models to infer hard-to-measure process variables in real time, updating themselves as process conditions drift. The patent activity mapped here spans 18,298 records filed or published between 2015 and mid-2026, drawn from claims and description text referencing drift detection, retraining triggers, local model selection and related maintenance mechanisms. The dataset captures both the core algorithmic techniques and their downstream applications in process industries.
Filing activity is concentrated less by company than by technique: the bulk of records sit in general data-processing and digital-transmission classes, with smaller but persistent pockets in materials handling, packaging and wireless networking that point to soft-sensor logic migrating into physical process equipment.
Filing trends and technology composition
Two views of the same 18,298 records: how filing volume has moved year over year, and which IPC subclasses carry the underlying technique.
A 2017 peak followed by a decline
Filings reached 1,404 in 2017, the high point of the period, then fell toward a 2022 midpoint of 327. That trajectory reads as flat-to-declining rather than growing, though the final one to two years are understated because publication typically lags filing by around 18 months.
Data processing dominates, but the tail is wide
G06F (electric digital data processing) carries the largest share at 14.6% of all records, followed by H04L (digital information transmission) at 6.4%. Below those two, shares drop off quickly: G06Q, B65G, B65D, H04N, G01N and H04W each sit in the 2.5%-4.2% range, showing the technique diffusing into commerce systems, physical handling equipment, video pipelines, material testing and wireless networks rather than staying confined to a single application class.
Shares are the percentage of the 18,298 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Go deeper on Adaptive and Just-in-Time Soft Sensors with Eureka
This page is one run against one query. Ask Eureka your own question about adaptive and just-in-time soft sensors and every answer comes back with the patent numbers behind it.
Try EurekaRepresentative filing and most-cited prior art
Validating programmed execution sequences for a robot working cell
This filing describes a robot and/or robot controller together with a method for validating programmed workflow sequences or teaching programs, applied to a robot mounted on or beside a processing machine such as an injection moulding machine, for handling parts that have just been produced. Travel parameters, equipment features and functionalities of the physical robot are stored in a configuration file on the control side, and the method checks a proposed sequence against that stored configuration before it runs.Abstract text is drawn directly from the filing as published.
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US8045952B2 | Method and device for obtaining playlist content over a network | 1,802 |
| 2 | US20060059253A1 | Architectures for netcentric computing systems | 1,376 |
| 3 | US6014643A | Interactive securities trading system | 1,067 |
| 4 | US20020002039A1 | Network-enabled audio device | 1,062 |
| 5 | US7020697B1 | Architectures for netcentric computing systems | 1,038 |
| 6 | US5984779A | Continuous real time Pari-Mutuel method | 889 |
| 7 | US20010044751A1 | System and method for displaying and selling goods and services | 861 |
| 8 | US4489574A | Apparatus for highly efficient laundering of textiles | 849 |
| 9 | US6163607A | Optimizing call-center performance by using predictive data to distribute agents among calls | 818 |
| 10 | US4489455A | Method for highly efficient laundering of textiles | 788 |
Citation counts favour older records simply because they have had more time to accumulate citations within the searched corpus; treat them as a signal of influence, not current relevance.
Each row carries its publication number; clicking a row searches Eureka by that number.
Put your own technology through the same analysis
Eureka on the web
When you want the answer in the next five minutes.
The agent works the prompt against patents and technical literature, citing every source.
Run your analysis now →MCP server & REST API
When it has to run inside your own pipeline.
Patent search, landscape analysis and assignee resolution as MCP tools. Drop them into any agent framework, or call REST directly.
Browse MCP servers →What the numbers mean for a filing decision
Three patterns stand out once concentration, timing and classification are read together.
No gatekeeper controls this space
The leading assignee holds 276 records out of 18,298, and the five leading assignees combined account for only 4.6% of all records in scope. That is a fragmented field rather than one guarded by a small number of blocking portfolios, which lowers freedom-to-operate risk for a new entrant relative to fields with a dominant top-5 concentration.
Activity has cooled since 2017
Filings peaked at 1,404 in 2017 and had fallen to a 2022 midpoint of 327; the most recent year shows only 76 records, though that figure is understated because publication lags filing by roughly 18 months. The overall shape is flat-to-declining rather than an emerging growth curve.
The technique reads across into physical handling
G06F and H04L cover the expected algorithmic core, but B65G (conveying and material handling) and B65D (containers and packaging) each carry a real share of records. That suggests adaptive sensing logic is being claimed inside physical process equipment, not just in standalone software.
Co-filing is rare and clustered
Only 10 co-assignee pairs appear across the dataset, and the strongest pairings repeat the same lead assignee against different partners. Joint filing is not a meaningful route into this space; most activity is single-assignee.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to adaptive and just-in-time soft sensors, with the prior art for and against each one.
Who is filing, and where the gaps sit
The assignee ranking spans 100 companies, none of which holds a commanding share; recent-year momentum has also cooled across the assignees with the largest historical volumes.
A modest lead, not a monopoly
The top-ranked assignee holds 276 records against a field of 18,298, and by fifth place the count has already dropped to 111. That gradient is shallow enough that a well-drafted new filing is competing against volume, not against a small set of dominant blocking patents.
Historical leaders have gone quiet
Several of the assignees with the deepest historical portfolios show zero filings in the latest year, with year-on-year declines of -100% recorded for at least two. That pattern is consistent with the broader post-2017 decline rather than a sign any one company has exited deliberately.
Filing is overwhelmingly solo
Co-assignee pairs number only 10 across the whole dataset, with the strongest pairs repeating one lead assignee against separate telecom and data-processing partners. This is not a field organised around joint ventures or shared portfolios.
| Assignee | Recent year | YoY |
|---|---|---|
| International Business Machines Corporation (IBM) | 0 | — |
| Microsoft Corporation | 0 | — |
| Hitachi, Ltd. | 0 | — |
| Telefonaktiebolaget LM Ericsson | 0 | -100% |
| Toshiba Corporation | 0 | — |
| Microsoft Technology Licensing, LLC | 0 | -100% |
| Canon Inc. | 0 | — |
| NEC Corporation | 0 | — |
Where to take this
Two follow-on moves for a team deciding whether and where to file.
Check freedom to operate against the ranked leaders
With no assignee above 276 records out of 18,298, a targeted clearance search against the ranked leaders is more tractable than in fields with a dominant top-5 concentration.
Explore assignee portfolios in EurekaMap claim language in the under-claimed branches
The IPC composition shows thinner activity outside the core data-processing classes; drafting claims that anchor drift-detection or retraining logic to a specific physical application may face less prior art.
Run a white-space search in EurekaCommon questions about this landscape
No single company dominates this landscape. The leading assignee holds 276 records out of 18,298 total, and the five leading assignees combined account for only 4.6% of all records in scope. This is a fragmented field with a long tail of filers rather than a small number of companies controlling the core claim space, which matters for freedom-to-operate assessments.
Filing activity peaked in 2017 at 1,404 records and had fallen to 327 by the 2022 midpoint, with only 76 records recorded in the most recent partial year. The overall trend reads as flat to declining rather than emerging growth. Note that the most recent one to two years are understated in any patent dataset because publication typically lags the original filing date by around 18 months.
The largest concentration sits in G06F, electric digital data processing, covering 14.6% of the 18,298 records in scope, followed by H04L for digital information transmission at 6.4%. Smaller but meaningful shares also appear in G06Q (business data processing), B65G (conveying and material handling), B65D (containers and packaging), H04N (pictorial communication), G01N (material analysis and testing) and H04W (wireless networks), showing the technique applied well beyond pure software contexts.
The clearest gaps sit at the intersection of the core algorithmic techniques and physical process equipment, where IPC shares are thinner than in the dominant G06F and H04L classes. Sub-areas such as drift-triggered local model re-selection tied to material handling lines, or adaptive calibration logic specific to packaging equipment, show lower filing density than the general data-processing core. A first claim in these areas would need to tie the retraining or drift-detection mechanism explicitly to the physical process step rather than claiming it as generic software.
Rarely. Only 10 co-assignee pairs appear across the entire 18,298-record dataset, and the strongest pairings all involve the same lead assignee paired with different individual partners. Joint filing is not a common strategy in this space; the overwhelming majority of records are filed by a single assignee, which means competitive mapping should focus on individual portfolios rather than alliance structures.
Research Adaptive and Just-in-Time Soft Sensors in depth with Eureka
Go past this page: query the whole adaptive and just-in-time soft sensors corpus yourself, in your own scope.
Every answer comes back with patent numbers you can open.
Disclaimer. This page is generated from Patsnap Eureka data drawn from a limited snapshot of global patent and scientific-literature records, and is provided for general information and reference only.
Patent data carries inherent limitations: recent filings (typically the most recent 18–24 months) are under-counted due to standard publication lag; counts may be reported at either a patent-family or a patent-record basis and are not always directly comparable; classification, applicant-name, and citation data may contain errors, duplicates, or omissions; and the underlying search query defines and constrains the scope shown. As a result, the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing on this page constitutes an exhaustive prior-art, novelty, freedom-to-operate, or validity search, nor does it constitute legal, financial, investment, or professional advice, and it should not be relied upon as such. Any patent, commercial, or strategic decision should be verified independently and reviewed with qualified patent, legal, and domain professionals. Patsnap makes no warranties, express or implied, as to the accuracy, completeness, or fitness for any particular purpose of the information presented.
Machine translation. Assignee and organisation names originally recorded in Chinese, Japanese or Korean have been rendered into English by an AI translation step so that the tables stay readable. These renderings are best-effort and may not match a company’s registered English name; the original name is what the underlying patent record carries, and it is what any Eureka query launched from this page uses.