Onboard Diagnostics Battery Fault Precursor Patents: Trends 2026
- Filing has plateaued. 2017 filings sat at 89 and 2025 peaked at 535, but the 2022 midpoint of 348 already shows growth flattening rather than compounding.
- No single filer dominates. The leader holds 104 records against 3,289 total, and the top 5 combine for just 10.8% of all records in scope — a long tail of single- and low-filing entrants fills the rest.
- Software classes outweigh sensing classes. G06F and G06N (data processing and AI models) each cover over a third of records, while G01R (electric/magnetic measurement) — the hardware sensing class — sits at only 8.1%.
What this landscape covers
Onboard diagnostics for battery fault precursors sits at the intersection of anomaly detection algorithms and in-vehicle sensing. This dataset spans 3,289 published records filed between 2015 and mid-2026, capturing everything from continuous behaviour-modelling systems adapted from IT anomaly detection to control-system signal analysis methods repurposed for early internal-short-circuit warning. Many of the most-cited records in the corpus originate outside automotive proper, in network and data-flow anomaly detection, and were later drawn into battery diagnostics claim language as filers reused established detection architectures.
Because publication lags filing by roughly 18 months, the 2025 and 2026 figures in any trend chart understate real filing activity for those years. Read the most recent two years as a floor, not a ceiling.
Filing trend and technology composition
Two views of the same 3,289-record corpus: the year-by-year filing count, and the IPC subclasses those records fall into. A record can carry more than one IPC class, so the composition shares add up to more than 100%.
Filing trend, 2017–2026
Filings rose from 89 in 2017 to a peak of 535 in 2025, but the 2022 midpoint of 348 sits well below a straight line to that peak — activity built early, then flattened rather than compounding year over year. The 2026 figure of 150 is a partial year and will revise upward as later publications post.
IPC subclass composition
G06F (electric digital data processing) and G06N (AI-based computing) each cover more than a third of all records, and H04L (digital information transmission) covers close to 30% — together they mark this as a software- and algorithm-heavy field. G01R (electric and magnetic measurement), the class closest to physical cell sensing, covers only 8.1% of records, and G06V (image/video recognition) covers 6.2%.
Shares are the percentage of the 3,289 records in scope. A patent can carry several IPC classes, so the shares add up to more than 100%.
Go deeper on Onboard Diagnostics for Battery Fault Precursors with Eureka
This page is one run against one query. Ask Eureka your own question about onboard diagnostics for battery fault precursors and every answer comes back with the patent numbers behind it.
Try EurekaMost-cited records and a representative filing
Modbus TCP communication behaviour anomaly detection method based on OCSVM dual-outline model
Proposed is an anomaly detection method for communication behaviours in an industrial control system based on an OCSVM algorithm. A normal behaviour profile model and an abnormal behaviour profile model, i.e. a dual-outline model, of communication behaviours are established, parameter optimization is performed by means of a particle swarm optimization (PSO) algorithm, an optimal intrusion detection model is obtained, and abnormal Modbus TCP communication traffic is identified. The false alarm rate is reduced by means of cooperative discrimination of the dual-outline detection model.Filed by Shenyang Institute of Automation, Chinese Academy of Sciences (2017-11-16). Its dual-outline, false-alarm-reduction approach is representative of how industrial anomaly detection methods get adapted toward the signal-monitoring problem this field addresses.
| # | Publication no. | Patent title | Citations |
|---|---|---|---|
| 1 | US20120137367A1 | Continuous anomaly detection based on behavior modeling and heterogeneous information analysis | 1,758 |
| 2 | US20070192863A1 | Systems and methods for processing data flows | 959 |
| 3 | US20120240185A1 | Systems and methods for processing data flows | 859 |
| 4 | US20070121596A1 | System and method for providing network level and nodal level vulnerability protection in VoIP networks | 763 |
| 5 | US20080229415A1 | Systems and methods for processing data flows | 636 |
| 6 | US20140096249A1 | Continuous anomaly detection based on behavior modeling and heterogeneous information analysis | 598 |
| 7 | US20180288063A1 | Mechanisms for anomaly detection and access management | 457 |
| 8 | US20080262990A1 | Systems and methods for processing data flows | 441 |
| 9 | US20080262991A1 | Systems and methods for processing data flows | 430 |
| 10 | US8135657B2 | Systems and methods for processing data flows | 426 |
Citation counts favour older records simply because they have had longer to accumulate citations inside the searched corpus — treat this table as a map of influential prior art, not of current filing activity. Several of the highest-cited records originate in general data-flow and network anomaly detection rather than battery diagnostics specifically, which is itself informative about where the underlying detection architectures came from.
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 data tells you
Three patterns worth acting on before drafting claims or scoping freedom-to-operate in this space.
The field is not gatekept by a handful of filers
The leading assignee holds 104 records and the top 5 combined cover 10.8% of all 3,289 records in scope. That is a modest concentration for a field this size — most of the corpus is a long tail of entrants with a handful of filings each, which leaves room for new filers to establish position without displacing an entrenched leader.
Recent-year activity has cooled among established filers
Several of the higher-ranked assignees show zero filings in the latest year, with year-over-year changes of -100% against prior activity. Combined with a 2025 peak that already outpaces the 2022 midpoint by a wide margin, this points to filing having concentrated in a burst rather than sustaining a steady climb.
Claim language skews toward algorithms, not sensors
G06F and G06N each appear in over a third of records and H04L in nearly 30%, while G01R — the class tied to actual electric and magnetic measurement — covers only 8.1%. Filers are largely claiming the detection and processing layer; the physical sensing and measurement layer is comparatively thin.
Eureka can read the same corpus for gaps instead of for coverage: under-claimed branches adjacent to onboard diagnostics for battery fault precursors, with the prior art for and against each one.
Who is filing, and where the openings sit
The ranked assignees skew toward large diversified technology and telecom filers whose detection architectures were adapted into this space, rather than dedicated battery or automotive specialists holding a commanding position.
A diversified technology filer sets the pace
The top-ranked assignee's 104 records outpace the rest of the field, but that share is still a modest slice of the 3,289-record corpus. Its portfolio draws on detection architectures built for IT and network anomaly detection, applied to the fault-precursor problem.
The drop-off from top to mid-field is steep
Fifth place holds 46 records and tenth place holds 22 — roughly half again by the time you reach the tenth ranked filer. This is a field where a modest, focused filing programme can still land inside the ranked leaders.
Co-filing is limited and concentrated in a few relationships
Only 10 co-assignee pairs appear in the corpus, and the strongest pairs recur around the same telecom and IT-conglomerate filers rather than spanning automotive OEMs and battery makers. Cross-sector collaboration between battery hardware specialists and detection-algorithm specialists is not yet a visible pattern here.
| Assignee | Recent year | YoY |
|---|---|---|
| AT&T Intellectual Property I, L.P. | 1 | -50% |
| Oracle International Corporation | 0 | -100% |
| Microsoft Technology Licensing, LLC | 0 | -100% |
| International Business Machines Corporation | 0 | -100% |
| Panasonic Intellectual Property Corporation of America | 0 | — |
| Telefonaktiebolaget LM Ericsson (publ) | 0 | -100% |
| Huawei Technologies Co., Ltd. | 0 | — |
| Tata Consultancy Services Ltd. | 0 | — |
Where to take this
The dataset points to a field with occupied algorithmic claim space and thinner coverage on the physical sensing and action-policy layers.
Map freedom-to-operate against the software-heavy classes
With G06F, G06N and H04L covering the bulk of records, any new detection-algorithm claim should be checked against this dense prior art before drafting, particularly where it touches behaviour-modelling or false-alarm-rate reduction language.
Run a claim-level search in Eureka →Explore the sensing and action-policy gap
G01R coverage sits at 8.1% of records, well below the algorithmic classes, suggesting under-claimed room in physical precursor sensing and in what a system does once an anomaly is flagged.
Explore white space in Eureka →Track the assignees still filing after the 2025 peak
Several ranked leaders show zero filings in the latest year. Watching which filers keep filing past the 2025 peak will show who is committing to this space versus who filed opportunistically.
Set up assignee monitoring in Eureka →Common questions
In this corpus, a fault precursor is a signal feature or pattern detected before an internal short circuit or other battery failure becomes an observable fault, rather than the fault itself. The search terms behind this dataset specifically target early-stage detection language such as signal features preceding a fault and internal short circuit early detection, distinguishing it from post-fault diagnostic trouble codes. Most of the claim language sits in the detection algorithm layer (anomaly detection, behaviour modelling) rather than in a specific sensor design, so precursor detection here is largely a data-processing problem applied to battery signals.
The leading assignee in this 3,289-record corpus holds 104 records, with the top 5 assignees combined covering 10.8% of all records in scope. That is a modest concentration for a field this size: most of the ranked leaders hold well under 50 records each, and the ranking includes 100 companies total, not just a handful of dominant players. Several of the leading filers are diversified technology and telecom firms whose detection architectures originated in IT and network anomaly detection before being applied to this problem.
Filing rose from 89 records in 2017 to a peak of 535 in 2025, but the 2022 midpoint of 348 shows growth had already begun flattening well before that peak. Several of the higher-ranked assignees also show zero filings in the latest year, with year-over-year changes of -100% against their prior activity, which points to cooling momentum among established filers rather than a sustained upward trend. Because publication lags filing by roughly 18 months, the most recent one to two years in any trend will always look lower than they eventually turn out to be, so this plateau should be read cautiously rather than treated as a firm peak.
The measurement and sensing class, G01R, covers only 8.1% of the 3,289 records in scope, well below the algorithmic classes like G06F and G06N which each exceed a third of records. That gap suggests physical cell-level sensing methods for internal short-circuit precursors, and the action-policy layer governing what a system does once an anomaly is flagged, are comparatively open relative to detection-algorithm claims. Co-assignee filing is also limited to just 10 pairs in this corpus, so cross-sector collaboration between battery hardware specialists and algorithm specialists is not yet a crowded space either.
Not necessarily. Citation counts inside a searched corpus accumulate over time, so older records have simply had more opportunity to be cited than recent filings, which biases the most-cited list toward age rather than current relevance. Several of the highest-cited records in this dataset originate in general data-flow and network anomaly detection rather than battery diagnostics specifically, reflecting where the underlying detection techniques were first patented. Citation rank is best read as a map of influential prior art architecture, not as a signal of which claims matter most for today's product decisions.
Research Onboard Diagnostics for Battery Fault Precursors in depth with Eureka
Go past this page: query the whole onboard diagnostics for battery fault precursors 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.