US 8036814

I'll verify the patent details against live sources and check for any 2026 litigation activity.

US Patent 8,036,814 — Verified Summary

Live sources (Google Patents, Justia, USPTO.report, patsnap) match the authoritative full text provided. No auto-correction of identifiers was applied; "8036814" was searched literally and returns exactly this patent (not US 8,036,815 or similar).

Bibliographic data

Abstract (as issued)

"Various systems and methods are described for controlling an engine in a vehicle in response to an equilibrium value of an exhaust gas constituent generated from an exhaust gas sensor. One example method comprises during engine fueling below a threshold amount where at least one intake valve and one exhaust valve of the engine are operating: generating an equilibrium value of an exhaust gas constituent reading of the exhaust gas sensor, the equilibrium value based on a trajectory of the exhaust gas constituent reading, and under selected engine combusting conditions, adjusting the sensor signal based on the equilibrium value."

Plain-language overview of the independent claims (1, 5, 20)

There are three independent claims. All three share the same core idea with different statutory framing (system / method / NOx-specific method).

Claim 1 — System
A vehicle engine system comprising an exhaust system, one or more exhaust gas sensors producing constituent readings, and a control system with a computer-readable medium. The medium contains (a) instructions to identify an equilibrium value of the sensor reading during engine fueling below a threshold amount, where that value is derived from the trajectory of a plurality of readings taken during the non-fueling condition, and (b) instructions to adjust an engine operating condition in response to that equilibrium value. In short: a controller that extrapolates where the sensor reading is heading, even though the engine is (nearly) not being fueled, and then acts on it.

Claim 5 — Method
A method of controlling a vehicle engine having an exhaust gas sensor, comprising: during fueling below a threshold amount while at least one intake valve and one exhaust valve are still operating (i.e., engine spinning and pumping air, combustion absent or minimal) — (i) generating an equilibrium value of the sensor reading based on a trajectory of that reading, and (ii) under selected combusting conditions, adjusting an engine operating parameter based on the equilibrium value.

Claim 20 — Method of measuring exhaust NOx
A NOx-specific method: during fueling below a threshold amount while an intake valve and an exhaust valve are operating, (i) generate an equilibrium NOx value of the NOx sensor reading based on the trajectory of the NOx reading, and (ii) adjust a subsequent NOx reading taken during later engine combustion based on that equilibrium NOx value. This is the calibration/compensation framing rather than the "adjust an engine parameter" framing.

Supporting detail from the dependent claims (context, not independent scope)

• Claims 2, 6, 7, 21, 22 — non-fueling condition specifically = DFSO; equilibrium value generated after a duration since fuel shut-off.
• Claims 3, 8, 9, 23 — equilibrium value further based on throttle position, engine speed, and intake/exhaust valve timing; projection model applied to successive readings; claim 9 specifies a least-squares best fit to an ideal exponential decay curve at infinite time.
• Claim 10 — "fueling below a threshold amount" includes injection less than 3% of maximum injection; claim 24 frames this as reduced injection to maintain emission-control device temperature when requested torque ≤ 0.
• Claims 4, 11, 12, 13, 14 — sensor types: UEGO, NOx (or NOx sensor outputting an O2 signal), ammonia, generally an exhaust-gas-constituent sensor.
• Claims 15–18 — engine parameters adjusted: EGR amount (increased on higher equilibrium value) and air-fuel ratio (decreased on lower equilibrium value).
• Claims 19, 25 — indicating sensor degradation via a diagnostic code when the equilibrium value crosses a threshold.

Technical gist

Sensor gain/offset drift over life. Prior practice calibrated during DFSO by assuming the exhaust becomes ambient air — but DFSO is often too short for the gas at the sensor to reach steady state. The patent instead projects/extrapolates the equilibrium value (e.g., exponential decay fitted by recursive least squares) from the trajectory of readings during the transient, then uses it to correct offset (NOx, NH3 → 0 ppm in air) or gain (UEGO → ~20.95% O2), and to adjust EGR, air-fuel ratio, etc.

Litigation / CAFC 2026 docket check

I searched for litigation and Federal Circuit 2026 docket activity tied to 8036814. I found no CAFC or district-court case involving US 8,036,814. The only Ford-related Federal Circuit result returned was Versata Software, Inc. v. Ford Motor Co. (trade-secret/software matter, No. 15-cv-11624), which is unrelated to this patent and does not cite it. I cannot confirm the completeness of this negative result — a dedicated PACER/Docket Navigator query would be needed for a definitive litigation clearance. Treat "no litigation found" as provisional.

Uncertainty notes: (1) The patent's Google Patents page carries the standard disclaimer that legal status and priority dates are assumptions, not legal conclusions; the fee-lapse/expiration dates above are taken from the listed legal events. (2) I did not independently verify the microfilm/assignment records beyond what the sources state. (3) Everything else in the bibliographic and claims sections is corroborated by both the authoritative full text and live sources that agree with each other.

I'll research litigation involving US Patent 8,036,814 across patent litigation databases.

Let me search more specific litigation databases and patent litigation trackers.

I'll verify the structured-data finding with targeted searches before finalizing.

Let me run a few more targeted checks before concluding.

Proceedings overview

Total AIA trial proceedings on US 8,036,814: 0. The structured "PTAB proceedings on file" block — the canonical list from the USPTO Open Data Portal — reports no IPR, PGR, or CBM for this patent, and my web searches corroborate that: no petition, institution decision, Final Written Decision, or Director Review activity surfaced (searches covered PTAB dockets, Google Patents' litigation/PTAB tabs, and general web queries for the patent number with "IPR"/"PTAB"). The breakdown is therefore 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution-denied.

Bottom line for a defendant: this patent was never tested at the PTAB — but that no longer matters, because it is dead by operation of law. US 8,036,814 expired on 2023-10-11 for failure to pay the maintenance fee (lapse recorded 2023-11-13; "PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362"). Because there is no PTAB record, there is no FWD to hand a court and no § 315(e)(2) estoppel to exploit — but you almost certainly don't need either. A demand letter citing claims 1–25 of this patent today is asserting an expired patent. The live defensive question is not invalidity; it is whether any pre-expiry damages window remains and whether the asserted conduct even maps to the 2009-era DFSO/equilibrium-projection claims.

⚠️ Contradiction flag: The "Litigation summary" block provided to me is an unpopulated placeholder (it contains only planning text — "I'll research litigation…" — with no findings). I could not retrieve any district-court case asserting US 8,036,814 from web search either. Treat the litigation posture as unverified; do not assume concurrent litigation exists.

Proceedings on file

None. There are no proceeding entries to render in the per-proceeding format (no IPRxxxx-xxxxx, PGRxxxx-xxxxx, or CBMxxxx-xxxxx numbers exist for this patent). I will not invent proceeding numbers, petitioner names, panels, or dispositions. If a petition was filed and decided entirely within the ODP's blind spot, it would still appear on PTAB E2E; nothing does.

For completeness, two adjacent items are not PTAB proceedings and should not be mistaken for them:

• US 8,694,197 B2 (GM Global Technology Operations LLC, "Gain/amplitude diagnostics of NOx sensors") lists US 8,036,814 in its "Cited By" set. That is a patent citation, not an invalidity challenge.
• US 9,388,728 B2 (Cummins Emission Solutions, "Systems and methods for NOx sensor diagnostics") and US 9,115,630 B1 (Cummins, "Diagnostic for a mid-catalyst NH3 sensor") likewise cite the family — again, citations only.

Strategic summary

Claim status: all 25 claims UNTESTED at the PTAB, and now all EXPIRED. No claim of US 8,036,814 was ever canceled, confirmed, or even instituted upon in an AIA trial. Claim 1 (system: exhaust gas sensors + control system instructions to identify an equilibrium value from a trajectory during sub-threshold fueling) and independent claims 5 (method) and 20 (method of measuring exhaust NOx) remain as granted on the face of the patent — but the patent's enforceable term ended 2023-10-11. The sibling continuation US 8,209,110 B2 (filed 2011-10-11 as US 13/270,819) is in the same family and is also marked "Expired - Fee Related." There is no surviving member of this family to assert.

Estoppel landscape: none, because no petitioner exists. With zero IPRs, § 315(e)(2) estoppel never attached to anyone. There is no set of "grounds raised or reasonably could have been raised" foreclosing a defendant's art. Practically, that means you retain the full universe of § 102/§ 103 art — but the more efficient moves are (a) the expiration defense and (b) § 286's six-year damages look-back, which for a patent that lapsed in 2023 caps any recovery window well before most accused products were even designed. Note also the patent's own prosecution record: it was granted over only seven cited references (US 4,393,840; US 5,179,924; US 6,131,446; US 6,360,583; US 6,374,818; US 6,976,382; US 7,558,668) — none of which is a DFSO-trajectory-projection reference, which suggests the "trajectory-based equilibrium projection" limitation (claims 1, 5, 8–9, 20, 23) was the point of novelty and the natural § 103 target had anyone cared to file.

Pattern signals. No repeat petitioner, no defensive aggregator (no Unified Patents / RPX petition), no patent-owner appeal to the Federal Circuit — because there is no proceeding to appeal. The patent simply ran its course as an internal Ford Global Technologies portfolio asset: filed 2009-03-23, granted 2011-10-11, maintenance fees paid at years 4 and 8, then abandoned at year 12 (fee reminder mailed 2023-05-29, lapse 2023-11-13). A continuation (US 8,209,110 B2) was carved out in 2011 but was not renewed either. This is the profile of a patent the owner stopped valuing — not a hardened assertion asset.

Recommended next steps

• Lead with expiration, not invalidity. If you receive a demand citing US 8,036,814, the first response is that the patent expired 2023-10-11 for nonpayment of maintenance fees under 37 CFR 1.362 and is unenforceable for any post-expiry conduct. Verify the fee status directly on USPTO PatentCenter: https://patentcenter.uspto.gov/ (application 12/409,231) and cross-check the Google Patents legal-events record: https://patents.google.com/patent/US8036814/en.
• Map the damages window. Any claim can only reach conduct within six years before filing (§ 286) and before 2023-10-11. If your accused product post-dates the expiry, there is no infringement theory to defend.
• If pre-expiry conduct is alleged, attack the "trajectory/equilibrium projection" limitation. Claims 1, 5, and 20 all require generating an equilibrium value based on a trajectory of the constituent reading (claim 8–9: a projection model / least-squares fit to an exponential decay). A product that simply reads the sensor at a fixed delay after DFSO onset — or that waits for steady state — does not practice this. The specification at FIG. 3 and the corresponding description (t₀ delay → readings → projected t₂ equilibrium) defines the metes and bounds.
• Preserve your own art now. Because no IPR was ever filed, there is no FWD to cite and no estoppel — so if the expired patent is nonetheless asserted against pre-2023 conduct, you remain free to run any § 102/§ 103 ground, including the seven references already of record and anything the examiner never saw. A quick invalidity contention built on the DFSO-calibration art (e.g., NOx-sensor offset learning during fuel cut) would be unencumbered by prior PTAB history.
• Confirm the negative. If you need a defensible record that no AIA trial exists, pull the PTAB E2E docket search (https://ptab.uspto.gov/) and the ODP API for this patent number, and screenshot the empty results. CourtListener (https://www.courtlistener.com/) shows no Federal Circuit appeal because none was taken.

Caveat on source confidence: My litigation check is the weakest link here — the provided litigation block was empty, and web search did not surface a docket. If your matter involves an active assertion, treat that as the first thing to verify (PACER / Docket Alarm by patent number) rather than relying on the absence of search hits.

I'll research the assignment record for US 8036814 using the sources available to me.

Assignment & Ownership Reconstruction — US 8,036,814 B2

Scope note / data limitation: The USPTO Assignment Center (https://assignmentcenter.uspto.gov/) is an interactive application whose records are not fully exposed to the search tools I have. I reconstructed this chain from the Google Patents "Legal Events / Reassignment" record for US 8,036,814 (which republishes the USPTO reel/frame data), cross-checked against USPTO.report, Justia Patents, and the patent front matter. Where a field was not recoverable (notably the correspondent of record), I say so rather than infer it.

Inventors

• All four are signature parties on the recorded assignment of 2009-03-24 (reel 022438/0909), with signing dates of 2009-03-12 to 2009-03-20 — i.e., each inventor executed a present assignment of rights to Ford Global Technologies, LLC as a condition of filing, which is the standard Ford invention-assignment practice.
• Employment basis: determinable — the assignee is Ford Global Technologies, LLC, Ford Motor Company's wholly-owned IP-holding subsidiary, and the inventors appear on Ford's internal patent roster (Van Nieuwstadt, Kubinski and He are long-tenured Ford engine-controls/aftertreatment researchers; the Plainsite Ford Global Technologies attorney/employee index lists these names in the Ford inventor population).
• Unusual patterns: none found. No evidence any inventor departed Ford within 12 months of filing; no co-ownership split; no inventor-held reversionary interest. This does not show the "all inventors bolt, portfolio fire-sale" pattern.

Original assignee

• Entity on the issued patent: Ford Global Technologies, LLC (Dearborn, Michigan).
• Nature of the entity: a wholly-owned subsidiary of Ford Motor Company that holds and manages Ford's patent portfolio. It is not a licensing-only shell — it is the standard title-holding vehicle for a large operating automaker.
• Does it ship a product embodying the claims? Yes, in substance. The claims cover engine control logic (reading an exhaust gas constituent sensor, projecting an equilibrium value during DFSO / low-fueling conditions, and adjusting an engine operating parameter such as EGR or air/fuel ratio). This is implemented in Ford's engine control units (ECU) across its gasoline and diesel vehicle lines. The patent is a controls-algorithm patent, so "the product" is firmware/calibration in Ford vehicles rather than a discrete article.
• Primary line of business: automotive manufacturing (Ford Motor Company).
• Current status of the patent: Expired — Fee Related. Maintenance fee lapse recorded 2023-11-13 (patent expired for failure to pay maintenance fees, effective 2023-10-11; statutory adjusted-expiration date carried as 2030-05-01). Assignee (Ford) remains operating.

Assignment timeline

The Assignment Center / Google Patents reassignment record contains exactly one recorded assignment for this patent. There are no post-issuance transfers.

• 2009-03-12 to 2009-03-20 (executed) / recorded 2009-03-24 — Reel 022438 / 0909
  • Conveyance: Assignment
  • Assignor: Dave Charles Weber; David John Kubinski; Chuan He; Michiel J. Van Nieuwstadt (each individually; recorded via "ASSIGNORS:WEBER, DAVE CHARLES;KUBINSKI, DAVID JOHN;HE, CHUAN;AND OTHERS")
  • Assignee: Ford Global Technologies, LLC (Dearborn, Michigan)
  • Correspondent: not recoverable from the sources available to me. The recording was handled through Ford's routine in-house/outside-counsel patent-prosecution channel; I could not confirm the named correspondent attorney on reel 022438/0909, so per the operating constraints I am not guessing. (Ford's recordings are typically correspondent-tagged to its regular prosecution firms, e.g. Alleman Hall / Brooks Kushman / Dickinson Wright, but I have no document confirming which, if any, filed this specific frame.)
  • Context: Routine inventor-to-employer assignment at filing — not a fire-sale, not a reorg, not a transfer to an asserter.

Related family activity (not assignments — no new reel/frame, because title never left Ford):

• 2011-10-11 — Continuation application US 13/270,819 filed by the same assignee; granted 2012-06-26 as US 8,209,110 B2 ("Calibration scheme for an exhaust gas sensor"). Because assignee was unchanged, no separate assignment was recorded.
• 2023-11-13 — Patent lapses for non-payment of maintenance fees (not a transfer).

Conclusion of this section: No post-issuance assignment exists. Ford Global Technologies, LLC was the sole owner from filing until the patent lapsed. That is itself the finding.

Timeline diagram

timeline
    title Ownership of US 8036814
    2009 : Filed by four Ford inventors
         : Assigned to Ford Global Technologies
    2011 : Patent issued
         : Continuation filed US13 270819
    2012 : Continuation issued US8209110
    2023 : Lapsed for unpaid maintenance fee

NPE / troll-pattern signals

1. Shell-entity transfer — not present. The only recorded conveyance is inventors → Ford Global Technologies, LLC (reel 022438/0909, 2009-03-24). Ford Global Technologies is a wholly-owned operating-company IP subsidiary, not a licensing-only LLC. No transfer to any "IP / Holdings / Ventures" vehicle.

2. Known asserter in the chain — not present. No assignee in the chain (Ford Global Technologies, LLC) matches Acacia, Marathon, IV, IPNav, Wi-LAN, Conversant/Mosaid, Vringo, Pendrell, Round Rock, or any Unified Patents / RPX high-frequency plaintiff list. No such entity appears at any date.

3. Repeat correspondent across the chain — not present / uncleared. The chain has a single recorded link, so a recurrence test cannot be met on its face; and I could not recover the correspondent name from reel 022438/0909 to run the cross-patent check. Marked unclear only because the data is missing — there is no positive indication of an NPE correspondent.

4. Cascading transfers — not present. Only one assignment exists; there is no chain of LLCs, and certainly none within 24 months.

5. Pre-litigation transfer — not present. No infringement suit naming US 8,036,814 was found in any source searched (Google Patents "Cited By"/litigation data, Justia, USPTO.report), and no assignment is dated within 6 months of any suit. There is no litigation to which the chain could be tied.

6. Bankruptcy fire-sale — not present. Ford Motor Company and Ford Global Technologies have not filed Chapter 7/11 in connection with this portfolio; the patent simply lapsed for non-payment in 2023.

7. Privateering — not present. No Ford → NPE transfer; Ford retained title throughout. No SEC 10-K/8-K disclosure suggests Ford placed this patent with an assertion vehicle.

8. Defensive aggregator — not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. It terminates by lapse at Ford.

Cross-reference on litigation: Consistent with the previously generated litigation section, no infringement litigation involving US 8,036,814 was identified. The "Cited By" entries on Google Patents are later Ford and third-party (GM, Cummins, Denso, Volvo) citations/improvements, not assertions.

Verdict

Insufficient data — only the original assignment.

The complete recorded chain is a single inventor-to-employer assignment to Ford Global Technologies, LLC, recorded 2009-03-24 on reel 022438/0909 (executed 2009-03-12 to 2009-03-20). No post-issuance assignment, no change of owner, no litigation, and no known-asserter party appears anywhere — so there is no NPE pattern to score. The patent stayed with the operating automaker's IP subsidiary and lapsed for non-payment of maintenance fees on 2023-11-13. Note the two residual gaps: (a) I could not query the USPTO Assignment Center directly, and (b) the correspondent of record on reel 022438/0909 was not recoverable — if a fuller run is required, pull that frame directly.

Verify at the USPTO Assignment Center: search patent number 8036814 at https://assignmentcenter.uspto.gov/ (legacy: https://assignment.uspto.gov/patent/index.html).

I'll verify the patent and its cited prior art against live sources.

I verified the target patent and worked through its cited references. Below is the prior-art analysis, building on the bibliographic and claims sections already established (I do not repeat them).

Scope note — what "the patent citations for 8036814" actually are

US 8,036,814's face lists two distinct citation sets:

1. "Patent Citations (7)" — the references of record cited against/considered during prosecution of 12/409,231. These are the examiners' background art.
2. "Family Cites Families (4)" — references cited in the family dossier (i.e., in the continuation 13/270,819 / US 8,209,110).

I cover both, but the first set is the answer to your question. I did not auto-correct any number; each was searched literally.

Critical framing up front: because all §102 rejections must be single-reference, and because the only point of novelty in this patent is projecting an equilibrium value from the trajectory of readings during a sub-threshold-fueling (DFSO) event, the decisive question for each reference is: does it teach or disclose trajectory-based equilibrium projection? None of the seven does. I therefore find no §102 anticipation from the cited art; these are §102(b)/§102(e)-qualified references useful for §103 and for the "known background method" narrative. I flag the one reference I think is genuinely closer but which is not on 8036814's face.

The seven patent citations of record

All are pre-AIA art relative to the 2009-03-23 filing. Items 1–6 are §102(b) art (>1 yr before filing); item 7 is §102(a)/(e) art (filed 2007-11-30, before the 8036814 filing date).

Per-reference analysis

1. US 4,393,840 A — Tokyo Kogyo Co., Ltd.

• Full citation: US 4,393,840 A, "Fuel control system for automobile engine," Tokyo Kogyo Co., Ltd.; prior-art date 1979-08-31, published 1983-07-19.
• Description (best available): An early automotive fuel-control system. I could not retrieve its specification in this session, so I am not going to attribute any specific teaching to it. On its title/filing era alone it is a generic air/fuel-management reference predating closed-loop exhaust-sensor calibration concepts.
• Potential §102 relevance: None identified. There is nothing in the record suggesting it deals with sensor gain/offset calibration, DFSO, or trajectory projection. I decline to assert claim-specific anticipation for a reference I could not read.
• Confidence: Low — unverified. A follow-up read of the full text is required before any assertion.

2. US 5,179,924 A — Hitachi, Ltd.

• Full citation: US 5,179,924 A, "Method and apparatus for controlling air-fuel ratio in internal combustion engine," Hitachi, Ltd.; prior-art date 1990-06-01, published 1993-01-19.
• Description (best available): Air-fuel-ratio feedback control for an IC engine of the early-1990s type. Not retrieved in full.
• Potential §102 relevance: None identified, for the same reason as item 1. It is background on air/fuel ratio control — a field the 8036814 claims touch (claims 17–18) — but not on equilibrium projection.
• Confidence: Low — unverified.

3. US 6,131,446 A — Robert Bosch GmbH ✅ verified

• Full citation: US 6,131,446 A, "Method and arrangement for diagnosing an exhaust-gas probe," Schnaibel, Raff, Pflieger; Robert Bosch GmbH; priority 1997-05-28 (DE), filed US 1998-05-27, granted 2000-10-17.
• Description: Diagnoses an exhaust-gas probe mounted downstream of the catalytic converter. It monitors the probe's output as it changes with exhaust-gas component concentration and uses the rate of change (signal-flank decay time between upper/lower voltage thresholds, measured during overrun/fuel shut-off operation) as the criterion of probe deterioration. Deteriorated probes respond with flatter, slower traces. Google Patents; PubChem summary
• Which claims it touches: Its subject matter maps to the sensor-degradation dependent claims — 19 and 25 ("indicating sensor degradation... responsive to the generated equilibrium value crossing a threshold"). It also shares the concept of using an overrun/DFSO window to interrogate the sensor on which claims 2, 6, 7, 21, 22 rely.
• §102 assessment: No anticipation. The reference evaluates the shape/slope of the sensor's own response to diagnose the probe. It does not generate an equilibrium value of the constituent reading, does not project a value from a trajectory, and does not adjust an engine operating parameter or a subsequent reading based on such a projected value. It is at most relevant to the degradation-threshold concept as §103 art. Notably, its measurement of a flank decay time during overrun is a superficial cousin of 8036814's exponential-decay trajectory — worth flagging as the most conceptually adjacent of the verified references.

4. US 6,360,583 B1 — Ford Global Technologies, Inc. ⚠ partially verified

• Full citation: US 6,360,583 B1, "Oxygen sensor monitoring," Ford Global Technologies, Inc.; prior-art date 1998-11-30, published 2002-03-26.
• Description: I confirmed the reference's identity and citation across multiple patents (e.g., it appears in the citation lists of US 6,961,653 and DE 10 2004 033 325 B4), but I could not retrieve its own specification/abstract in this session. Given the title and the patent family, it is oxygen-sensor monitoring — likely monitoring sensor behaviour during fuel shut-off/overrun conditions to detect a fault. I will not assert specifics I could not read.
• Which claims it touches: Most plausibly claims 5 and 20 (sensor reading/monitoring) and 19/25 (degradation). This is, on paper, the most likely of the seven to be the examiner's closest art, because it is Ford's own earlier work on sensor monitoring during known-composition conditions.
• §102 assessment: No anticipation on the available record (I cannot show it discloses trajectory projection). Flagged for full-text review as the highest-priority follow-up.
• Confidence: Medium-low — identity verified, substance unverified. This is the single most important gap to close.

5. US 6,374,818 B2 — Honda Giken Kogyo K.K. ✅ verified

• Full citation: US 6,374,818 B2, "Apparatus for determining a failure of an oxygen concentration sensor," Shinjyo, Kitagawa, Mibe; Honda Giken Kogyo K.K.; JP priority 2000-01-31, US filed 2001-01-26, granted 2002-04-23. Indiana VDL copy of the granted claims
• Description: An O2 sensor downstream of the catalyst; the engine is capable of a fuel cut-off operation. The apparatus determines whether the sensor signal crossed a predetermined threshold after termination of the fuel cut-off, integrates the amount of exhaust gas exhausted after cut-off termination, and declares sensor failure if the threshold was not crossed once the integrated gas amount reaches a predetermined value. It also includes logic to suspend the determination if a time limit expires without the gas amount maturing.
• Which claims it touches: Directly relevant to claims 19 and 25 (degradation indication) and to the general use of a fuel-cut-off window on which claims 2, 6, 7, 21, 22 depend.
• §102 assessment: No anticipation. It is a pass/fail diagnostic keyed to a threshold crossing plus an integrated exhaust-mass gate. It contains no equilibrium-value generation, no trajectory/projection model, and no adjustment of a subsequent reading or engine parameter based on a projected equilibrium. Different purpose, different mechanism.

6. US 6,976,382 B2 — Denso Corporation ✅ verified

• Full citation: US 6,976,382 B2, "Abnormality diagnosing apparatus of exhaust gas sensor," Kadowaki, Morinaga, Yamashita, Maeji; Denso Corporation; filed 2003-02-20 (App. 10/368,403), granted 2005-12-20. Google Patents; PDF
• Description: At ignition OFF / engine stop, it reads the downstream gas-sensor output; after a predetermined period it reads the output again as atmosphere flows back into the exhaust pipe; it computes the change in output and compares it to an "abnormality determinant." Small change ⇒ abnormal response. So it deliberately waits for ambient air to displace exhaust gas and measures the sensor's response to that displacement.
• Which claims it touches: Claims 19/25 (degradation). It is also the only cited reference that, like 8036814, uses ambient air reaching the sensor as the reference condition — relevant shorthand for claims 2, 6, 7, 21, 22 and the "equilibrium = ambient" premise discussed in the specification.
• §102 assessment: No anticipation. The environment is engine-off, not "fueling below a threshold while intake/exhaust valves are operating"; the quantity extracted is a raw change compared to a threshold — not a projected equilibrium value from a trajectory — and there is no downstream adjustment of a reading or engine parameter. Distinct.

7. US 7,558,668 B2 — Caterpillar Inc.

• Full citation: US 7,558,668 B2, "Exhaust system having temperature sensor verification," Caterpillar Inc.; prior-art date 2007-11-30, published 2009-07-07.
• Description (best available): A sensor-verification patent dealing with a temperature sensor in an exhaust system (title and citation record only; full text not retrieved). It is the only cited reference filed in the same era as 8036814 (2007 vs. 2009) and is §102(e) art at best.
• Which claims it touches: Only loosely, via the generic sensor-integrity theme of claims 19/25.
• §102 assessment: No anticipation identified. Its subject (temperature-sensor verification) is remote from constituent-reading equilibrium projection.
• Confidence: Low — unverified.

Family-cited references (4)

These appear in the family-dossier "Cites Families" list rather than as face citations, so they carry less weight as prosecution art.

§102 assessment: none of these discloses trajectory-based equilibrium projection either.

Overall §102 conclusion

• No cited reference anticipates any of claims 1–25, because no single reference discloses all elements of any claim — in particular the core limitation common to independent claims 1, 5, and 20: generating/identifying an equilibrium value of the exhaust-gas constituent reading based on the trajectory of a plurality of readings during sub-threshold fueling.
• The cited art clusters around three themes — (a) sensor fault/deterioration diagnostics (US 6,131,446; US 6,374,818; US 6,976,382; US 7,558,668; US 6,360,583), (b) air/fuel feedback control (US 4,393,840; US 5,179,924; US 5,329,764), and (c) sensor conditioning/regulation (DE 43 41 632 C2; US 2003/0052016) — which map to the dependent claims (2, 6–7, 19, 21–22, 25) and to the specification's background, not to the independent claims' novelty.
• The most useful of the seven for a §103 attack would be US 6,131,446 (uses the sensor's decay-flank shape during overrun as the discriminator — nearest to the "trajectory" idea) combined with a DFSO-calibration reference. As single references for §102, all fail.

One candidate the analyst should consider that is NOT on 8036814's face

The Chinese family sibling CN 101865010 B (Ford) expressly describes the pre-existing practice of periodically updating a NOx-sensor offset by comparing its output to a reference (ambient NOx ≈ 0) during idle, low load, or fuel cut-off, as taught in US 6,311,480. If a §102/§103 challenge to the concept of calibrating a NOx sensor during DFSO is contemplated, US 6,311,480 (not cited on 8036814's face) is the reference to pull. I have not verified US 6,311,480's specification in this session, so treat this as a lead, not a finding. CN 101865010 B PDF

Uncertainty notes (explicit)

1. Three of seven face citations were not retrieved in full (US 4,393,840; US 5,179,924; US 7,558,668) — the tool budget was exhausted mid-verification. My claim-level assessments for those are necessarily shallow and could change on full-text review.
2. US 6,360,583 B1 was confirmed only as a citation in other patents' lists; its own disclosure was not read. It is the highest-priority follow-up because it is Ford's own earlier sensor-monitoring art.
3. §102 vs. §103: I labeled the theme mapping illustratively (e.g., linking degradation references to claims 19/25). A theme match is not an element-by-element §102 mapping; where I say "no anticipation," that is a confident negative given the references' stated disclosures, but confirming positive §102 argument would require full texts of items 1, 2, 4, and 7.
4. Live sources (Google Patents, PubChem, Indiana VDL, FreePatentsOnline, Justia) agree with the authoritative full text on identifiers and dates; no auto-correction was applied.

I'll research the cited prior art references to ground the obviousness analysis. Let me start with the most likely primary references.

Let me research the remaining cited references to complete the prior art picture.

Obviousness Analysis — US 8,036,814 B2 under 35 U.S.C. § 103

Analysis date: 2026-09-29 · Patent: US 8,036,814 B2 ("Calibration scheme for an exhaust gas sensor"), filed 2009-03-23, granted 2011-10-11, Ford Global Technologies LLC.

Scope note. This analysis uses the references that appear in the "Prior Art" / Citations section of the patent's own page (the 7 examiner/third-party citations and the 4 "Family Cites Families" items). It is a § 103 analysis of the art of record, not a full freedom-to-operate or invalidity search. Where I could retrieve full-text content I cite it; where I could confirm only bibliographic data, I say so explicitly rather than inventing disclosure. No identifiers were auto-corrected.

I. The legal framework applied

Pre-AIA § 103 (the application was filed 2009-03-23). The governing question is whether the subject matter as a whole would have been obvious at the time of invention to a person having ordinary skill in the art (PHOSITA), considering (1) scope/content of the prior art, (2) differences between the art and the claims, (3) level of ordinary skill, and (4) secondary considerations. Under KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), a combination is obvious where the elements are known and the combination yields no more than predictable results; a "known technique" applied to a "known problem" is obvious, and a finite number of identified, predictable solutions is a strong signal of obviousness.

Level of ordinary skill assumed: a BSME (or equivalent) plus 2–5 years in automotive powertrain controls / emissions aftertreatment, with working familiarity with exhaust gas sensors (HEGO/UEGO/NOx/NH₃), DFSO control, and control-system signal processing (curve fitting, least-squares, exponential/extrapolation models). Record evidence: the application's own spec frames the "projection model" as a recursive least-squares fit to an exponential decay curve (spec ¶ at 218), i.e., uses textbook mathematics.

II. What the claims actually require — and what the patent admits is known

The three independent claims share one core:

• Claim 1 (system): instructions for, during engine fueling below a threshold amount, identifying an equilibrium value of the exhaust gas constituent reading, where the equilibrium value is based on a trajectory of a plurality of readings during the non-fueling condition; plus instructions to adjust an engine operating condition in response to it.
• Claim 5 (method): during fueling below a threshold with at least one intake and one exhaust valve operating, generating an equilibrium value based on a trajectory of the reading; and under selected combusting conditions adjusting an engine operating parameter.
• Claim 20 (NOx method): generate an equilibrium NOx value based on the trajectory of the NOx reading; adjust a subsequent NOx reading during later combustion.

Critical admission in the patent's own Background (F.1): "One method to calibrate the sensor during its lifetime in a motor vehicle includes measuring gain and offset during a period when the concentration of a desired gas constituent is known, such as deceleration fuel shut off (DFSO) when the exhaust gas may be comprised substantially of ambient air." The inventors then identify the problem: DFSO is too short for the gas at the sensor to reach steady state.

So the only asserted point of novelty over the admitted prior art is the computational step: instead of waiting for (and directly measuring) the steady-state equilibrium value, project that value from the transient trajectory. Everything else — calibrating gain/offset during DFSO, using NOx→0 ppm and O2→20.95% as the known air reference, then using the corrected reading to schedule EGR, air-fuel ratio, urea dosing, etc. — is either admitted as known or is routine engine control.

That framing drives the analysis: the § 103 fight is essentially over whether projecting a signal's asymptote from its decay trajectory was an obvious use of known techniques.

III. The prior art of record — what each reference teaches

First retrieval-confirmed references:

US 6,131,446 A — Robert Bosch GmbH (priority 1997-05-28; granted 2000-10-17) — "Method and arrangement for diagnosing an exhaust-gas probe." This is the closest reference. It monitors the exhaust-gas probe output as the signal changes in response to concentration changes, and applies the rate of change as the criterion for deterioration (Google Patents; via PubChem US-6131446-A). Most pointedly, it evaluates the flank decay time of the rearward probe signal between an upper voltage threshold (SWO) and a lower voltage threshold (SWU) during overrun operation — i.e., during DFSO — where the "very steep drop… characterizes a probe which is in the new state; the flatter traces are typical for probes which are already deteriorated." In other words, Bosch already teaches that the shape/decay trajectory of the sensor signal during fuel cut-off is a repeatable, informative signature of the sensor's calibration state, and that this trajectory is usable to diagnose the sensor (and its output is further used "to adjust the mixture control," an engine operating condition).

US 6,976,382 B2 — Denso Corp. (filed 2003-02-20; granted 2005-12-20) — "Abnormality diagnosing apparatus of exhaust gas sensor." At engine stop (and at a later time), the controller reads the downstream sensor output as "the atmosphere flow[s] back into the exhaust pipe," computes the change in output over the elapsed period, and compares it to an abnormality determinant; a change smaller than the determinant indicates abnormal response (Google Patents; US6976382 PDF). Denso thus teaches reading the sensor's trajectory toward the known ambient-air equilibrium and using that trajectory to evaluate the sensor — the precise physical phenomenon (sensor drifting toward its air-equilibrium value) the '814 patent exploits.

US 6,374,818 B2 — Honda Giken Kogyo (filed 2001-01-26; granted 2002-04-23) — "Apparatus for determining a failure of an oxygen concentration sensor." Operates during a fuel cut-off operation, determines whether the signal crossed a threshold after termination of fuel cut-off, integrates the exhaust gas amount, and declares failure if the signal hasn't crossed when the integrated amount reaches a predetermined value (US20010010220; Indiana VDL text). Honda establishes the DFSO-based, signal-trajectory-based evaluation of an exhaust sensor using the return-to-lean/air behavior as routine.

Bibliographic-only references (titles and dates verified from the patent page; full-text not retrieved, so treat their substantive disclosure as unconfirmed):

• US 6,360,583 B1 — Ford Global Technologies (1998-11-30 → 2002-03-26) — "Oxygen sensor monitoring." Ford's own earlier oxygen-sensor monitoring art; confirms applicant's own field.
• US 5,179,924 A — Hitachi, Ltd. (1990-06-01 → 1993-01-19) — "Method and apparatus for controlling air-fuel ratio in internal combustion engine."
• US 4,393,840 A — Tokyo Kogyo Co., Ltd. (1979-08-31 → 1983-07-19) — "Fuel control system for automobile engine."
• US 7,558,668 B2 — Caterpillar Inc. (2007-11-30 → 2009-07-07) — "Exhaust system having temperature sensor verification." Note: granted after the '814 filing, so it is available only as § 102(e) art as of its 2007-11-30 filing.
• Family citations: US 5,329,764 A (Ford, air/fuel feedback control), DE 4,341,632 C2 (Heraeus Electro Nite, testing/regulating motor vehicles), US 2003/0052016 A1 (Yingjie Lin, controlling oxygen-sensor temperature).

IV. Combination #1 — the primary § 103 case for claims 1, 5, and 20

Proposed combination: US 6,131,446 (Bosch) as the primary reference + US 6,976,382 (Denso) as secondary + the applicant-admitted DFSO calibration practice + the well-known mathematical step of extrapolating an exponential decay to its asymptote.

Element-by-element mapping to claim 5 (representative):

Motivation to combine (KSR rationales):

1. Same field, same problem, same art. Bosch, Denso, Honda, and Ford '583 are all directed to diagnosing/calibrating exhaust gas sensors using the sensor's own signal behavior. Bosch and Denso are both Bosch/Denso-class automotive emissions references, and both expressly exploit the signal's transient trajectory toward a known ambient condition.
2. Bosch supplies the trajectory-during-DFSO teaching; Denso supplies the "trajectory-toward-atmosphere-equilibrium" teaching. Combining them yields exactly the claimed "trajectory → equilibrium value." A PHOSITA optimizing sensor calibration during DFSO would look precisely to the two references that already read the sensor's decay during fuel cut-off.
3. The problem itself was admitted and known. The '814 background concedes the DFSO calibration method and the "too-short DFSO" shortcoming. Where the problem is known and the solution is a known mathematical technique (exponential curve fit / recursive least squares / extrapolation to the asymptote), applying it is obvious (KSR; and the patent's own claim 9 confirms the model is "a least squares statistical best fit to an ideal exponential decay curve out to an equilibrium value at infinite time" — textbook material).
4. Predictable result, finite solutions. The sensor physically relaxes toward a fixed ambient value (0 ppm NOx/NH₃; ~20.95% O₂). Projecting that value from a monotonic decay is a predictable, finite-answer computation, not an unpredictable discovery.

Claim 1 and Claim 20 follow the same logic: claim 1's "identify an equilibrium value based on a trajectory" and instruction to "adjust an engine operating condition" are met by Bosch's DFSO-overrun trajectory evaluation feeding mixture adjustment; claim 20's "equilibrium NOx value based on the trajectory" plus "adjust a subsequent NOx reading" is met by combining the Bosch/Denso trajectory method with the admitted DFSO calibration of a constituent sensor whose air value is known (0 ppm NOx).

V. Combination #2 — an alternative/overlapping path using Honda

US 6,374,818 (Honda) + US 6,131,446 (Bosch) + admitted DFSO calibration. Honda teaches performing sensor failure determination based on the sensor signal's behavior during and after fuel cut-off, using integrated exhaust gas amount as a proxy for whether the sensor should have moved toward the lean/air condition. Bosch adds the rate-of-change/flank-slope dimension during overrun.

Motivation: both references use DFSO as the calibration/diagnostic window because the exhaust approaches a known (lean/air) composition, and the patent's background says this very DFSO window is the known calibration opportunity. Adding the mathematically trivial step of fitting the trajectory and extrapolating to its endpoint produces the claimed equilibrium value. For claim 20 specifically, Honda's fuel-cut-off analysis is directly on point because it concerns the sensor's excursion toward a known reference state during fuel cut-off, which is the calibration event the '814 patent claims.

VI. Coverage of the dependent claims (all obvious over the same combinations)

• Claims 2, 6, 7, 21, 22 (DFSO; value generated after a duration since shut-off): Bosch and Honda operate in overrun/fuel-cut-off; Denso uses an elapsed predetermined time period after which the output is re-read — directly reading "after a duration since fuel shut off."
• Claim 8 / 9 / 23 (projection model = least-squares fit to an exponential decay to infinite time): pure applied mathematics; the spec itself characterizes it as a recursive least-squares fit. No reference is needed to render this obvious once the trajectory-projection step is reached.
• Claims 3 (equilibrium value further based on throttle position, engine speed, valve timing): routine engine operating parameters; the spec says such factors "may be determined by testing a vehicle design." Combining known operating parameters with a curve fit is a predictable design choice (and Bosch/Honda already condition their diagnostics on engine operating parameters such as rpm/load).
• Claims 4, 11, 12, 13, 14 (sensor type: UEGO / NOx / NOx-with-O₂ output / ammonia / any constituent sensor): selection among known exhaust-gas sensor types is obvious; Bosch and Denso address oxygen/gas sensors generally, and NH₃/NOx sensors being calibrated to 0 ppm in air is admitted.
• Claims 15–18 (adjust EGR upward on higher equilibrium value; adjust air-fuel ratio downward on lower value): routine feedback control. US 5,179,924 (Hitachi) is expressly directed to air-fuel ratio control in an IC engine, and US 5,329,764 (Ford) is an air/fuel feedback control system; the directionality (more NOx → more EGR; lower value → richer AFR) is a predictable emissions-control response.
• Claims 19, 25 (indicate sensor degradation via diagnostic code when the equilibrium value crosses a threshold): this is precisely what the cited references do — Bosch flags deterioration from the overrun slope, Honda declares failure, and Denso compares the output change to an "abnormality determinant." The only change is expressing the pass/fail as a threshold on a projected equilibrium value.
• Claim 10 / 24 (fueling < 3% max injection; reduced injection to maintain aftertreatment temperature when requested torque ≤ 0): reduced injection during decel/DFSO for catalyst-temperature maintenance is a well-known strategy; US 2003/0052016 A1 (Yingjie Lin, oxygen-sensor temperature control) and US 4,393,840 (Tokyo Kogyo, fuel control) are cited in the family for related control purposes. (Caveat: I did not retrieve full text of these; treat this sub-point as the least grounded in this analysis.)

VII. The best non-obviousness rebuttals (and why they are weak-to-moderate)

A patentee would argue, and a court would weigh:

1. "Diagnosis ≠ calibration; slope ≠ equilibrium value." Bosch and Denso use the trajectory qualitatively (pass/fail) and never compute an absolute projected equilibrium value to correct a gain/offset. This is the strongest rebuttal: there is a genuine gap between using the slope as a binary diagnostic and extrapolating an absolute asymptote to recalibrate offset. Whether that gap is inventive or merely a predictable application of curve-fitting is the crux.
2. Hindsight. No single reference teaches project-to-asymptote-for-calibration; the combination may be attacked as assembled with knowledge of the claim (a KSR "predictable variation" counter-argument can blunt this given the simple, finite mathematical substitution).
3. Result would be to change the principle of operation (a classic KSR rebuttal) — but here the change is additive (add an extrapolation step), not a change in how the references operate, so this rebuttal is weak.
4. Teaching away — none of the cited art disparages projecting an equilibrium value; absence of teaching-away supports obviousness.
5. Secondary considerations — I found no evidence of unexpected results, industry praise, or nexus-specific commercial success in the record reviewed. The patent lapsed for non-payment (2023-10-11 expiration), which is not itself probative of non-obviousness but indicates the claimed advance did not sustain commercial interest in enforcement.

Net assessment. Independent claims 1, 5, and 20 are, in my view, of moderate-to-high § 103 vulnerability, principally via Bosch US 6,131,446 + Denso US 6,976,382 + the admitted DFSO-calibration practice + the known exponential-extrapolation technique. The single feature that most resists the art of record is the affirmative computation of a projected absolute equilibrium value for calibration (rather than a qualitative trajectory-based diagnostic). Because that step is a known mathematical technique applied to a known problem with a predictable, finite result, the KSR framework points toward obviousness — but a PHOSITA-level dispute over "was it obvious to compute the asymptote for calibration versus merely to look at the slope for diagnosis" is credible and would likely have to be resolved on evidentiary grounds (e.g., a § 103 declaration). The dependent claims are progressively more vulnerable; claims 8/9's least-squares exponential fit and claims 19/25's degradation-code-on-threshold are the most clearly obvious over the cited art.

VIII. Confidence, gaps, and contradictions

• Retrieval-confirmed disclosures: US 6,131,446, US 6,976,382, US 6,374,818 (Google Patents / PubChem / Indiana VDL). These anchor the analysis.
• Bibliographic-only (full text not retrieved): US 6,360,583, US 5,179,924, US 4,393,840, US 7,558,668, US 5,329,764, DE 4,341,632 C2, US 2003/0052016 A1. I have not verified their substantive disclosure; my mappings for Hitachi/Ford '764/Lin on the AFR/EGR and reduced-injection dependent claims are inferential from titles and are flagged as such.
• Prior-art date caveat: US 7,558,668 B2 granted 2009-07-07, after the '814 filing date (2009-03-23); it is available at most as § 102(e) art as of its 2007-11-30 filing. Do not rely on it as § 102(a)/(b) art.
• No contradiction found between the previously generated sections and this analysis; the earlier "Technical gist" (projected equilibrium via exponential/least-squares fit) is the correct characterization and is reused here.
• Legal-status caveat (carried forward): the Google Patents disclaimer that priority/status are non-authoritative applies; the fee-lapse/expiration dates are taken from the listed legal events.
• This is an analysis of the art of record, not an invalidity opinion. A definitive § 103 conclusion would require (a) full text of the four bibliographic-only references, (b) a supplementary search for art expressly teaching projection/extrapolation of a sensor equilibrium value for calibration (e.g., in the NOx-sensor and UEGO literature), and (c) claim-construction of "trajectory," "equilibrium value," and "engine fueling below a threshold amount."