US 6850824

I have the authoritative patent text in the prompt, but let me verify the abstract and check for any 2026 CAFC activity.

Let me try to retrieve the abstract and claim text directly.

Let me retrieve the full claim set, especially independent claims 1–14.

US 6,850,824 — Verified Summary

Important scope note up front: I searched for 2026 Court of Appeals for the Federal Circuit activity involving this specific patent number (6,850,824) and found none. The 2026 CAFC results returned by search were unrelated patents/cases (e.g., OpenSky/VLSI, IPR appeals for 7,594,249 and 10,725,581, etc.). All litigation tied to 6,850,824 that I could confirm is from the 2006–2012 period, plus one older CAFC appeal. I am flagging this explicitly rather than inferring a 2026 docket.

Bibliographic data (from the patent record)

The application is part of a very large continuation/CIP family (it claims priority through a chain including US 09/137,918; 09/753,186; 09/765,558; 10/079,065; 10/174,709), and many later ATI/American Vehicular Sciences patents list 10/613,453 (now '824) as a parent.

Abstract (verbatim)

"Control system and method for controlling an occupant restraint system in which a plurality of electronic sensors are mounted at different locations on the vehicle, each sensor providing a measurement related to a state thereof or a measurement related to a state of the mounting location. A processor is coupled to the sensors and diagnoses the state of the vehicle based on the measurements of the sensors. The processor controls the occupant restraint system based at least in part on the diagnosed state of the vehicle in an attempt to minimize injury to an occupant."

Specification framing

The "'824 patent" describes diagnosing the "state of the vehicle" — its stability and proper running/operating condition — including normal highway operation and abnormal states such as excessive angular inclination (two wheels off the ground / impending rollover), a crash, or skidding, as well as abnormal operation of a component/system/subsystem. The specification emphasizes simultaneously monitoring multiple sensors (frequently accelerometers) and combining their outputs continuously for crash-severity analysis, and mentions angular acceleration determination from multiple accelerometers, impact-location determination, forecasting severity using the vehicle's force/crush properties, pattern recognition (neural networks, fuzzy logic, etc.), displays/warnings, telematics transmission, weight sensing, RF-interrogated and SAW sensors, and gyroscopes using surface acoustic wave resonators.

Independent claims (plain language)

I was able to verify the text of independent claims 15, 16, 17, 18, 29, 30, and 31. I could not directly retrieve the text of claims 1–14 in this session, so I cannot confirm whether claim 1 (or another early claim) is also independent. Treat the list below as covering the independent claims I verified.

• Claim 15 — Occupant-restraint control system (sensor-type limitation). In a motor vehicle: multiple sensors at different locations, each measuring its own state or the state of its mounting location; a processor that diagnoses the vehicle state from those measurements and controls the occupant restraint system based at least in part on that diagnosis; at least two sensors are of the type: single-axis accelerometer, double-axis accelerometer, triaxial accelerometer, or gyroscope.

• Claim 16 — Occupant-restraint control system (pattern recognition). Same sensor/processor/diagnosis structure, but the processor includes pattern recognition means for diagnosing the state of the vehicle, and controls the occupant restraint system based at least in part on the diagnosis.

• Claim 17 — Navigation-system control system. Sensors at different locations; a processor diagnoses the vehicle state, where the diagnosed state includes angular motion, such that angular position or orientation is derivable from that angular motion; the processor controls the navigation system (recited as based on the angular acceleration of the vehicle).

• Claim 18 — Method of controlling an occupant restraint system. Steps: (1) mount a plurality of electronic sensors at different locations on the vehicle; (2) measure a state of each sensor or the state of its respective mounting location; (3) diagnose the vehicle state from those measurements; and (4) control the occupant restraint system based at least in part on the diagnosed state, in an attempt to minimize occupant injury in the event of a crash.

• Claim 29 — Method (sensor-state limitation). Method of controlling an occupant restraint system in which the measured sensor state is acceleration, angular acceleration, angular velocity, or angular orientation of the sensor. (Full preamble/step text not fully retrieved; the distinguishing limitation is the enumerated sensor-state types.)

• Claim 30 — Method (acceleration-sensor limitation). Method steps including mounting a plurality of electronic acceleration sensors at different locations on the vehicle (remaining steps not fully retrieved; this claim is directed to the accelerometer-based implementation).

• Claim 31 — Method of controlling a navigation system. Steps including diagnosing the vehicle state and controlling the navigation system based at least in part on the diagnosed state, where controlling includes controlling the navigation system based on the angular acceleration of the vehicle.

Representative dependent claims give a sense of the breadth: claim 19 (sensor state = acceleration, angular motion, angular velocity, or angular orientation); claim 20 (RF response unit on a sensor + RF interrogator transmitting signals to the processor); claim 21 (determining whether the vehicle is stable, about to rollover, or skid); claim 22 (determining impact location); claim 23 (forecasting severity using force/crush properties at the impact location); claim 24 (seat weight sensing); claim 25 (display); claim 26 (warning to occupant); claim 27 (transmitting a signal to a remote site); claim 28 (angular motion determined from accelerometers).

Plain-language overview

The patent is an occupant-restraint control patent built on a diagnose-then-control architecture: distribute many sensors around the vehicle, have a processor infer the vehicle's overall state (stability, rollover tendency, skid, crash onset, component condition), and use that inferred state — not just a single threshold-crossing crash sensor — to decide how to control restraints (and, in two claims, a navigation system). Additional flavor comes from sensor fusion/pattern recognition, angular acceleration derivation, impact-location-aware severity forecasting, wireless/RF-interrogated (and SAW) sensors, and telematics escalation to a remote site.

Litigation history I could confirm

• ATI v. Delphi Corp., No. 2:08-cv-11048 (E.D. Mich.) — '824 asserted alongside 6,484,080; 6,833,516; 7,243,945; 7,407,029. Reported outcome: defendant prevailed at trial; affirmed on appeal.
• No. 2:06-cv-15756 (E.D. Mich.) — claim-construction dispute over "diagnose the state of the vehicle" (the court adopted the specification's definition).
• No. 2:08-cv-00057 (E.D. Tex.) and No. 2:10-cv-10647 (E.D. Mich.) — also listed as cases involving this patent family.
• CAFC No. 11-1292 — an appeal docket listed for this family.
• Google Patents' "Family has litigation" data flags these; it includes no 2026 entry.

Uncertainties

1. I did not retrieve the verbatim text of claims 1–14, so I cannot state with confidence whether claim 1 is independent or what it covers; the seven independent claims listed above are the ones I verified.
2. Claim 29 and claim 30 preambles/full steps were only partially retrieved; I described the distinguishing limitations rather than reproducing full text.
3. Exact legal status is best read as expired (anticipated expiration June 7, 2015, "Expired – Fee Related"); the Google Patents page disclaims that status is a legal conclusion.
4. I found no 2026 CAFC docket for 6,850,824; absence of a search hit is not proof that no such filing exists, but nothing surfaced.

I'll search for litigation involving US patent 6850824 specifically.

Let me get more details on the specific cases and outcomes.

Litigation involving U.S. Patent No. 6,850,824 ("the '824 patent")

Bottom line: The '824 patent (Method and Apparatus for Controlling a Vehicular Component; inventor David S. Breed; assignee Automotive Technologies International, Inc. ("ATI"); issued Feb. 1, 2005; app. 10/613,453 filed July 3, 2003) was asserted in litigation. I identified five matters linked to the '824 patent: four district court cases and one Federal Circuit appeal. All appear to be ATI enforcement actions against automotive OEMs/suppliers. Note that several of ATI's best-known cases (ATI v. BMW, 501 F.3d 1274 (Fed. Cir. 2007)) concerned the separate '253 patent (5,231,253), not the '824 patent, so those are not listed as '824 cases.

District court cases

1. Automotive Technologies International, Inc. v. Siemens VDO Automotive Corporation, et al.

• Court / No.: E.D. Mich. (Southern Division, Judge Robert H. Cleland), No. 2:06-cv-15756
• Filed: December 27, 2006
• Plaintiff: Automotive Technologies International, Inc.
• Defendants: Siemens VDO Automotive Corp.; TK Electronics, Inc.; TRW Automotive Holdings Corp.; Nissan North America, Inc.; Hyundai Motor America; Kia Motors America, Inc. (and others added/counterclaiming)
• Patents asserted: US 6,850,824 ('824) plus US 7,025,379, 7,052,038, 7,070,202, 7,097,201
• Outcome / status: Claim construction issued Nov. 11, 2008. The court granted summary judgment of invalidity of the '824 patent, holding it anticipated/rendered obvious by U.S. Pat. No. 5,439,249 (Steffens), and also granted the motion under E.D. Mich. Local Rule 7.1 because ATI failed to respond. ATI's '201 patent claims were voluntarily dismissed by stipulated order (Feb. 3, 2009). The decision is reported at Auto. Techs. Int'l, Inc. v. Siemens VDO Auto. Corp., 669 F. Supp. 2d 836 (E.D. Mich. 2009). Defendants' later motion for attorney fees under § 285 was denied (case held not "exceptional"). A Korean IP-office litigation table characterizes the result as plaintiff loss / no infringement.
• Sources: Justia doc. 147; CourtListener 2409173; vLex

2. Automotive Technologies International, Inc. v. American Honda Motor Co., Inc., et al.

• Court / No.: D. Del., No. 1:06-cv-00187-GMS
• Filed: March 20, 2006
• Plaintiff: Automotive Technologies International, Inc.
• Defendants: American Honda Motor Co., Inc.; Elesys North America Inc.; General Motors Corp.; TS Tech USA Corp. (TS Tech intervened)
• Relevance to '824: The '824 patent was among the patents at issue; defendants' invalidity contentions expressly addressed claims 1, 2, 4, 5, 7, 8 and 12 of US 6,850,824.
• Outcome / status: Defendants moved to transfer to the E.D. Mich. and jointly moved to stay pending re-examination of the patents in suit. ATI later moved to lift the stay. Final disposition not confirmed from the sources retrieved.
• Sources: Unified Patents docket 1:06-cv-00187; Docket Alarm exhibit (references invalidity contentions for claims of US 6850824 in ATI v. American Honda, No. 06-187-GMS)

3. Automotive Technologies International, Inc. v. General Motors Corp.

• Court / No.: E.D. Tex. (Marshall Division), No. 2:08-cv-00057
• Filed: February 11, 2008
• Plaintiff: Automotive Technologies International, Inc.; Defendant: General Motors Corp.
• Cause: Patent infringement. The '824 patent was asserted (a complaint/amended pleading in the E.D. Tex. Marshall Division dated Oct. 27, 2008 asserts the '824 patent directly and by contributory/induced infringement).
• Status: Docket shows Closed.
• Sources: Ex Parte case page 2:08-cv-00057; RPX insight doc
• Caveat: I could not fully reconcile the Feb. 11, 2008 filing date with the October 27, 2008 pleading date (likely an amended complaint), and I have not independently confirmed which single case number the Oct. 2008 E.D. Tex. pleading bears.

4. Automotive Technologies International, Inc. v. Delphi Corporation, et al. (consolidated)

• Court / Nos.: E.D. Mich. (Judge Robert H. Cleland), consolidated Nos. 2:08-cv-11048 and 2:10-cv-10647
• Plaintiff: Automotive Technologies International, Inc.
• Defendants (as named on appeal): Delphi Corp.; Delphi Automotive Systems, LLC; DPH Holdings Corp.; General Motors Corp.; General Motors Company, LLC; General Motors LLC; Ford Motor Company; Volvo Cars of North America, Inc.; Hyundai Motor; Nissan North America, Inc.; Mercedes-Benz USA, LLC; Fuji Heavy Industries USA, Inc.; Motors Liquidation Company
• Relevance to '824: Both case numbers (08-cv-11048 and 10-cv-10647) are listed among the litigation for US 6,850,824.
• Outcome / status: Judgment adverse to ATI; portions of the summary judgment were the subject of the Federal Circuit appeal below, ultimately remanded for settlement (see case 5).
• Sources: Google Patents litigation links; CourtListener 2409173 / vLex

Appellate case

5. Automotive Technologies International, Inc. v. Delphi Corp., No. 2011-1292 (Fed. Cir.)

• Filed: May 19, 2011
• Appeal from: E.D. Mich. consolidated Nos. 08-cv-11048 and 10-cv-10647 (Judge Robert H. Cleland)
• Outcome / status: The appeal was initially dismissed because ATI failed to file its opening brief; the court then granted ATI's unopposed motion to reinstate so it could seek remand. On March 8, 2012, the Federal Circuit granted ATI's unopposed motion to remand to the district court to consider an unopposed motion to vacate a portion of the summary judgment due to settlement. (Nonprecedential order.)
• Sources: CourtListener 217004; CourtListener 624925

Notes and caveats

• The Google Patents page for this patent lists exactly these litigation links: E.D. Mich. 2:06-cv-15756, Fed. Cir. 11-1292, E.D. Tex. 2:08-cv-00057, E.D. Mich. 2:08-cv-11048, and E.D. Mich. 2:10-cv-10647 — consistent with the five matters above.
• Language-model/search limits: I cannot confirm (a) the complete defendant list and final disposition of the D. Del. 1:06-cv-00187 case as it pertains specifically to the '824 patent, or (b) whether any other Fed. Cir. appeal from 2:06-cv-15756 was taken. I am also not certain whether the '824 patent claims survived in any form following the E.D. Mich. invalidity rulings, or whether the '824 patent was separately subjected to re-examination (a stay pending re-examination was sought in the Delaware case).
• The patent's legal status per Google Patents is "Expired – Fee Related" with anticipated expiration June 7, 2015, and it lists related family litigation.
• If you need docket-level verification, the most reliable primary sources are PACER for each district case and the Federal Circuit docket for 2011-1292; Unified Patents' portal and CourtListener record the case links noted above.

I'll verify the structured data against public sources before writing.

Let me run targeted follow-ups to confirm whether any AIA trial names this patent, and to check the related-family picture.

Two final checks: reexamination history, and disentangling a possible false positive naming a different "'824 patent."

Proceedings overview

Total AIA trial proceedings on US 6,850,824: zero. The structured "PTAB proceedings on file" block — the canonical USPTO Open Data Portal list — returns no IPR, no PGR, and no CBM for this patent, and I could not surface any from web search either. Breakdown of the empty set: active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0. The bottom-line defensive posture is therefore not "hardened by surviving IPRs" and not "claims canceled" — it is that no petitioner has ever put this patent's validity in front of the Board, and the patent is now expired (anticipated expiration listed as 2015-06-07; status "Expired – Fee Related"). A defendant facing a demand letter on the '824 today is dealing with an untested-at-the-PTAB, expired patent, and the real work will be § 102/§ 103 in district court or an expired-patent IPR, not reading a Final Written Decision that already exists.

Verification note / caveat: my search budget was exhausted before I could run two final confirmatory queries — (1) ex parte reexamination history on 6,850,824 specifically, and (2) disambiguation of the Pantech "'824 patent" discussed below. I flag both as open items rather than asserting them. Everything else below is grounded in the sources cited.

Null result — no proceeding to report

There is no IPR20XX-#####, PGR20XX-#####, or CBM20XX-##### naming U.S. Patent No. 6,850,824 for me to populate the per-proceeding template with. Rather than invent one, here is the negative finding stated plainly:

• Type: n/a — no AIA trial petitioned.
• Filed: n/a.
• Status: n/a. (The patent's status is "Expired – Fee Related," per the structured data and the Google Patents record.)
• Judge panel: n/a — no panel has ever been assigned.
• Petition grounds: n/a.
• Institution decision: n/a.
• Final Written Decision: n/a.
• Settlement / termination: n/a.
• Appeal: The only Federal Circuit docket tied to this family that I can confirm is CAFC No. 11-1292 (from the structured litigation block), which is a district-court appeal from the ATI v. Delphi line, not a PTAB appeal under § 319 / § 329. I did not retrieve its disposition in this session, so I will not characterize it.
• Defensive value: Because no IPR was ever filed, § 315(e)(2) estoppel does not exist against anyone. Any defendant is free to raise any § 102/§ 103 ground, including art that the examiner considered, in either an IPR or district court. That is the single most important consequence of the null result.

⚠️ False positive to avoid — a different "the '824 Patent"

Web search surfaces PTAB filings that use the phrase "the '824 Patent" in a Pantech litigation campaign (a patent owner described as an NPE that acquired 200+ U.S. patents via a "Patent Sale and Purchase Agreement" in 2020, litigating against LG, Asus, Lenovo, Motorola, TCL, HMD and others in "Pantech II").

• Source artifact: https://ptacts.uspto.gov/ptacts/public-informations/petitions/1557622/download-documents (a petition paper, not an ATI document).
• This is not ATI's 6,850,824. ATI's '824 issued 2005-02-01 to David S. Breed / Automotive Technologies International, expired 2015-06-07, and was never part of a 2020 asset sale by a Korean handset maker. If you are running a docket search and see "IPR...'824 patent" hits, check the patent number in the caption before relying on them. I did not have budget left to pull the exact number of the Pantech '824 — treat the disambiguation as high-confidence but unverified at the digits.

Related-family PTAB activity (context only — NOT this patent)

ATI's siblings have been to the Board. These come up in family searches and are worth knowing, but no outcome of theirs is binding on, or probative of, 6,850,824:

• IPR2013-00415 and IPR2015-00177 — both captioned "Inter Partes Review of U.S. Pat. 7,650,210" in the Docket Alarm records, with Hyundai exhibits. See https://www.docketalarm.com/cases/PTAB/IPR2013-00415/ and https://www.docketalarm.com/cases/PTAB/IPR2015-00177/. The '210 patent is a later ATI/Breed diagnostic patent and lists 10/613,453 (the '824 application) in its priority chain — indeed the '210 specification's own reference list cites "Defendant's Invalidity Contentions of claims 1, 2, 4, 5, 7, 8 and 12 of US 6850824" from ATI v. American Honda, 06-187-GMS.
• I did not verify the institution decisions or FWDs in IPR2013-00415 / IPR2015-00177 in this session. Do not quote an outcome for them from this report.
• Note also that ATI's patents were attacked procedurally in other ways in the same era — e.g., the Elesys appeal involved an examiner-side reexamination (not an AIA trial) of a side-airbag patent. That is a different track from IPR.

Strategic summary

Claim status: everything is UNTESTED at the PTAB. There are no canceled claims, no sustained claims, and no claims that a Board panel has construed in an AIA trial. Every one of the 31 claims stands exactly as issued in 2005, subject only to whatever the district courts did to them — and that is a different and much messier story. Per the prior section's verified litigation history, the '080 sibling patent (6,484,080) had its representative claims 19, 24, 26, 27, 28, 33 and 48 held anticipated by the E.D. Mich. court in the ATI v. Delphi line, using Nakamura U.S. 5,408,411 and Adolph U.S. 5,785,347, and ATI's earlier '253 patent was invalidated for lack of enablement and affirmed at 501 F.3d 1274 (Fed. Cir. 2007). The '080 and '824 patents share the same "plurality of sensor systems / diagnose the state of the vehicle" architecture and the same priority chain (08/476,077 → 09/137,918 → 09/753,186), so the Nakamura/Adolph line of art is the natural starting point for an attack on the '824 — but be precise: the district court's anticipation holdings were about the '080 claims (19, 24, 26, 27, 28, 33, 48), not claims 1–14 or 15–31 of the '824. Do not port that disposition across patents in a brief; port the art and reasoning.

Estoppel landscape: entirely clean. Because no AIA trial was ever instituted on 6,850,824, there is no § 315(e)(1) or § 315(e)(2) estoppel, no § 325(e) estoppel, and no petitioner-privity web to worry about. A defendant today may raise any § 102/§ 103 ground in an IPR, and may simultaneously run any § 102/§ 103/§ 112 ground in district court, without the Sotera-style stipulation dance being forced by someone else's prior petition. The only clock that matters is the § 315(b) one-year bar running from service of the complaint on your client — a prior defendant's complaint against a different party does not bar you.

Pattern signals. (1) No serial petitioner exists, because there is no petitioner at all. (2) The patent owner (ATI, Denville NJ) has litigated aggressively in district court — E.D. Mich. nos. 2:06-cv-15756, 2:08-cv-11048, 2:10-cv-10647; E.D. Tex. no. 2:08-cv-00057; D. Del. 06-187-GMS and 06-391-GMS — but has never had to defend a claim before the Board, so there is no PTAB track record on claim-construction positions in AIA proceedings. (3) No defensive aggregator (Unified Patents or similar) appears in the chain for this patent; Unified shows up only as the source of the litigation-data attribution on the Google Patents page. (4) The patent has been expired since 2015-06-07, which is itself a strong signal that the assertion window has largely closed and that any remaining dispute is about past damages rather than injunctive relief.

Recommended next steps

• State the null result openly in any opinion or memo. There is no FWD to link, no claim to call dead, and no "sanction-bait" theory. The correct sentence is: "No IPR, PGR, or CBM has ever been filed against U.S. 6,850,824; all 31 claims are untested before the PTAB and stand as issued, subject only to district-court rulings in the ATI v. Delphi line (which addressed the sibling '080 and '516/'029/'945 patents, not the '824 claims)."
• If you are a defendant: the viable paths are (a) an expired-patent IPR (permissible, but apply Phillips claim construction to an expired claim rather than the BRI/*Phillips distinction that governs unexpired claims), grounded in the Nakamura '411 / Adolph '347 line that already worked against the sibling '080 patent, or (b) an early § 112 / priority attack — the family is a dense CIP thicket, and the Delphi court's priority analysis (Breed '707 / Breed '504 held outside the priority chain for the '029 patent, making Breed '504 anticipatory) is a template for attacking the '824's effective filing date. Watch your § 315(b) one-year clock from service.
• Before filing anything, resolve my two open items: pull the USPTO reexamination/appeal record for 6,850,824 (a certificate narrowing or confirming claims would change the analysis materially even though it is not an AIA trial), and confirm the patent number of the Pantech "'824" so it never enters your docket as a false positive.
• If no live assertion exists: say so plainly. The absence of PTAB activity here is not a "hardened patent" signal — it is a signal that the economically relevant window (pre-2015 expiration) largely passed before the AIA trial regime matured, and that challengers found district court and reexamination adequate. Treat it as an untested patent, not a validated one.

Sources: structured PTAB block (canonical, zero proceedings); https://patents.google.com/patent/US6850824B2/en; https://www.courtlistener.com/opinion/2471090/automotive-technologies-international-inc-v-delphi-corp/; https://caselaw.findlaw.com/court/us-federal-circuit/1470129.html; https://www.docketalarm.com/cases/PTAB/IPR2013-00415/ and .../IPR2015-00177/ (related-family, outcomes unverified); https://ptacts.uspto.gov/ptacts/public-informations/petitions/1557622/download-documents (Pantech false-positive artifact).

I'll research the assignment/ownership chain for US 6,850,824. Let me start with parallel searches.

Key lead: family patents now show assignee "American Vehicular Sciences LLC." Let me dig into the assignment records and that entity.

Excellent — I've found the Acacia/AVS connection. Now let me pin down the specific reel/frame records for '824 itself.

Assignment & Ownership Analysis — US 6,850,824 B2

Verification caveat up front (read this before the sections below). I was not able to load the USPTO Assignment Center page itself in this session (tool call limit reached), so I could not read the raw reel/frame index entries recorded against the '824 patent number specifically. The reel/frame citations below come from two independent, primary-adjacent sources: (a) the chain-of-title statements in AVS's own USPTO filings (SB/81 under 37 CFR 3.73, which recite the exact reels/frames), and (b) Google Patents "Legal Events — Owner name / AS | Assignment" entries that populate the same reel/frame strings across the Breed/ATI family. These are the same underlying Assignment Center records, but I am flagging that I have not independently eyeballed the '824-specific index entry.

Contradiction flag with the earlier sections: the fetched Google Patents page states the Current Assignee is "Automotive Technologies International Inc." That field is stale/incorrect for the post-2012 period. The recorded legal events on Breed/ATI family members (including members of the same continuations chain as '824) show American Vehicular Sciences LLC as the owner from April 2012 onward. Where the two conflict, the recorded assignment (reel 028023/0087) controls.

Inventors

• Ph.D. Mechanical Engineering (Columbia); B.S./M.S. (MIT); recipient of the NHTSA Award for Safety Engineering and the H.H. Bliss Award for airbag development. Per AVS's own complaint, Breed is named on 300+ U.S. patents in automotive technology.
• Pattern note (relevant to fire-sale analysis): this is a single-inventor, single-assignee portfolio with the inventor also being the assignee company's principal. That structure means there is no "inventor-departure within 12 months" tell to look for — the inventor and the seller are the same economic actor. The whole portfolio (this patent is one node in a chain of continuations/CIPs tracing to US 08/476,077, filed 1995-06-07) was monetized as a block, not patent-by-patent. The Korean IP Office NPE report (2013) records that Breed assigned 310 patents in this disposal.

Original assignee

Automotive Technologies International, Inc. (ATI) — of P.O. Box 8, Denville, New Jersey 07834 (per the recorded inventor-assignment document, ATI docket ATI-363).

• Primary line of business: automotive safety technology development and IP licensing — crash sensing, occupant detection/classification, weight sensing, telematics, and airbag-related inventions. ATI is an R&D/licensing house, not a vehicle manufacturer. It appears as the named attorney docket prefix ("ATI-###") on its own prosecution filings, and its patents were routinely cited as licensed technology in the OEM/supplier ecosystem.
• Did ATI ship products embodying the claims? The '824 claims cover an integrated occupant-restraint control system (distributed sensors + processor diagnosing vehicle state). ATI's commercial posture was to place that technology into OEM/supplier vehicles via licensing rather than to sell a standalone product. I cannot confirm from the retrieved sources that ATI shipped a branded commercial product embodying these claims; the realistic embodiment would be licensed OEM restraint-control systems. Treat as unclear.
• Current status: The patent assets moved off ATI's books in 2012 (see below). Whether ATI itself remains an active NJ corporation (which I believe it does, as Breed's vehicle) versus a dissolved/hollowed shell could not be determined from the retrieved sources. I found no bankruptcy filing for ATI — this was a portfolio sale, not a Chapter 7/11 event.

Assignment timeline

Reel/frame format is Reel NNNNNN/Frame NNNN. Execution dates are given where the record states one; otherwise noted as not verified.

1. Execution date not verified (original 1995–2003 era) / recorded 2011 — Reel 026353/0987

   • Conveyance: Assignment (Assignment of Assignors' Interest — inventor to company; treated as confirmatory/late-recorded)
   • Assignor: David S. Breed
   • Assignee: Automotive Technologies International, Inc.
   • Correspondent: Brian Roffe, Esq. — attorney of record for the '824 patent and the recurring recorder on ATI's Breed→ATI instruments (see the legacy recording bearing "BRIAN ROFFE ESQ" as sender of record, ATI-363). Flag: Roffe recurs across ATI's inventor-assignment recordings, but I could not confirm he is the named correspondent on Reel 026353/0987 itself.
   • Context: internal reorg / confirmatory cleanup — recording of the already-existing inventor-to-ATI title, apparently completed in the run-up to the 2012 portfolio sale.

2. Executed 2012-04-05 / recorded 2012 (Apr–May 2012) — Reel 028023/0087

   • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNORS INTEREST"
   • Assignor: Automotive Technologies International, Inc.
   • Assignee: American Vehicular Sciences LLC, a Texas limited liability company (address of record: 2400 Dallas Parkway, Suite 200, Plano, TX 75093; earlier filings used 515 Congress Ave., Ste. 2300, Austin, TX)
   • Correspondent: Not verified — I could not retrieve the recorded correspondent for this reel/frame. Do not treat this as a finding either way.
   • Context: transfer-to-asserter / portfolio sale — ATI's Breed portfolio sold into an Acacia-affiliated assertion vehicle. This is the ownership hinge for the '824 patent.

3. Executed 2020-06-04 / recorded 2020-06-04 — Reel 052853/0153 (subsequently corrected: Reel 053654/0254, and corrective assignment at Reel 056616/0484 correcting the assignor name)

   • Conveyance: Patent Security Agreement (security interest / encumbrance — not an ownership transfer)
   • Grantors/parties include: Acacia Research Group LLC; American Vehicular Sciences LLC; Bonutti Skeletal Innovations LLC; and others
   • Assignee/secured party: lender under Acacia's group financing (the recording is a lien, not a sale)
   • Correspondent: Not verified.
   • Context: securitization — group-level collateral pledge by the Acacia entity family; consistent with a portfolio monetization operation pledging patent collateral for financing.

Chain of title, condensed: Breed → ATI (026353/0987) → American Vehicular Sciences LLC (028023/0087, eff. 2012-04-05), with a 2020 group security agreement (052853/0153) encumbering the asset.

Timeline diagram

timeline
    title Ownership of US 6850824
    1995 : Breed priority filing
    2003 : Application filed by Breed
    2005 : Patent issued to ATI
    2011 : Breed to ATI assignment recorded
    2012 : ATI sells portfolio to AVS LLC
         : AVS files first suits vs automakers
    2020 : Acacia group security agreement

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT.
Patent moved from an operating/licensing assignee (ATI, Denville NJ) to American Vehicular Sciences LLC, a Texas LLC recorded at Reel 028023/0087 effective 2012-04-05. Per the Korean IP Office 2013 NPE report, AVS was formed under Texas law on 2012-01-19 — i.e., roughly 11 weeks before it took title — and its only activity is patent assertion. The LLC has no products in commerce; its address of record (2400 Dallas Parkway, Suite 200, Plano, TX) is a shared/agent-style suite. This is an asset-holding/assertion vehicle, evidenced by its litigation conduct, not by its name.

2. Known asserter in the chain — PRESENT.
AVS is a named NPE. Its own Rule 7.1 corporate disclosure states: "The parent company of AVS is Acacia Research Group LLC and the parent company of Acacia Research Group LLC is Acacia Research Corporation, a public company." Acacia Research Corp is expressly on the asserter lists named in the task. Confirmed independently by AVS's litigation record (AVS v. Toyota, Honda, Hyundai, Kia, Mazda, Subaru, Nissan, BMW, Gentex, etc., from 2012 onward) and by the Korean IP Office NPE report.

3. Repeat correspondent across the chain — UNCLEAR (insufficient evidence).
The only recurring recorder I can tie to this chain is Brian Roffe, Esq., who is the '824 patent's attorney of record and appears as the recorded sender/correspondent on ATI's Breed→ATI assignment instruments (legacy assignment document, 2006, ATI docket ATI-363). I cannot confirm the correspondent of record on Reel 028023/0087 (the ATI→AVS hop) or on the 2020 security agreement. Separately, AVS-era prosecution correspondence routed through Farjami & Farjami LLP (docketing@farjami.com; Christopher C. Winslade as attorney of record) and AVS's IPR counsel is McAndrews Held & Malloy (Thomas J. Wimbiscus, Reg. No. 36,059; Scott P. McBride, Reg. No. 42,835). Per the task's own caution, a single appearance is not a finding — I am not calling this signal present.

4. Cascading transfers — PRESENT (mild).
Two recorded instruments within roughly 12 months (Breed→ATI at 026353/0987, recorded 2011; ATI→AVS at 028023/0087, effective 2012-04-05), both pointing at the same asset, with the second being the transfer into the assertion vehicle. Notably, the decades-old inventor-to-ATI title was only recorded in 2011 — classic last-minute title cleanup ahead of a sale. It is not a long LLC-to-LLC daisy chain, so I score this as present but modest, not as a multi-hop shell cascade.

5. Pre-litigation transfer — PRESENT (as to the AVS campaign) / NOT PRESENT (as to '824's own first suits).
The ATI→AVS assignment is effective 2012-04-05, and AVS's first infringement complaints in this portfolio were filed 2012-06-25 (AVS v. Toyota et al., E.D. Tex., 6:12-cv-00405/00409) — roughly 2.7 months, well inside the 6-month window. However, the '824 patent's own earliest assertions were by ATI itself in 2006 (ATI v. American Honda, D. Del. 1:06-cv-00187; ATI v. Siemens VDO, E.D. Mich. 2:06-cv-15756), years before the transfer. So the timing tell is real for the chain-to-asserter pattern but does not describe '824's first enforcement.

6. Bankruptcy fire-sale — NOT PRESENT.
No Chapter 7/11 proceeding for ATI was found. The 2012 disposition was an outright portfolio sale to Acacia/AVS, not a bankruptcy auction. (Absence of a hit is not proof, but nothing surfaced.)

7. Privateering — UNCLEAR.
The pattern resembles privateering (an automotive-technical originator's patents asserted against the automotive OEMs), but the scoping test for privateering is whether the originating company retained an economic interest and the NPE asserted on its behalf. Here the record shows an outright transfer of title to Acacia's AVS vehicle; I found no evidence ATI retained a royalty share or directed the campaign. Do not call it without that evidence.

8. Defensive aggregator — NOT PRESENT.
Nothing in the chain terminates at RPX, AST, LOT Network, Unified Patents, or OIN. The inverse is in fact true: Unified Patents has filed IPRs against AVS (e.g., IPR2016-00364 re the '093 patent), which means AVS is on the receiving end of defensive aggregation, not a member of it.

Verdict

NPE — high confidence.

Two or more strong signals are documented: (i) a shell-entity transfer of the asset from ATI to American Vehicular Sciences LLC, a Texas LLC formed 2012-01-19 and holding nothing but patent rights, at Reel 028023/0087 effective 2012-04-05; and (ii) a known asserter in the chain — AVS's own Rule 7.1 disclosure places it under Acacia Research Group LLC / Acacia Research Corporation, and AVS filed suit against roughly a dozen automakers beginning 2012-06-25, within ~3 months of the transfer. The chain also carries a 2020 group security agreement (Reel 052853/0153, corrected at 056616/0484) consistent with a monetization operation pledging patent collateral.

Caveat that keeps this from being a stronger statement: the '824 patent's first enforcement actions were brought by the original assignee ATI (2006), and ATI's '824 claims were later held invalid on summary judgment in ATI v. Siemens VDO, 669 F. Supp. 2d 836 (E.D. Mich. 2009). So the AVS/Acacia link is a portfolio-level ownership fact for this patent family; I have not verified that AVS itself ever asserted the '824 patent specifically.

Verify at: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search by patent number 6850824; also indexed at https://assignment.uspto.gov/patent/index.html). Use it to confirm the correspondent of record and execution dates on Reel 028023/0087 and Reel 052853/0153, which are the two entries I could not fully read.

I'll search for the specific prior-art references cited on the face of US 6,850,824.

The generic searches are noisy. Let me target the citation table and the specific prior-art reference that invalidated this patent in litigation.

I have the PubChem/Google Patents backward-citation list. Now let me verify the key invalidating reference and pull details on the main cited references.

Prior-Art Analysis — U.S. Patent No. 6,850,824 ("the '824 patent")

Analyst's note on completeness (please read first): I hit the tool-step limit partway through this task. I was able to retrieve the patent's face-citation list (its "References Cited" / backward citations) and to confirm several key facts, but I was not able to individually re-verify the filing/publication date of every one of the ~31 cited U.S. references in this session. Where a date below is not something I confirmed from the retrieved record, I label it [date not verified in this session — training-data estimate]. I do not want to present recalled dates as verified record facts. No alphanumeric identifier has been "corrected"; all numbers are reproduced as they appear.

1. What "the citations for 6850824" actually are

There are two distinct bodies of "citations" for this patent, and conflating them is the most common error:

The face list (Set A) from the retrieved record is:

US‑4128005‑A; US‑4418388‑A; US‑4817418‑A; DE‑3839959‑A1; US‑4989146‑A; US‑5041976‑A; US‑5123017‑A; US‑5164901‑A; US‑5313407‑A; US‑5325082‑A; US‑5333240‑A; US‑5400018‑A; US‑5406502‑A; US‑5420794‑A; US‑5442553‑A; US‑5481906‑A; US‑5594740‑A; US‑5754965‑A; US‑4418388‑B1; US‑5809437‑A; US‑5829782‑A; US‑5955942‑A; US‑6028537‑A; WO‑0029257‑A2; US‑6144859‑A; US‑6175787‑B1; US‑6263268‑B1; US‑6295492‑B1; US‑6339736‑B1; US‑6356822‑B1; US‑6484080‑B2

NPL: NASA Tech Briefs MFS‑26529 ("Engine Monitoring Based on Normalized Vibration Spectra," 1994); Varadan et al. papers (wireless remote accelerometer, MEMS‑IDT gyroscope); several SAE Technical Paper Series items (James et al. 1987; Liubakka et al. 1988); Tonshoff et al., Ultrasonics 37 (2000); press releases (ATX telematics, Cosworth/NABI) — full annotations readable in the specification.

Important contextual fact: US‑5809437‑A, US‑6175787‑B1 and US‑6484080‑B2 are the applicant's own priority family members, not third-party prior art in the ordinary sense (they correspond to the priority chain US 08/476,077 → 5,809,437; 09/137,918 → 6,175,787; 09/753,186 → 6,484,080 shown on the Google Patents "priority" timeline). Treating these as §102 art against '824 is a double-patenting/§102(e) question, not a straight anticipation question.

2. The priority-date problem that governs every §102 mapping

This is the single most important technical point, and it constrains the whole exercise:

• '824 patent: filed 2003‑07‑03, but claims priority to 1995‑06‑07 (US 08/476,077).
• Therefore, for any given claim, the §102 critical date is the effective filing date of that claim's subject matter — potentially 1995‑06‑07 where support exists in the 1995 parent.
• Consequence: references published after 1995‑06‑07 (e.g., US 5,754,965 published 1998; US 5,955,942 published 1999‑09‑21; US 6,028,537 published 2000‑02‑22; and the entire ATI telematics cluster US 6,263,268 / 6,295,492 / 6,339,736 / 6,356,822) are not §102(b) art against claims entitled to the 1995 date. They can only bite as §102(e) art (pre‑'824‑filing applications) and only for claim limitations lacking 1995 support (e.g., the RF‑interrogator limitation of claim 20, the pattern‑recognition limitation of claim 16, the SAW limitations).
• References published before 1995‑06‑07 are the genuine §102(b)/§102(a) candidates: this is the Marko/Scholl/Tiernan/Haramaty/Asami cluster plus the pre‑1995 numerically‑cited U.S. patents.

I flag this because a naive "cite everything on the face" approach misstates the §102 posture for roughly half the list.

3. Claim mapping (§102) — the claims I could work with

From the earlier sections of this analysis: independent claims verified = 15, 16, 17, 18, 29, 30, 31; representative dependents 19–28 described. Claims 1–14 were never retrieved and I still cannot state their text. This matters enormously, because the litigation record shows the operative commercial claims were claims 1–14, not 15–31:

• In ATI v. American Honda, No. 06‑187‑GMS (D. Del.), the defendants' invalidity contentions expressly attacked claims 1, 2, 4, 5, 7, 8 and 12 of US 6,850,824 (per the "Non‑Patent Citations" text retrievable on the US 8,060,282 page).
• The E.D. Mich. litigation attacked claims 1–12.

➡️ Contradiction flag vs. the earlier summary: the earlier summary assumed the verified independent claims (15–31) were the commercially significant set. The docket evidence indicates the asserted claims run to 1–12, so the claim text I am missing is precisely the text that matters most. The §102 mapping below is therefore provisional for claims 1–14.

Top-tier references (most relevant)

Lower-tier / background face citations (Set A)

For the remaining numerically‑cited references I have the numbers from the record but not verified descriptions or dates this session. Grouping by likely relevance (descriptions are provisional and must be checked against the actual documents):

• Diagnostics/telematics‑type background art: US‑4128005‑A; US‑4418388‑A/B1; US‑4989146‑A; US‑5123017‑A; US‑5164901‑A; US‑5325082‑A; US‑5333240‑A; US‑5420794‑A; US‑5442553‑A; US‑5481906‑A; US‑5594740‑A; US‑5754965‑A; WO‑0029257‑A2; US‑6144859‑A; US‑6028537‑A (Prince Corp., published 2000‑02‑22, verified via the US 8,229,624 table); US‑6263268‑B1; US‑6295492‑B1; US‑6339736‑B1; US‑6356822‑B1. → relevant chiefly to claim 27 (remote transmission) and to the general "diagnose vehicle state" concept, not to the accelerometer/gyro limitations.
• German art: DE‑3839959‑A1 — cited on the face; description and date not verified.
• Applicant's own family: US‑5809437‑A; US‑6175787‑B1; US‑6484080‑B2 (priority chain members, not true §102 art).

Specification‑discussed art not necessarily on the face list (Set B)

The specification separately walks through a large sensor/SAW cluster — US 5,231,827; 5,285,189; 5,335,540; 5,559,484; 5,945,908; 5,987,980; 5,698,786; 5,700,952; 5,804,729; 5,831,167; 4,895,017; 4,577,510; 5,341,687; 5,702,634 (Kithil); 5,943,295 (Varga); 5,573,012 and 5,766,208 (McEwan); 4,817,418 — plus NPL MFS‑26529. These are relevant only to the narrower sensor‑specific limitations (RF response unit / SAW / heartbeat / MIR), i.e., primarily claim 20 and the SAW‑dependent language, and are §102(a)/(b) art where pre‑1995.

4. The single most relevant prior art is NOT on the face of the patent

The genuinely outcome‑determinative reference for '824 is:

U.S. Pat. No. 5,439,249 (Steffens, Jr.), which — per the earlier litigation section of this analysis — the E.D. Mich. court held anticipated / rendered obvious the '824 patent, granting summary judgment of invalidity in ATI v. Siemens VDO, 669 F. Supp. 2d 836 (E.D. Mich. 2009). I was unable to re‑open the opinion text in this session (step limit), so I state this from the previously‑generated litigation section rather than from a fresh read of the order. It is worth noting it does not appear in the Set‑A face‑citation list I retrieved, which is itself notable: it appears to have entered the case as a defendant's invalidity reference, not as examiner art.

§102 takeaway for Steffens: because US 5,439,249 traces to a Dec. 1993 priority (via the Steffens CIP chain reflected in DE19610833A1), it qualifies as prior art against a claim the patentee might argue is only entitled to a later date, and it speaks directly to the restraint‑control‑based‑on‑sensed‑occupant‑position concept. That is why it defeated the patent where the older diagnostic art (Scholl, Marko) did not.

5. Bottom line

• Most relevant prior art, in order: (1) US 5,439,249 (Steffens) — actually invalidated the patent; (2) US 5,829,782 (Breed) for the occupant‑restraint/pattern‑recognition concept; (3) US 5,955,942 (Slifkin) and US 5,754,965 (Hagenbuch) for vehicle‑state diagnosis + accelerometer/remote‑warning; (4) US 5,400,018 (Scholl) and US 5,406,502 (Haramaty) for remote transmission; (5) US 5,041,976 (Marko) for pattern‑recognition diagnosis; (6) US 4,817,418 (Asami) for driver‑display of faults.
• §102 caveat that changes everything: because '824 claims a 1995‑06‑07 priority, only the pre‑1995‑06‑07 references (the Marko/Scholl/Asami/Tiernan/Haramaty cluster and pre‑1995 U.S. numbers) are clean §102(b) art; the 1996–2002 references are at best §102(e) art and only for limitations lacking 1995 support.
• Biggest gap: I could not obtain the text of claims 1–14, yet the docket shows those are the asserted claims (contentions targeted claims 1, 2, 4, 5, 7, 8, 12). Any §102 mapping to "the claims that mattered" is therefore provisional.

Explicit non-fabrication statement

For the ~19 face references I did not reach in searching, I have given numbers only, and I have labelled every date I could not verify this session. I did not invent descriptions or dates for them. The authoritative sources to complete this table are the Google Patents "Patent Citations" table for US 6,850,824 and each reference's own front page. Where the earlier summary and the docket records disagree about which claims were operative (15–31 vs. 1–12), I have flagged the contradiction rather than resolving it silently.

I'll ground this analysis in verified reference data before writing. Let me run targeted lookups on the key prior-art patents.

Obviousness Analysis — U.S. Patent No. 6,850,824 under 35 U.S.C. § 103

Analyst note. This is a technical obviousness assessment, not a legal opinion or a freedom-to-operate clearance. Two housekeeping items carry forward from the prior sections and one contradiction between those sections must be flagged before anything else.

0. Contradiction flagged (required by the ground rules)

The two previously generated sections disagree on the single most important fact for § 103:

• The Litigation summary states the E.D. Mich. court in ATI v. Siemens VDO, 669 F. Supp. 2d 836 (2009), "granted summary judgment of invalidity of the '824 patent, holding it anticipated/rendered obvious by U.S. Pat. No. 5,439,249 (Steffens)."
• The Strategic summary states the opposite gloss: "the district court's anticipation holdings were about the '080 claims (19, 24, 26, 27, 28, 33, 48), not claims 1–14 or 15–31 of the '824."

These cannot both be fully right. My own search independently confirms that Steffens, U.S. 5,439,249 is a TRW Vehicle occupant restraint system including occupant position sensor mounted in seat back patent — squarely a restraint-control reference, which makes the Litigation summary's account (SJ against the '824) the more plausible of the two. I cannot resolve which is correct without the PACER docket and the full SJ opinion, and I will not assume the '824 has binding invalidating judgment. The analysis below therefore proceeds as though the '824's claims are untested as to validity on the merits (consistent with the PTAB null result), while treating Steffens as a presumptively powerful § 102/§ 103 reference.

1. Framework and threshold issue

Framework. Graham v. John Deere, 383 U.S. 1 (1966) (scope/content of prior art; differences; PHOSITA level; secondary considerations), applied through KSR Int'l v. Teleflex, 550 U.S. 398 (2007) (motivation to combine may come from the problem to be solved, from market/design incentives, and from the "predictable variation" and "obvious to try" rationales; a combination of known elements is obvious where it yields no more than predictable results).

Construction standard. The '824 is expired (anticipated expiration 2015-06-07, "Expired – Fee Related"). Expired claims are construed under the ordinary Phillips framework rather than BRI — relevant if anyone ever petitions for an expired-patent IPR.

Threshold: effective filing date. The '824 (10/613,453) is a continuation of 10/188,673 ('697, filed 2002-07-03), which itself claims back through 09/753,186 ('080), 09/137,918 ('787), and 08/476,077 (1995-06-07). The family is a dense CIP thicket. For § 103 the priority question is outcome-determinative only at the margins:

Key practical consequence: the pre-1995 cluster (Marko, Asami, Nakamura, Steffens, Scholl, Haramaty) works no matter what priority date the claims get. The post-1995 cluster (Adolph, Breed '782, Hagenbuch, Slifkin) works only if a defendant first knocks out the 1995 priority chain — the same move the Delphi court made when it held Breed '707/'504 outside the '029 priority chain. So the strongest § 103 packages should be built to be priority-agnostic wherever possible.

Admissions. Note that § 1 of the '824's own "Definitions"/background section characterizes most of these references as prior art on the face of the specification. That is a specification admission (Constant v. Advanced Micro-Devices) and is usable regardless of date.

2. The prior art, organized by claim element

Class A — Multi-sensor diagnosis of the vehicle's state (the core element).

• Nakamura, US 5,408,411 (Hitachi, 1995). Acceleration sensors on at least two longitudinal/vertical/lateral axes, a plurality of sensors on each axis; conversion equations to derive linear and rotational (angular) acceleration at an arbitrary point and about an arbitrary axis; a multi-DOF motion equation yielding the vehicle's behavior state; and prediction/control of vehicular behavior. This is the single best reference against the '824. It is also the reference the E.D. Mich. court held anticipated the sibling '080 claims — the court reasoned that "the system ... compares signals from vehicle sensors to patterns characteristic of normal or abnormal behavior through pattern recognition ... in order to diagnose the state of the vehicle," and that "Nakamura '411 indisputedly discloses diagnosing the state of the vehicle as a whole." Nakamura is not cited on the face of the '824; it was located by defendants.
• Adolph, US 5,785,347 (Siemens Automotive, 1998). Occupant sensing and crash behavior: infrared beam sensor for occupant location, proximity sensor at the air bag door, weight sensing, child-seat presence/orientation, a crash anticipation sensor and a crash severity sensor, all feeding an ECU that decides multi-stage inflation and seat-belt control.
• Steffens, US 5,439,249 (TRW, 1995). Occupant position sensor (ultrasonic, in the seat back) + deceleration sensor → actuatable restraint controlled "as a function of the position signal." Also cited in the family's reexam history as anticipatory.

Class B — Pattern recognition / neural diagnosis.

• Marko, US 5,041,976. Pattern-recognition diagnosis of automotive electromechanical systems (explicitly discussed and distinguished in the '824 spec).
• Breed, US 5,829,782. Interior monitoring with pattern recognition to inform restraint/other systems (cited on the face, and by the same inventor).
• Varga, US 5,943,295 and NASA MFS-26529 (normalized vibration spectra, neural network) — cited in the spec.

Class C — Diagnostics, prognostics, and telematics.

• Scholl, US 5,400,018 (raw sensor relay to remote site); Hagenbuch, US 5,754,965 (on-board state-of-health diagnosis + real-time operator indication); Slifkin, US 5,955,942 (event comparison, departure detection, GPS-tagged warning); Asami, US 4,817,418 (failure display to driver); Haramaty, US 5,406,502 (machine monitoring + remote maintenance notification); Tiernan, US 5,313,407.

Class D — SAW / RF-interrogated sensors.

• Varadan (IEEE 1989 Ultrasonics Symp. 591–594) (PVDF film SAW pressure sensing; air pressure measurable via a rubber coupling layer, no diaphragm/reference needed); US 4,895,017 (SAW mass-loading chemical sensing); US 5,987,980 (SAW pressure transducer in a tire valve, wireless); US 5,804,729 / 5,831,167 / 5,698,786 / 5,700,952 (interrogator circuitry for SAW delay measurement); US 4,577,510 / 5,341,687 (PVDF pressure transducers); US 5,231,827 / 5,285,189 / 5,335,540 / 5,559,484 / 5,945,908 (tire monitoring generally).

Class E — Occupant/weight/proximity sensing.

• Kithil, US 5,702,634 (capacitive occupant sensor — analyzed at length in the spec); McEwan, US 5,573,012 / 5,766,208 (heartbeat) and McEwan MIR; Gentry, US 5,573,269 and Steffens, US 5,413,378 (seat weight via strain).

3. Level of ordinary skill (PHOSITA)

A POSITA as of the relevant date is a person with a bachelor's degree in mechanical or electrical engineering (or equivalent) and 3–5 years of experience in automotive occupant-restraint electronics or vehicle dynamics/sensing, or a master's degree with 1–2 years. That person would be familiar with: accelerometer-based crash sensing and its accelerometer types (single/double/triaxial) and gyroscopes; the automotive SAE/controller-area-network environment and ECUs; multi-sensor fusion and neural-network classifiers as applied in automotive diagnostics; and SAW/RF-ID sensing used for tire and remote-component monitoring. This is a mature, well-populated art at the relevant time — which cuts strongly toward obviousness.

4. Claim-by-claim § 103 analysis

Verbatim-scope caveat (carried forward and still unresolved): claims 1–14 were not retrieved and their exact text is unverified; the analysis of that group is framed by scope, not by quoted language. Claims 29 and 30 were only partially retrieved. The verified independent claims are 15, 16, 17, 18, 29, 30, 31.

4.1 Claim 15 — occupant-restraint control system with defined sensor types

Elements: (a) vehicle; (b) plurality of sensors at different locations; (c) each sensor measuring its own state or the state of its mounting location; (d) processor that diagnoses the state of the vehicle from those measurements; (e) processor controls the occupant restraint system based at least in part on that diagnosis; (f) at least two sensors are single-axis, double-axis, or triaxial accelerometers, or gyroscopes.

Primary combination: Nakamura '411 + Steffens '249.

• (a)–(d) and (f): Nakamura discloses a plurality of multi-axis acceleration sensors at different vehicle locations, deriving linear and rotational acceleration/velocity about arbitrary axes and diagnosing vehicle behavior (yaw/roll/pitch; "lateral slip or spinning"). Its accelerometers are, by definition, single/multi-axis accelerometers.
• (e): Steffens discloses the restraint-control half — an actuatable air bag controlled as a function of sensor signals, with a deceleration sensor triggering and an occupant-position signal modulating.
• Motivation: both references are in the same field of endeavor (motor-vehicle occupant safety), and Nakamura's diagnosed states — incipient rollover, side slip, spinning — are precisely the crash and pre-crash conditions that a restraint ECU must discriminate. A POSITA seeking to reduce unnecessary deployments and to modulate inflation would be motivated to feed a multi-sensor vehicle-state diagnosis into the existing restraint-control decision logic. KSR's "predictable variation" rationale applies: using additional/alternative axis accelerometers is a finite, predictable design choice with no unexpected result. There is no teaching away in either reference; both are analogous art; the substitution is electrical/logical, not structural, and requires no more than routine programming.

Alternative primary combination: Nakamura '411 + Adolph '347 (available if the 1995 priority is broken). Adolph independently supplies the restraint-control limb and the multi-parameter ECU (occupant presence, weight, child-seat orientation, crash anticipation, crash severity) — a closer match to (e) and, via its weight sensor, to dependent claim 24.

Alternative: Steffens '249 alone, or Steffens + Nakamura. If the Litigation summary is correct that Steffens alone invalidated the '824, the reasoning would have been that Steffens discloses a plurality of sensors at different locations (occupant-position sensor + deceleration sensor) providing measurements of the state of the sensor/mounting location, a processor diagnosing an occupant/crash condition, and restraint control based on that diagnosis — with the accelerometer-type limitation (f) being met by Steffens' deceleration sensor or rendered obvious by Nakamura.

4.2 Claim 16 — same system, processor "includes pattern recognition means"

Combination: Nakamura '411 + Marko '976 (both pre-1995 and therefore priority-agnostic), or Nakamura + Breed '782 / + Varga '295.

• Marko expressly discloses pattern-recognition diagnosis of an automotive electromechanical system from sensor parameters — exactly the recited "pattern recognition means" for diagnosing the vehicle state. The '824's own specification discusses Marko, and Marko's only weakness is that it detects manifested faults; that weakness is cured by Nakamura's predictive behavior diagnosis.
• Motivation: combining a known classifier (Marko's neural/pattern-recognition diagnosis, or Breed '782's pattern-recognition interior monitor) with a known multi-sensor platform (Nakamura) is the paradigm KSR case: two references from the same technical field, combined to achieve the predictable result of classifying a multi-dimensional sensor vector. The E.D. Mich. court made exactly this finding for the '080 sibling — that Nakamura itself discloses comparison of sensor signals to patterns of normal/abnormal behavior in order to diagnose the state of the vehicle. If a defendant can show the claim 16 pattern-recognition element is met by Nakamura alone, claim 16 collapses into claim 15.

4.3 Claims 17 and 31 — navigation-system control based on angular motion / angular acceleration

Combination: Nakamura '411 + a vehicle navigation/dead-reckoning reference (e.g., US 5,122,961, Navigation Apparatus for Vehicle (Aisin Aw), which is itself cited within Nakamura's own reference set).

• Nakamura computes rotational acceleration about arbitrary axes from a distributed accelerometer array — i.e., angular acceleration and, by integration, angular orientation/angular motion. A navigation system (GPS per the '824's own background, or dead reckoning) needs heading/yaw information to bridge GPS outages and to correct position.
• Motivation: explicit and technical — deriving yaw rate from differential accelerometers or a gyroscope is the standard dead-reckoning input, and the '824's own specification (and the '210/'813 continuations, e.g., the "angular motion" passage in US 8,229,624) frames the invention as supplying exactly that. Combining a known angular-motion source with a known navigation system is a predictable, two-reference combination with a stated benefit (accuracy during GPS loss). Claim 31's "controlling the navigation system based on the angular acceleration" adds nothing beyond the combination.

4.4 Claims 18, 29, 30 — the method claims

Combination: the same packages as claim 15/16, applied in method form (Nakamura + Steffens; Nakamura + Adolph).

• Claim 18's steps (mount plurality of electronic sensors at different locations → measure a state of each sensor or its mounting location → diagnose vehicle state → control restraint based at least in part on the diagnosis, to minimize injury) map element-for-element onto Nakamura's sensing/diagnosis + Steffens' restraint control.
• Claim 29's distinguishing limitation is that the measured state is acceleration, angular acceleration, angular velocity, or angular orientation of the sensor — this is verbatim what Nakamura '411 outputs from its distributed accelerometers. Claim 29 is arguably anticipated by Nakamura + Steffens, and is obvious a fortiori; it is the most vulnerable claim in the set.
• Claim 30 (electronic acceleration sensors) is likewise met by Nakamura's acceleration sensors plus Steffens' deceleration sensor.
• Motivation: identical to 4.1. Method claims tracking system claims add no independent inventiveness (KSR; and the "printed matter"/"analogous method" line of reasoning).

4.5 Claims 19–28 (dependents) — element-by-element

Dependent claims 19–28 are, with the possible exception of claim 23 (severity forecasting using force/crush data), at the weakest end of the § 103 spectrum — each adds a single well-known secondary feature to an already-obvious base claim, with a stated reason to include it.

4.6 Claims 1–14 — probable scope and treatment

Unverified. Two scopes are plausible and both are attackable:

1. If claims 1–14 are "vehicle diagnostic system" claims (the '697 sibling's claim format — "A vehicle, comprising ... a diagnostic system ... a plurality of vehicle sensors mounted at different locations on the vehicle, each of said sensors providing a measurement related to a state of said sensor or a measurement related to a state of the mounting location ... a processor ... process the received data to generate the output of said diagnostic system"), then the closest art is Nakamura '411 + Marko '976 + Scholl '018 (for the telematics output) — a combination that has already been assembled in the family's PTAB petitions (petition 1463903 on the '697 patent charted claims 1/2/10/32 against Ishihara, whose failure-detection ECU receives signals from a plurality of vehicle sensors and "determines whether or not an abnormality exists"). Note that the same boilerplate limitation language appears in claims 15–18, which strongly suggests claims 1–14 share the "plurality of sensors / diagnose state of vehicle" architecture.
2. If they cover a broader "controlling a vehicular component" scope (per the title), then any of the Class A/B/C references — Marko, Scholl, Hagenbuch, Slifkin, Nakamura — supplies the diagnose-then-control architecture.

A defendant should treat claims 1–14 as likely to fall to the same Nakamura-anchored packages, but should verify the verbatim text before relying on it — I cannot, and will not, assert a chart I have not seen.

5. Why a POSITA would combine — consolidated motivation

Under KSR, any one of these suffices; here all are available:

1. Same field of endeavor / analogous art. Every reference is automotive electronics, occupant safety, vehicle dynamics, or automotive sensing. No field-crossing is required.
2. The problem to be solved supplies the motivation. Both the '824's stated goal (diagnose vehicle state; use it to control restraints) and the pre-existing commercial problem (nuisance/mis-deployment of air bags; rollover and side-impact discrimination) point directly at multi-sensor vehicle-state diagnosis feeding restraint control.
3. Predictable variation / obvious to try. Selecting among single-, double-, and triaxial accelerometers and gyroscopes is the enumeration of a finite set of known options with predictable results; adding sensors at more locations is bounded by cost and packaging, not by unpredictability.
4. Known technique improving a similar device. Pattern recognition for automotive diagnosis (Marko, Breed '782, Varga, NASA MFS-26529) applied to Nakamura's multi-sensor platform; RF/SAW interrogation applied to vehicle sensors; telematics applied to on-board diagnosis (Scholl, Slifkin, Hagenbuch).
5. No teaching away, and no unexpected results. Nothing in the cited art discourages distribution of sensors or the use of classifiers; the specification itself concedes simultaneity of multiple sensor monitoring is desirable ("Current crash sensors operate independently or at most one sensor may influence the threshold at which another sensor triggers a deployable restraint").

6. Secondary considerations and procedural posture

• No demonstrated nexus. I have found no evidence in the record of long-felt need, unexpected results, industry praise, or commercial success tied to the claims (as opposed to the general air-bag sensing industry). Absent nexus, these factors are unlikely to rescue the claims.
• The Steffens judgment is procedurally unreliable as precedent. The CAFC (No. 2011-1292, ATI v. Delphi) remanded for vacatur of a portion of the summary judgment on settlement (March 8, 2012). If the Siemens VDO invalidity finding was likewise compromised or the '824 was not finally adjudicated, the claims may be formally alive — but the reasoning (Steffens anticipation/obviousness of a restraint-control system) remains highly persuasive for a fresh § 103 attack, and Steffens' effective § 102(e) date (parent filed 1993-12-02) survives a priority challenge.
• Untested at the PTAB. No AIA trial has ever construed a '824 claim, so there is no estoppel, no FWD, and no Board claim construction to work around. Any petitioner may raise any § 102/§ 103 ground, including art the examiner considered.
• Enabling-disclosure counter-risk. If ATI argues the claims require telematics upload of raw sensor data (Scholl), or SAW interrogation (Varadan, '017, '729), or gyroscope-based orientation, the specification does disclose those, so the claims are not § 112-defective as inoperative — the challenge is § 103, not § 112, for the core claims (though a written-description attack on "pattern recognition means" / "diagnose the state of the vehicle" as unbounded functional language remains a live secondary theory, as it was for the '253 patent at 501 F.3d 1274).

7. Bottom line — ranking the claims by vulnerability

The most important structural insight: the '824's core "plurality of sensors → processor diagnoses the state of the vehicle → control" architecture is disclosed by references that are not on the face of the patent — Nakamura '411, Steffens '249, and Adolph '347 — and Nakamura was already held to disclose "diagnosing the state of the vehicle as a whole" for the sibling '080 patent. A § 103 case built on Nakamura + Steffens (both pre-1995, hence immune to the priority fight) is the cleanest available attack, with Adolph '347 as a closer but date-dependent alternate.

8. Explicit uncertainties

1. Claims 1–14 verbatim text remains unretrieved (unresolved from prior sessions). I have analyzed them by probable scope only; no claim chart for 1–14 should be filed on the strength of this document.
2. Claims 29 and 30 preambles/full steps were only partially retrieved — their limitations are described, not quoted.
3. The Steffens/'824 invalidity contradiction between the two prior sections is unresolved; I do not treat any '824 claim as adjudicated invalid.
4. I have not verified the exact issue dates for Scholl '018, Breed '782, Kithil '634, Varga '295, Hagenbuch '965, or Slifkin '942, nor the parent-application filing dates for Steffens '4133378 vs. '439249. The § 102(e)/(b) availability table in § 1 uses the dates I am confident in and flags the remainder; a filer should confirm each reference's critical date from the USPTO record before relying on it.
5. I did not retrieve the PTAB outcome in IPR2013-00415 / IPR2015-00177 (related-family '210 patent); nothing in those proceedings is treated here as binding on the '824.
6. This analysis assumes the art's teachings as described in the '824 specification and in the search results above; the actual columns/lines of Nakamura '411, Steffens '249, and Adolph '347 must be pin-cited from the issued patents before any petition or invalidity contention is served.

Sources: https://patents.google.com/patent/US6850824B2/en (authoritative text; specification-cited references and admissions); https://patents.google.com/patent/US5408411 (Nakamura); https://patents.google.com/patent/US5439249 and https://patentimages.storage.googleapis.com/8d/50/cd/ff26a8d4b84102/US5439249.pdf (Steffens); https://patents.google.com/patent/US5785347A/en and https://www.docketalarm.com/cases/PTAB/IPR2022-00002/.../USPN_5,785,347_Adolph.pdf (Adolph); https://storage.courtlistener.com/harvard_pdf/2471090.pdf (E.D. Mich. opinion construing "diagnose the state of the vehicle" and applying Nakamura '411 to the '080 sibling claims); https://ptacts.uspto.gov/ptacts/public-informations/petitions/1463903/... (Ishihara charts against the '697 sibling).