US 6108594

I'll research US patent 6108594 across USPTO sources and Federal Circuit dockets.

Now let me check for any litigation or Federal Circuit activity involving this patent.

US Patent 6,108,594 — Summary

Important scoping note up front: I searched for Federal Circuit activity and litigation involving this exact number and found no CAFC 2026 docket, appeal, or district-court litigation tied to US 6,108,594. Searches for "6108594 + Federal Circuit," "6108594 + CAFC 2026," and "6,108,594 + litigation" returned only unrelated 2026 CAFC opinions (e.g., Ideahub v. Unified Patents, 2024-1684; a Google/Sonos hotword appeal, 24-2119) and generic litigation-trend reports — nothing naming this patent. Treat "no litigation found" as the current state, with the caveat that I cannot query PACER/CAFC dockets directly and absence of search hits is not proof of absence.

Bibliographic data

Abstract (as issued)

"A method and apparatus for autonomous acquisition of attitude in a stellar inertial spacecraft attitude system is disclosed. The present invention uses star trackers, an on-board star catalog, spacecraft steering and inertial sensors to determine spacecraft attitude. The present invention utilizes pattern match and pattern rejection methods and uses multiple stellar snap-shots in conjunction with spacecraft steering and spacecraft inertial measurements to acquire spacecraft attitude. Spacecraft inertial measurements are used to connect multiple stellar snap-shots to provide adequate star information that can be used to acquire spacecraft attitude. In an attitude determination system using star trackers, the star trackers may have a narrow field-of-view or few stars may be available for viewing. The present invention uses pattern matching and pattern rejection on different sets of stars, thereby allowing attitude acquisition when the number of stars in view is small."

Plain-language overview of the independent claims

There are two independent claims — claim 1 (method) and claim 7 (apparatus/spacecraft).

Claim 1 — Method of autonomous acquisition of spacecraft attitude (steps, in order):

1. Use a star tracker to observe a first set of stars at a first instant in time.
2. Form a set of candidate star triplets (three-star combinations) from that first set.
3. Pattern-match each candidate triplet against an on-board star catalog (OSC) to produce a set of pattern matches.
4. Form a set containing a plurality of spacecraft attitudes based on those matches (i.e., don't assume a single answer — keep the ambiguity).
5. Apply a rejection method to that candidate set to eliminate some of the candidate attitudes.

The core inventive idea: rather than requiring enough stars in a single snapshot to resolve attitude uniquely, the system deliberately retains multiple candidate attitudes and then prunes them over time and across successive star observations. The rejection alternatives are recited in dependent claims 2 and 3 (see below).

Claim 7 — Spacecraft (apparatus):

• A star tracker for observing stars;
• an inertial sensor for detecting body rates;
• an on-board star catalog containing star position and brightness information;
• a spacecraft control processor receiving attitude-related position information from the star tracker and inertial sensor, programmed to perform the same five-step sequence as claim 1 (observe first star set → form candidate triplets → pattern match to the OSC → form a plurality of candidate attitudes → run a rejection method to eliminate candidates).

The apparatus claim mirrors the method with the added structural elements (star tracker, inertial sensor, OSC, and a processor programmed to run the algorithm).

Dependent claims (context for the rejection step)

• Claims 2 / 8: rejection by new-sky consistency — observe a second set of stars and discard any candidate attitude for which too few of the newly observed stars match the OSC.
• Claims 3 / 9: rejection by temporal/gyro consistency — stamp each candidate with a propagated attitude and time, compute the difference between past attitude measurements and the propagated attitude, threshold it, build a table of "consistent/inconsistent" entries, find the largest string of consistent measurements, and repeat until that string reaches a critical length.
• Claims 4–6 / 10–12: add inertial-sensor (gyroscope) calibration — observe a star, check whether it was measured before, retrieve the prior reference measurements, and compute a bias estimate from the observed star position and its prior measurements. (Claims 6 and 12 refer to a "gryoscope" — that spelling appears verbatim in the issued document; I am reproducing it literally, not correcting it.)

Supporting detail worth noting

• The specification describes the rejection logic with Equation 1: ‖(Ĉ₁ᵀ·Ĉ₂)ᵀ(C₁ᵀ·C₂) − I₃ₓ₃‖ < Threshold, where C terms are candidate attitudes and "C-hat" terms are gyro-propagated attitudes. Table 1 in the patent illustrates a 4-candidate consistency matrix (entries "C"/"I") in which C₁ survives as the longest consistent string.
• Embodiments include an extended-Kalman-filter attitude processing function, steering commands that deliberately slew the spacecraft so new sky enters the tracker FOV (effectively widening the FOV), and star-tracker output combining circuitry.
• Priority art cited on the face includes U.S. Pat. Nos. 4,621,329 (Jacob), 4,944,587 (Harigae), 4,658,361 (Kosaka), 5,177,686 (Boinghoff), 5,745,869 (van Bezooijen), and 5,935,195 (Quinz) — 15 U.S. patent documents total.
• Patent family ID: 22205011 (single-member family; no foreign counterparts listed).

Confidence and caveats

• High confidence on title, dates, application number, inventors, original assignee, abstract, claim text, and classification — these are corroborated across Google Patents, Justia, EveryPatent, and the USPTO-face-data PDF (US6108594.pdf).
• Medium confidence on the current assignee (AT&T MVPD Group LLC): this comes from Google Patents' automated listing, which it flags as unverified. The Hughes → DirecTV → AT&T corporate lineage would make it plausible, but I did not confirm a recorded assignment.
• Low confidence / negative finding on CAFC 2026 docket activity: nothing was found, but I could not exhaustively query the CAFC docket system. If you need a definitive litigation check, that would require a PACER/Docket Navigator query rather than open web search.
• Because the patent expired May 29, 2018, any enforcement litigation in 2026 would concern pre-expiration damages only, which further reduces the likelihood of active 2026 CAFC briefing.

I'll search for litigation records involving US Patent 6108594 across patent litigation sources.

Let me search more specific litigation sources and databases.

Litigation Search Report — US Patent No. 6,108,594

Bottom line

No litigation involving US Patent 6,108,594 was found. Despite targeted searches of patent-litigation aggregators, PTAB/CAFC materials, and general web sources, I could not identify any district court case, Section 337 (ITC) action, PTAB proceeding, or Federal Circuit appeal in which US 6,108,594 was asserted, challenged, or otherwise at issue.

I want to be explicit about the limits of this finding: it is based on web-accessible sources reached via search, not a direct, authoritative query of PACER, Docket Navigator, or Lex Machina. A negative result here is evidence of no litigation, not proof.

The patent's identity (confirmed literally, not auto-corrected)

The expiration is significant: any infringement action would have had to be filed on or before the 2018 expiration date to seek prospective relief, and damages would be limited to the six-year lookback from filing.

What I searched and what came back

1. Direct patent number queries — US patent 6108594 litigation, "6,108,594" patent infringement complaint, patent 6108594 "attitude acquisition" litigation district court. These returned patent-profile pages (e.g., a Google Patents-family listing and inventor aggregator pages), not any case docket. No complaint, answer, or judgment referenced this number.

2. Unified Patents portal references — Searches returned Unified Patents PTAB case-list material, but none of the named proceedings involved US 6,108,594. Unified's case lists that surfaced concerned unrelated patents (e.g., 5,964,658; 10,311,073; 8,767,824; 10,621,228) appearing in other members' litigation contexts.

3. Assignee-oriented queries — "6108594" SpaceX OR DirecTV OR Hughes patent case. Returned only the inventor/profile pages, no litigation.

4. CAFC / court documents — A CAFC Rule 36 judgment surfaced (IdeaHub v. Unified Patents, 2024-1684, PGR2022-00044), but it involves a different patent entirely and is not related to US 6,108,594.

Note on the cited-by/similar-documents lists on the Google Patents page (34 forward citations; 34 "cited by" entries such as US 6,289,268, US 6,357,615, US 6,691,033, US 6,760,227, etc.). These are subsequent patents that cite 6,108,594 as prior art — they are not litigation. Do not conflate them with infringement suits.

Why a "no litigation" result is plausible for this patent

• It is a narrow, spacecraft-guidance-specific patent (stellar inertial attitude acquisition using star trackers, gyros, and pattern matching/rejection). The pool of potential infringers is small and largely limited to satellite bus/attitude-determination suppliers and operators.
• It expired in 2018 with no evidence of reissue, reexamination, or continuation that would have extended assertion power.
• The current assignee attribution to AT&T MVPD Group LLC appears to reflect a corporate-lineage/assignment database artifact (Hughes → DirecTV/AT&T corporate family) rather than a record of enforcement activity.

Recommended next steps to confirm definitively

To convert this "not found" into a verified negative, run:

1. PACER Case Locator (pacer.uscourts.gov) — nationwide party and patent-number search. This is the authoritative source for district court filings.
2. USPTO Patent Center / Assignment Search for application 09/087,409 — check for recorded assignments or security interests that would signal monetization.
3. USPTO PTAB API / Patent Trial & Appeal Board — confirm no IPR/PGR/CBM involving 6,108,594.
4. USPTO Patent Litigation Docket (the "Patent Litigation" dataset posted by USPTO) — bulk searchable by patent number.
5. Docket Navigator or Lex Machina — commercial databases that index by asserted patent number; these will resolve any gap left by the free web sources used here.

Caveat

If you need this for a legal proceeding (e.g., an invalidity, freedom-to-operate, or damages analysis), the web-based negative above is not a substitute for a certified PACER/Lex Machina search. I did not have direct access to PACER or Docket Navigator in this session, so a truly complete litigation check remains outstanding.

Proceedings overview

Total AIA trial proceedings on file for US 6,108,594: 0. No IPRs, PGRs, or CBMs have been instituted, denied, or terminated on this patent — the USPTO Open Data Portal returns no AIA trial activity, and nothing surfaced in web/PTAB-document searches that contradicts that. There is no breakdown by outcome because there are no proceedings: 0 active, 0 claims invalidated, 0 claims sustained, 0 settled, 0 institution denials. The bottom line for a defendant is unusual but clear: this is not a "hardened by IPR" patent — it is an un-petitioned patent. All 12 claims stand exactly as they issued on 2000-08-22, never tested at the Board, and the patent (priority 1998-05-29) reached its anticipated expiration on 2018-05-29. A defendant is therefore litigating an expired patent whose validity has never been adjudicated by the PTAB — which cuts both ways (no estoppel/no dead claims, but also no free kill from a prior FWD).

(Because there are no proceedings, the per-proceeding template below is not populated. I am not going to invent proceeding numbers, panels, or dispositions. The remainder of this memo gives the defensive-relevant context that a real proceedings list would otherwise supply.)

No proceedings to enumerate

There is no IPR20XX-#####, PGR20XX-#####, or CBM20XX-##### for this patent. I specifically checked for:

• Any IPR/PGR/CBM naming US 6,108,594 (also searched the "594 patent" family and the Hughes Electronics / AT&T assignee chain) — no hits.
• Any Federal Circuit appeal of an FWD on this patent — none found.
• Any Unified Patents / defensive-aggregator challenge — none found.

Searches for "594" + PTAB return different patents (e.g., U.S. 7,386,594 in an email-URL IPR, and a " '594 patent" in a services/enterprise-technology IPR). Those are not this patent and must not be attributed to it. Flagging this because the " '594" shorthand is a common trap when counsel runs a quick '594 docket search — verify by full number 6,108,594, not by the last three digits.

Strategic summary

Claim status — CANCELED vs. SUSTAINED vs. UNTESTED. There is no IPR narrowing here. Every claim is UNTESTED, and per the face of the patent all 12 claims (independent claims 1 and 7; dependent claims 2–6 and 8–12) issued and were maintained through the last fee payment on file (12-year fee, 2012-02-22). So there is no "surviving claim set" produced by a Board decision — the claim set is intact: claim 1 (autonomous acquisition method: observe a first set of stars → form candidate star triplets → pattern-match to the OSC → form a set of candidate attitudes → rejection method), claim 7 (the spacecraft/apparatus counterpart), and the dependent claims covering the two rejection variants (claim 2/8 "second set of stars, reject if <3 match the OSC"; claims 3/9 gyro-propagated-attitude consistency table + "largest string of consistent measurements" until a "critical length") and the inertial-sensor-calibration subject matter (claims 4/5, 10/11).

Estoppel landscape. Because no petitioner ever filed, § 315(e)(2) estoppel is a blank slate. No defendant is barred from raising any § 102/§ 103/§ 112 ground. That is the opposite of the usual post-IPR posture: there is no pre-screened, Board-approved art that a plaintiff can say "you already lost on," and equally no free invalidity finding. If you file an IPR now, you would be the first petitioner — expect the patent owner to have no prior-IPR playbook for this patent, but also be aware that the PTAB may exercise § 314(a)/Fintiv-style discretion (and the § 325(d)/§ 315(b) clocks) based on any co-pending district court case.

Pattern signals. None of the usual tell-tales are present: the same petitioner has not filed multiple IPRs (there are zero), the patent owner has not pursued a PTAB appeal (no FWD), and no defensive aggregator appears in the chain. Assignee history is Hughes Electronics → (via the AT&T/Loral/DIRECTV corporate lineage) AT&T MVPD Group LLC per the Google Patents current-assignee field. This is a 1998-era Hughes spacecraft-attitude patent; it has never been asserted in a way that drew an IPR, which is itself the signal — it reads like a portfolio/museum patent, not a litigation-funded assertion vehicle. That said, the absence of IPRs is not proof of non-assertion; it just means no one has taken it to the Board.

One timing caveat that matters for defense strategy. The patent's 20-year term ran from the 1998-05-29 priority date and expired on 2018-05-29 (Google Patents legal status: "Expired - Lifetime," anticipated expiration 2018-05-29). Damages exposure on an expired patent is bounded by the pre-expiration period plus the § 286 six-year look-back, so any assertion today is chasing a closed window. IPR of an expired patent is technically available (the Board can still cancel claims), but the ROI is usually poor because there is no prospective injunction value — the practical lever is the § 286 damages clock, not claim cancellation.

Recommended next steps

1. Treat the "no PTAB activity" result as confirmed but re-verify at the source before relying on it. Pull the patent's trial documents directly from USPTO PTAB E2E (https://ptab.uspto.gov) by entering 6,108,594 (and separately, the application number 09/087,409) — do not search by " '594." Also confirm via CourtListener (https://www.courtlistener.com) that no Federal Circuit appeal naming the patent exists.

2. If you are a defendant being asserted: you are not facing canceled claims, so an "the claims are already dead" argument is unavailable — do not assert it. Your freedom-to-operate defense is an expired-patent / § 286 damages-window argument plus an original invalidity case you would have to build from scratch (the patent cites Jacob U.S. 4,621,329; Harigae U.S. 4,944,587; Kosaka U.S. 4,658,361; Boinghoff U.S. 5,177,686 as the single-snapshot/pattern-matching prior art, and cites U.S. 5,935,195 (Matra Marconi, "Autonomous star identification") and U.S. 5,745,869 (Lockheed, "Techniques for optimizing an autonomous star tracker") among references of record — those are the natural prior-art starting points for a § 103 theory on the "multiple snapshots + gyro-propagated attitudes + rejection" combination).

3. If you are considering filing the first IPR: there is no § 315(b) one-year bar triggered unless/until you are served with a complaint, and no § 315(e) estoppel to worry about. But weigh § 314(a) discretionary denial against any parallel litigation, and note the expired-patent posture — the Board will still institute, but the commercial payoff is limited to neutralizing a damages claim.

4. If no one is asserting this patent: the correct read is the plain one — there is no PTAB activity on file, the patent is expired, and the claims were never tested at the Board. The absence of IPRs on a 1998 Hughes attitude-determination patent is unremarkable; it is not a "hardened by litigation" signal and not a "troll" signal. It is simply an untested,

I'll research the assignment record for US 6108594 across USPTO assignment sources and related databases.

Key lead found — the "AT&T MVPD Group LLC" name traces back to a Hughes→DIRECTV corporate lineage. Let me pin down the recorded reels and check for licensing-arm transfers and litigation.

I hit the tool-call limit while chasing the last few queries, so I'll flag exactly where the evidence stops. Critically, I could not query USPTO Assignment Center directly — it is an interactive application that doesn't render to web search or fetch. Everything below is reconstructed from (a) the assignment data printed on the Google Patents legal-events record for this patent, (b) INPADOC assignment/change-of-name records retrieved for sibling Hughes/DIRECTV patents, (c) a Second Circuit corporate disclosure statement, and (d) DIRECTV/S2 Licensing press releases. Where an inference crosses from one patent to this patent, I say so.

⚠️ Correction/upgrade to the prior section: The earlier summary flagged "current assignee = AT&T MVPD Group LLC" as medium confidence. I can now raise that to high confidence for the entity identity, because a court filing states it directly: "The DIRECTV Group, Inc. has changed its name to The AT&T MVPD Group, LLC" and "The DIRECTV Group, Inc. has been converted to a limited liability company under the name The AT&T MVPD Group, LLC" (Brief of Appellee, Hughes Communications India Pvt. Ltd. v. The DIRECTV Group, Inc., No. 21-3013 (2d Cir.), Corporate Disclosure Statement, filed Aug. 12, 2022, https://fingfx.thomsonreuters.com/gfx/legaldocs/movakqkjxva/CA2%20Hughes%20India%20brief%20of%20appellee.pdf). That is a name change / entity conversion, not a sale — which materially changes the NPE analysis.

Inventors

All five executed the assignment of record to Hughes Electronics Corporation (Reel 009482/0157, effective 1998-08-13), which is strong evidence of common employment, but I did not independently verify each inventor's employer-of-record at filing — treat the "employer" column as assignment-inferred, not payroll-verified.

Unusual-pattern check: I found no evidence of a mass inventor departure within 12 months of the 1998-05-29 filing, and no inventor-side assignment back out of the portfolio. The residences cluster in the Los Angeles basin (Hughes' El Segundo / space-and-communications footprint) with one outlier in Illinois — I cannot explain Didinsky's IL address on the evidence I have, and I won't speculate. Unclear / not determinable is the honest call.

Original assignee

Hughes Electronics Corporation, El Segundo, CA (assignment recorded 1998-09-04, effective 1998-08-13, Reel 009482/0157).

• Did it ship a product embodying the claims? Yes, in substance. Hughes Electronics' Hughes Space and Communications business built and flew geosynchronous communications satellites; the patent is a star-tracker/gyro attitude-acquisition algorithm of the kind embedded in spacecraft control processors (specification, FIGS. 1–3, describing CCD star trackers 20/25, inertial sensors 36, and spacecraft control processor 30). Direct documentary proof that this specific algorithm flew on a specific bus is not in my sources — flagging that.
• Primary line of business: satellite manufacturing (Hughes Space and Communications) and satellite television (DIRECTV), under the Hughes Electronics holding company.
• Current status: Not dissolved. Hughes Electronics was renamed The DIRECTV Group, Inc. (change of name effective 2004-03-16, per INPADOC change-of-name records on sibling Hughes patents), and that entity was later converted to The AT&T MVPD Group, LLC (a Delaware LLC and wholly owned indirect subsidiary of AT&T Inc., per the 2d Cir. corporate disclosure above).

⚠️ Note a divergence worth flagging to a diligence team: Hughes' spacecraft manufacturing business (Hughes Space and Communications) was sold to Boeing in 2000 — yet the assignee of record for this spacecraft-attitude patent did not travel with the space business; it stayed with the Hughes Electronics/DIRECTV television-holding lineage. I have not verified the 2000 Boeing transaction documents and cannot state how this exclusion was papered. Medium confidence on that point.

Assignment timeline

Recorded assignments on this patent's own record are limited to the original 1998 filing. The downstream events below are corporate successions whose reels I confirmed on other Hughes/DIRECTV patents; I could not confirm that any of those specific reels name US 6,108,594, because Assignment Center was not directly queryable in this session.

• 1998-05-29 (executed) / recorded 1998-09-04 — Reel 009482/0157

  • Conveyance: Assignment (ASSIGNMENT OF ASSIGNORS INTEREST)
  • Assignor: Garry Didinsky; Yeong-Wei Wu; Rongsheng Li; Arunkumar Nayak; Douglas Hein
  • Assignee: Hughes Electronics Corporation, El Segundo, CA
  • Correspondent: not visible in the sources I retrieved. The patent face lists agents T. Gudmestad and M.W. Sales (Hughes in-house style); I will not assert a correspondent name I did not read on the record.
  • Context: original employment/obligation assignment — the only conveyance on this patent's own chain. Nothing exotic.

• 2004-03-16 (executed/effective) / recorded circa 2010 on sibling records — Reels observed: 024160/0372, 025114/0964, 027058/0708, 029766/0873 (four different reels, all the same 2004-03-16 event, on four different Hughes patents)

  • Conveyance: Change of Name
  • Assignor: Hughes Electronics Corporation
  • Assignee: The DIRECTV Group, Inc.
  • Correspondent: not captured (INPADOC record shows only new-owner/effective date/reel).
  • Context: internal reorg / change of name only — no change in beneficial ownership. Note the late 2010 recording dates despite the 2004 effective date; that lag pattern is a chain-cleanup tell (typically done ahead of a financing or licensing campaign), not a transfer tell.

• 2013-02-19 (executed/effective) / recorded 2013-02-19 — Reel 029833/0363 (observed on sibling Hughes/DIRECTV patents)

  • Conveyance: Assignment
  • Assignor: The DIRECTV Group, Inc.
  • Assignee: DTVG Licensing, Inc. (wholly owned subsidiary of The DIRECTV Group, Inc.)
  • Correspondent: not captured.
  • Context: transfer to a captive licensing arm — but a licensing arm of an operating company, used to run the DVB-S2 standard-essential pool ("S2 Licensing," launched Aug. 25, 2008; https://portal.unifiedpatents.com/patents/patent/EP-[1487166](/patent/1487166)-A3 shows current assignee "Dtvg Licensing Inc" for DIRECTV DVB-S2 families). I have no evidence US 6,108,594 was included in Reel 029833/0363 — and in fact Google Patents reports the current assignee of this patent as AT&T MVPD Group LLC, not DTVG Licensing, which suggests this spacecraft patent was left behind with the parent.

• 2021 (conversion of DIRECTV Group, Inc. to LLC; AT&T/TPG video restructuring closed 2021-08-02)

  • Conveyance: entity conversion / change of name (stated in the 2d Cir. corporate disclosure; I did not read a USPTO reel for it)
  • Assignor: The DIRECTV Group, Inc.
  • Assignee: The AT&T MVPD Group, LLC
  • Context: internal reorg pursuant to the AT&T/TPG contribution transactions; also note "DIRECTV Group Holdings, LLC → AT&T MVPD Group Holdings, LLC." Real parties in the related IPR: AT&T Services, Inc., AT&T Communications, LLC, DIRECTV, LLC (IPR2021-00603, Petitioners' Revised Mandatory Notices, Aug. 20, 2021).

• 2018-05-29 — anticipated expiration; patent expired (fee-paid through the 12-year window; no lapse flag). The asset is off the assertion table regardless of ownership.

Litigation check: I found no lawsuit, ITC action, or PTAB proceeding naming US 6,108,594. Searches returned unrelated matters involving AT&T/DIRECTV as parties (e.g., AT&T Services v. Broadband iTV, IPR2021-00603) but nothing naming this patent. Consistent with the prior section's finding of no 2026 CAFC docket, though I could not query PACER/Docket Navigator.

Timeline diagram

timeline
    title Ownership of US 6108594
    1998 : Filed by five Hughes inventors
         : Assigned to Hughes Electronics Corp
    2000 : Patent issues 22 Aug 2000
    2004 : Hughes Electronics renamed The DIRECTV Group
    2013 : Sibling Hughes patents go to DTVG Licensing
    2018 : Patent expires 29 May 2018
    2021 : DIRECTV Group converted to AT&T MVPD Group LLC

NPE / troll-pattern signals

1. Shell-entity transfer — not present. No licensing-only LLC ever takes this asset. The terminal entity is The AT&T MVPD Group, LLC, an operating-company subsidiary whose address of record is the DIRECTV campus at 2230 East Imperial Highway, El Segundo, CA 90245 (the same address used in DIRECTV Group's SEC filings), not a registered-agent maildrop. The only LLC-suffix entity in the vicinity, DTVG Licensing, Inc., is expressly described by its parent as "a wholly owned subsidiary of The DIRECTV Group, Inc. … the owner by assignment of [DIRECTV's] essential DVB-S2 patents" — a captive pool-licensing arm of an operating company, not a third-party assertion vehicle, and there is no evidence this patent is in that pool.
2. Known asserter in the chain — not present. No match to Acacia, Marathon, IV, IPNav, Wi-LAN, Conversant/Mosaid, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation, or any Spangenberg entity. Every name in the chain is Hughes → DIRECTV → AT&T, all operating companies.
3. Repeat correspondent across the chain — unclear. This is the one signal I genuinely could not test: USPTO Assignment Center was not directly queryable, and none of the sources I reached printed the correspondent name for Reel 009482/0157 or the downstream reels. I will not guess. Note the pattern that is visible: multiple different reels (024160/0372, 025114/0964, 027058/0708, 029766/0873) all recording the same 2004-03-16 change of name on different sibling patents — that is a chain-cleanup/re-recording signature worth a targeted reel-by-reel pull.
4. Cascading transfers — not present. No chained LLC-to-LLC sequence on this patent; the downstream changes are name changes and entity conversions of a single corporate family.
5. Pre-litigation transfer — not present. No suit found, so there is no litigation date to measure against. The 2013-02-19 sibling assignment to DTVG Licensing occurred roughly five years after the 2008 S2 Licensing campaign launch and was pool-driven, not litigation-driven.
6. Bankruptcy fire-sale — not present. No Chapter 7/11 for Hughes Electronics, The DIRECTV Group, or AT&T MVPD Group in the record; the DIRECTV/TPG transaction (2021) was a contribution/spin, not a bankruptcy sale.
7. Privateering — not present. DTVG Licensing's DVB-S2 program is standard-essential pool licensing at fixed, published, non-discriminatory rates ($0.50/$1.00 per encoder or decoder) with RAI, ESA, France Telecom/TDF — the opposite of a privateering campaign, and in any event unrelated to this spacecraft patent.
8. Defensive aggregator — not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN.

Verdict

Insufficient data — with a strong subsidiary finding that no NPE signal is present.

Justification: the only assignment of record on this patent's own chain is the original inventor→Hughes assignment, Reel 009482/0157, effective 1998-08-13, recorded 1998-09-04; everything downstream is a change of name (Hughes Electronics → The DIRECTV Group, effective 2004-03-16) or an entity conversion (The DIRECTV Group, Inc. → The AT&T MVPD Group, LLC, per the 2d Cir. corporate disclosure), neither of which is an arm's-length transfer, and I could not verify these events on US 6,108,594's own reel/record because Assignment Center was not directly queryable. I also could not retrieve the correspondent for Reel 009482/0157 — the single most diagnostic field in this task — so the honest classification is insufficient data rather than a clean "non-NPE." On the evidence available, the patent never left the Hughes→DIRECTV→AT&T operating-company family, and it expired 2018-05-29, so it is not a live assertion candidate.

Verification links:

• USPTO Assignment Center (search by patent number 6108594): https://assignmentcenter.uspto.gov/
• USPTO Patent Assignment Search (legacy interface): https://assignment.uspto.gov/patent/index.html
• Google Patents legal events / current-assignee listing for this patent: https://patents.google.com/patent/US6108594/en
• 2d Cir. corporate disclosure establishing the DIRECTV Group → AT&T MVPD Group name change: https://fingfx.thomsonreuters.com/gfx/legaldocs/movakqkjxva/CA2%20Hughes%20India%20brief%20of%20appellee.pdf
• DTVG Licensing's captive status and DVB-S2 pool: https://www.digitaltvnews.net/?p=3725 and https://portal.unifiedpatents.com/patents/patent/EP-1487166-A3

To close the two open gaps, a manual Assignment Center pull is required for: (i) the correspondent of record on Reel 009482/0157, and (ii) whether Reel 029833/0363 (DIRECTV Group → DTVG Licensing) or any post-2021 reel names application 09/087,409. Both require interactive querying that this session could not perform.

I'll research the prior art references cited on the face of US 6,108,594. Let me verify details across sources.

Let me verify the remaining cited references.

Prior-Art Analysis — US Patent 6,108,594

Patent analyzed: US 6,108,594 B1 — "Autonomous attitude acquisition for a stellar inertial attitude determination system"; App. 09/087,409; filed/priority 1998-05-29; granted 2000-08-22; Hughes Electronics Corp.
Basis: The 15 U.S. patent documents listed under "Citations (15)" and "Patent Citations (15)" on the face of US 6,108,594 (Google Patents), plus author/inventor data cross-checked against Google Patents, FreePatentsOnline, Espacenet, Justia, and patent full-text PDFs.

Legal framework note (do not skip): Because this application was filed 1998-05-29, pre-AIA 35 U.S.C. § 102 governs. Two of the references have date profiles that matter:

• References published before 1997-05-29 qualify under § 102(b) (more than one year before filing).
• US 5,745,869 (granted 1998-04-28) does not qualify under § 102(b) but qualifies under § 102(a) and § 102(e).
• US 5,935,195 (granted 1999-08-10, filed 1997-03-27) published after the filing date, so it is not a § 102(a)/(b) printed publication. It can only operate as § 102(e) art as of its 1997-03-27 filing date.

Analyst caution up front: Anticipation under § 102 requires a single reference to disclose every limitation of a claim. None of the 15 references, read literally, appears to disclose the full combination of independent claim 1 or claim 7 — in particular the pairing of (i) a set of a plurality of candidate spacecraft attitudes with (ii) a rejection method that prunes that set. The references below are therefore characterized by the claim limitations they most nearly reach, with the unmet limitations flagged. This is a screening/anticipation assessment, not a legal opinion.

A. Complete cited-reference table (the 15 examiner-cited U.S. patents)

B. Per-reference analysis against claims 1–12

Below, "reaches" = limitations the reference appears to disclose; "gap" = limitations absent from that reference alone.

2. US 4,621,329 — Itek, "Star catalog navigation system" (1979-12-31 → 1986-11-04) — high relevance

• Description (from the granted text): Stores a geometric pattern of stars imaged through a tunable acousto-optic filter plus each star's spectral ("frequency") signature, then sequentially compares that stored pattern/signature against digital data in a star-catalog memory until a match occurs, yielding spacecraft orientation. Uses a four-star exemplary pattern.
• Reaches: Claim 1 steps (a) observe a set of stars, (c) pattern-match observed pattern to an on-board-style catalog, and the general "determine orientation from the match." Also relevant to claim 7's catalog/processor structure.
• Gap vs. claim 1: No "candidate star triplets" (it uses a fixed four-star geometric/spectral pattern) and no "set of a plurality of candidate attitudes" pruned by a "rejection method." It matches to a single orientation rather than maintaining an ambiguity set.

3. US 4,658,361 — Hitachi (Kosaka et al.), "Method and apparatus for determining satellite attitude by using star sensor" (1983-07-25 → 1987-04-14) — high relevance

• Description (verified from PDF): Two-mode scheme. Mode I builds a star sub-catalog from rough attitude (from gyros/earth sensors) and identifies observed stars against it; Mode II searches the whole star catalog when attitude is unstable/unavailable. Estimates attitude angles (roll/pitch/yaw) from identified stars.
• Reaches: Claim 7 structure (star sensor + on-board catalog + processing unit + attitude estimation); claim 1 steps (a)–(c) and (e) at high level; the "lost attitude → search whole catalog" concept is conceptually related to acquisition.
• Gap vs. claim 1: Mode II relies on a ground-terminal data exchange for the search ("data transaction with the ground terminal"), so it is not truly autonomous; it does not form a set of a plurality of candidate attitudes nor apply a rejection method to prune that set. Also no inertial-rate-connected multi-snapshot fusion.
• Note: This is one of the two references the 6,108,594 background section names directly ("U.S. Pat. No. 4,658,361 … to Kosaka et al.").

4. US 4,680,718 — Hitachi (Sasaki et al.), "Method and apparatus of determining an attitude of a satellite" (1983-11-11 → 1987-07-14) — high relevance to claim 1(b)/(c)

• Description: Per US 5,177,686 (Boinghoff), US 4,680,718 discloses a star-identification method in which at least three catalogued stars are selected (one as guide star), and the catalog stores information on different three-star combinations — brightness totals and the triangle surface defined by the three stars — which is compared to the same quantities computed for observed three-star combinations.
• Reaches: Directly reaches claim 1 step (b) ("forming a set of candidate star triplets") and step (c) (pattern matching each triplet against the catalog).
• Gap vs. claim 1: No explicit "set of a plurality of spacecraft attitudes" pruned by "a rejection method"; it identifies a three-star configuration rather than retaining/rejecting multiple attitude candidates over time.

7. US 4,944,587 — Toshiba (Harigae), "Star sensor" (1988-03-17 → 1990-07-31) — relevant (background)

• Description: Pattern-matching star sensor based on a single snapshot of the sky. The 6,108,594 background cites it as an example of single-snapshot pattern matching that fails when too few catalog stars are in the FOV.
• Reaches: Claim 1 steps (a)–(c).
• Gap: Single snapshot only; no multi-snapshot/inertial fusion and no candidate-set rejection — indeed it is the specific prior-art approach the patent distinguishes. Anticipation of claim 1 is very unlikely; it is § 102 background art.

8. US 5,177,686 — Messerschmitt-Bölkow-Blohm (Boinghoff et al.), "Method for determining the position of a space vehicle by means of star acquisition and star identification" (1989-10-26 → 1993-01-05) — high relevance

• Description (verified): After a vehicle is rotated from a known starting attitude to a desired attitude, a reduced visual field is defined; at least three catalogued reference stars (one = guide star) are selected; the brightest-star positions are found in windows; distances between sensed stars are compared with the catalog's known distances; if the distances do not coincide, the process repeats with a different candidate star until a match is found; attitude error is then derived.
• Reaches: Claim 1 steps (a), (b) (three catalogued stars), (c) (pattern matching geometry to catalog), and (e) — the iterate-on-a-different-candidate-until-match loop is an analog of a rejection method. Also claim 7 structural elements (star sensor, catalog, on-board computer).
• Gap vs. claim 1: It presupposes a known starting attitude and a commanded rotation ("desired new attitude"), so it is not autonomous acquisition when lost in space. It also does not define a retained set of a plurality of attitudes formed from pattern matches.

14. US 5,745,869 — Lockheed (van Bezooijen), "Techniques for optimizing an autonomous star tracker" (1995-09-28 → 1998-04-28) — strongest § 102(e)/(a) candidate

• Description (verified from granted text/claims): On-board detector + computer + guide-star database. Generates a set of star match groups by matching pairs of observed stars to pairs of guide stars; "Match groups having small numbers of stars are eliminated," non-eliminated groups are validated, "Redundant match groups are eliminated," the "match group having the largest number of stars is selected," and attitude is derived from the selected group. Uses spacecraft angular rate to predict which guide stars ("propagation guide stars") will remain in the FOV at a later time, and updates attitude autonomously without a priori attitude knowledge.
• Reaches (my assessment):
  • Claim 1: (a) observe first set ✔; (c) match to on-board database ✔; (d) produce multiple match groups (~candidate attitudes) ✔; (e) eliminate small/ redundant groups = rejection method ✔.
  • Claim 3 / 9: "selecting the match group having the largest number of stars" is closely analogous to "determining a largest string of consistent measurements"; iterating while groups are below a threshold maps to "repeating … while the largest string is below a critical length."
  • Claim 7: detector + on-board computer + guide-star database, with spacecraft rate used for propagation — arguably reaches the "inertial sensor for detecting body rates" limitation, though the rate is used for star prediction rather than being a discrete inertial sensor.
• Gaps: It matches pairs, not "triplets"; the rejection is by group size/mutual redundancy, not the claim-2 "second set of stars not matching the OSC." Literal anticipation of claim 1 could fail on "triplets," but this is the closest single-reference disclosure of the candidate-set + pruning concept.

15. US 5,935,195 — Matra Marconi (Quine), "Autonomous star identification" (1997-03-27 → 1999-08-10) — strongest § 102(e) candidate

• Description (verified): Receives an electronic star image, selects a target star plus at least two others forming a triad, determines the star separation values of the triad, geometrically orders them, and compares against a database derived from a star catalog. Claim 4 recites "rejecting any identification which is not feasible relative to the majority of target star identifications." Candidate matches are stored and cross-checked; the patent expressly targets autonomous operation (no a priori attitude) aboard a spacecraft.
• Reaches (my assessment):
  • Claim 1: (a) image of a region of sky ✔; (b) target + two others = triple/triad ✔; (c) compare to catalog-derived database ✔; (d) multiple candidate matches identified and stored ✔; (e) rejection of infeasible identifications ✔.
  • Claim 2 / 8: the "reject relative to the majority" logic is a functional analog of "eliminating … when a number of the stars … do not match the on-board star catalog."
  • Claim 7: star camera + on-board microprocessor + database; no explicit inertial sensor for body rates.
• Note on § 102 category: Because it granted 1999-08-10 (after the 1998-05-29 filing), it is not a § 102(a)/(b) publication; it is available only as § 102(e) art as of its 1997-03-27 filing date. This matters if the applicant's date of invention precedes 1997-03-27.

13. US 5,412,574 — Hughes Aircraft (Bender et al.), "Method of attitude determination using earth and star sensors" (1993-05-14 → 1995-05-02) — relevant to calibration claims

• Description: Combines earth-sensor and star-sensor data to determine spacecraft attitude.
• Reaches: Generally the claim-7 architecture (star tracker + processing) and, more specifically, the calibration/attitude-update subject matter of claims 4–6 / 10–12 (attitude determination that updates sensor parameters). Same assignee family as the patent (Hughes).
• Gap: No candidate-triplet pattern matching/rejection; no inertial body-rate sensor in the claim-7 sense; the calibration of claims 5/11 is a gyro-bias estimate, which this reference does not literally recite.

9. US 5,109,346 — Microcosm (Wertz), "Autonomous spacecraft navigation system" (1990-02-01 → 1992-04-28) — relevant (background/architecture)

• Description: Autonomous on-board spacecraft navigation (position/attitude) system.
• Reaches: Claim 7's broad architecture (on-board processing for autonomous attitude/navigation); background support for claims 4–6 / 10–12 (sensor calibration).
• Gap: Does not disclose the triplet pattern-match-then-reject pipeline of claim 1/7. (Independent re-verification of this reference's full text was not completed this session — see caveats.)

1, 6, 10, 11, 12 — background/technological references

• US 4,617,634 (Mitsubishi; satellite attitude control) — attitude-control background; not anticipatory of the acquisition pipeline.
• US 4,746,976 (Ford Aerospace; Kamel et al.; star sightings for image navigation) — star-sighting/image-navigation background.
• US 5,204,818 (USAF; surveying satellite apparatus) — satellite attitude/navigation background.
• US 5,508,932 (Aérospatiale; earth acquisition via pole star) — attitude-acquisition background (whole-spacecraft slew to a known reference), but not pattern-match/rejection from an on-board star catalog of ambiguous candidates.
• US 5,473,746 (Loral; Pritt & Greening; interactive graphics planning tool) — cited as technological background (a ground-based maneuver-planning GUI). Contains no autonomous on-board acquisition algorithm; not anticipatory.

5. US 4,680,716 — U.S. Philips, "Automatic method and apparatus for dosing samples" (1983-12-02 → 1987-07-14) — non-analogous art; flag

• Description: Dosing/sample-handling apparatus (per its title). This is chemically/mechanically unrelated to spacecraft attitude determination.
• Reaches: Nothing in claims 1–12. Its presence in the citation list is almost certainly an artifact of the citation-matching/copying process (a well-known Google-Patents/OCRed-citation-list issue) rather than substantive prior art. I recommend treating this entry as non-prior art for this analysis.

C. Consolidated anticipation map (what would a § 102 rejection actually look like?)

Bottom line: The two references most capable of supporting a § 102 challenge to independent claims 1 and 7 are US 5,745,869 (van Bezooijen) and US 5,935,195 (Quine) — the former for the "candidate-set + rejection/largest-group" concept, the latter for the "triplet (triad) → catalog match → reject infeasible" concept. Each alone appears to miss at least one limitation (pairs vs. triplets for the former; absence of an explicit "plurality of spacecraft attitudes" set for the latter), which is consistent with the patent's novelty residing in the combination of multi-snapshot candidate-attitude formation and rejection using inertial (gyro) propagation — a combination none of the cited references discloses in a single document.

D. Corrections and flags relative to the earlier generated sections

1. Inventor of US 5,935,195 is "Quine," not "Quinz." The earlier Patent-summary section listed "US 5,935,195 (Quinz)." Search results (Justia/Google Patents citation lists) show the inventor as Quine. Per the operating rule to prefer search results, the name is Quine.
2. US 4,680,716 is non-analogous art. The earlier sections did not flag it; its title ("dosing samples," Philips) shows it is unrelated to spacecraft attitude and should be excluded from any substantive § 102 analysis.
3. The patent's own background-section attributions are internally inconsistent with the face citation list. The specification names only four pattern-matching references (4,621,329 "Jacob"; 4,944,587 "Harigae"; 4,658,361 "Kosaka et al."; 5,177,686 "Boinghoff et al.") while the face lists 15 references, several of which (e.g., 4,680,716) are unrelated. The four named references are the substantive ones the examiner and applicant treated as the closest art; the remainder are largely background/technological citations.
4. Reconciliation with the litigation finding: Nothing here changes the earlier "no litigation found" conclusion. Note that several forward citations (US 6,289,268; US 6,357,615; US 6,691,033; US 6,760,227; US 7,216,036; US 9,073,648; etc.) are later patents citing 6,108,594 as prior art — not litigation and not prior art to 6,108,594.

E. Caveats on verification

• Verified this session via live search: US 4,621,329, US 4,658,361, US 5,177,686, US 5,745,869, US 5,935,195 (full text / abstracts / claims).
• Descriptions drawn from secondary citation records and titles (not independently re-verified in full text this session): US 4,617,634, US 4,680,718, US 4,680,716, US 4,746,976, US 4,944,587, US 5,109,346, US 5,204,818, US 5,508,932, US 5,473,746, US 5,412,574. For US 4,680,718 the three-star-combination description is corroborated by US 5,177,686's own background text (verified). Three planned verification searches returned a tool step-limit error before I could complete them; treat those entries as medium confidence on description and high confidence on the citation strings/dates taken directly from the patent's face.
• § 102 categorization is based on the dates as printed on the patent's face and assumes pre-AIA law (filing 1998-05-29). Any sworn date of invention earlier than a reference's filing/publication date can change which references are available.
• This is a technical prior-art screen, not a legal opinion on validity, and is not a substitute for a claim-chart-based invalidity analysis or a PACER/Lex Machina litigation check.

I'll ground this § 103 analysis by verifying the key prior-art references of record before applying the Graham v. John Deere framework.

Let me verify the two "single-snapshot" references the patent itself distinguishes over, plus the gyro-calibration prior art.

I've reached my search limit, but I gathered enough grounding (the references of record plus the sibling PCT's own search report) to complete the § 103 analysis. Below I build on — rather than repeat — the claim summary from the earlier sections.

§ 103 Obviousness Analysis — US 6,108,594

0. Method and framing

I apply the Graham v. John Deere framework (scope/content of prior art → differences → level of ordinary skill → secondary considerations) with the KSR rationales layered on top (predictable combination of known elements, known design need, "obvious to try," finite number of identified solutions).

Priority/filing date: 1998‑05‑29 (no earlier priority claimed). Every reference below must therefore pre-date that date as prior art, which requires care for one reference (US 5,935,195) addressed below.

One correction to flag against the earlier sections: the earlier "Patent summary" lists the priority art as including "U.S. Pat. No. 4,621,329 (Jacob)" and "4,944,587 (Harigae)" — this is confirmed by the specification text, which expressly distinguishes Jacob and Harigae as single‑snapshot methods. No contradiction; I'm relying on that framing.

1. Scope and content of the prior art (references of record)

Sibling‑application evidence of the art's level. The European search report for the related van Bezooijen PCT (WO 97/11882, priority US 08/534,378) categorized US 4,658,361 (Kosaka) as "Y" and the van Bezooijen "True‑sky demonstration" paper as "Y" against a substantially similar claim 1 — i.e., the examiner of the same technology treated a Kosaka + van Bezooijen‑paper combination as rendering that claim lacking an inventive step. See WO1997011882A3 PDF. That is direct, citable evidence that combining a pattern‑match star‑ID reference with a gyro‑propagation reference was within the ordinary skill level by 1997.

2. Level of ordinary skill (PHOSITA)

A person with a B.S./M.S. in aerospace engineering, physics, or a related field and approximately 2–5 years of experience in spacecraft attitude determination (star trackers, star catalogs, gyro propagation, attitude estimators). By 1998 this person would have been familiar with: (a) triplet/triad angular‑separation star ID (Jacob, Quine, Junkins‑type work); (b) match‑group generation and pruning (van Bezooijen); (c) gyro‑propagated attitude updates and star‑referenced gyro‑bias calibration (NASA 1979; Wertz). The 1996 NASA report confirms that multi‑star, multi‑snapshot, gyro‑assisted operation was the state of the art heading into the '594 filing.

3. Differences between the claims and the prior art, and the combinations

Combination A — Claim 1: van Bezooijen '869 (primary) in view of Quine '195 (or Jacob '329)

Motivation to combine. (1) Both references are in the identical field (autonomous, no‑a‑priori star‑based attitude determination) and address the same "lost in space" problem. (2) The triplet/triad is a known, orientation‑invariant star‑pattern primitive — Quine explains why triads (geometric separation identity is independent of image orientation and reduces reliance on magnitude). (3) '869's own framework is pair‑based match groups sharing a kernel star; upgrading from a pair to a triplet is a predictable substitution of one known star‑ID primitive for another with an expected improvement in disambiguation (this is the KSR "known technique, predictable result" rationale). (4) The '869 background itself describes the pair‑based candidate‑match approach (van Bezooijen "True‑sky demonstration"), so the two references are not merely combinable but expressly cross‑referenced in the same line of development.

Bottom line for claim 1: the only applicant‑added feature over '869 is "triplets" (vs. pairs) and the explicit retention of a plurality of attitudes before pruning. Both are shown; claim 1 is a strong § 103 candidate. (Independently, claim 1 is arguably anticipated by '869 but for the "triplet" wording — that gap is what makes it obvious rather than anticipated.)

Combination B — Claim 2: van Bezooijen '869, optionally + Harigae '587/Jacob '329

• "observe a second set of stars" → '869 cl. 1: "at said subsequent time, determining which stars in the field of view correspond to the selected propagation guide stars."
• "eliminating certain spacecraft attitudes… when a number of the stars… do not match the on‑board star catalog" → '869: eliminating match groups with "small numbers of stars" (i.e., too few observed stars matched the guide‑star database). This is the same operation the '594 specification labels "step 205."

Motivation: the patent's own stated purpose — "Disclosed prior systems, which use only a single snap‑shot of the sky, require a sufficient number of stars…" — is the classic KSR "known problem in the field," and the known remedy (take another snapshot and require agreement) is exactly what '869 teaches. Repeated single‑snapshot matching to build confidence is also the natural use of the Jacob/Harigae snapshot technique.

Combination C — Claim 3 (and claim 9): van Bezooijen '869 + Kosaka '361 and/or NASA 1979

Motivation: discriminating among multiple retained candidate attitudes by requiring temporal consistency under gyro propagation is the natural discriminator once you decide (per Combination A) to keep several candidates. The 1996 NASA report confirms the field already used "hybrid" algorithms "robust to extremes in gyro propagation errors" that retain pattern‑match candidates and refine them with gyro‑propagated a priori attitude. A PHOSITA would have been motivated to add a thresholded, time‑propagated consistency check across successive star sightings — an "obvious to try" step with predictable results. The "table of C/I entries" (patent Table 1) is a routine implementation choice for recording pairwise consistency.

Combination D — Claims 4–6 (and 10–12): inertial‑sensor / gyro calibration over NASA 1979 and Wertz '346 (optionally Bender '574)

• "observing a star at a first time; determining when the observed star has been measured previously; retrieving reference measurements; computing a bias estimate" → NASA 1979 discloses a Kalman filter that, from star‑pattern attitude measurements, estimates attitude and biases (gyro drift); star‑referenced gyro‑bias calibration was routine. Wertz '346 (autonomous spacecraft navigation) and Bender '574 (Hughes; star tracker + on‑board catalog) provide the star/gyro combination context.
• Claims 4–6 do not require any novel calibration math — they recite the generic loop: see a star, recognize it was seen before, recall the reference, and update the bias. That is textbook star‑referenced gyro calibration.

Motivation: gyro drift must be estimated and removed before gyro propagation is trustworthy — which is a necessary predicate for the '594's own gyro‑propagated rejection (claim 3). The motivation is therefore intrinsic to the system the applicant himself built; combining calibration with star ID is a routine, expected design requirement, not an inventive leap.

Combination E — Claim 7 (apparatus), built from the above

Claim 7 adds structure that maps directly:

• "star tracker for observing stars" → '869 detector/camera; Jacob vidicon; Bender star tracker.
• "inertial sensor for detecting body rates" → the angular‑rate source of '869 cl. 1 + the gyros of Wertz/NASA 1979/Izumida.
• "on‑board star catalog which contains information on star position and brightness" → '869 guide‑star database (positions) + Quine/Jacob (magnitudes/intensities).
• "spacecraft control processor… programmed to [perform steps]" → the '869 computer/module structure.

Motivation: combining a star tracker, an inertial sensor, an on‑board catalog, and a processor to run the pattern‑match/prune loop is the aggregation of familiar elements according to their known functions (KSR), with no change in their respective principles of operation.

4. Why a PHOSITA would combine — consolidated motivations

1. Common field and common problem. All primary references address "lost‑in‑space" autonomous star attitude acquisition; '869 and Quine are pre‑1998 and squarely on point.
2. Express cross‑referencing. Quine's background discusses van Bezooijen's pair‑candidate method; '869 descends from the van Bezooijen "True‑sky demonstration." The references are in the same design lineage.
3. Known design need. The '594 specification itself concedes the problem: single‑snapshot methods fail with narrow‑FOV/low‑sensitivity trackers. A recognized problem whose solution is taught elsewhere is the paradigm KSR motivation.
4. Predictable, finite solution space. "Require more stars in one snapshot" vs. "retain multiple candidates and prune across successive snapshots using gyro propagation" — a small, identified set of options with predictable results = "obvious to try."
5. Necessary integration. Gyro calibration (claims 4–6) is a prerequisite for the gyro‑propagated rejection of claim 3, so the two are not merely combinable but functionally linked.
6. Examiner‑level confirmation. The sibling PCT's own search report treated a Kosaka + van‑Bezooijen‑paper combination as addressing the same subject matter at the "Y" level, evidencing that such combinations were within the ordinary skill level.

5. Secondary considerations (weak on this record)

• Long‑felt need? The need (C narrow‑FOV trackers) is real, but the '594 admits it was already recognized, and others (van Bezooijen, Quine, NASA) were addressing it — this cuts toward obviousness, not away.
• Unexpected results? None identified; the claims recite a systematic pruning of candidates, an expected consequence of retaining them.
• Commercial success / copying / licensing / industry praise? No evidence surfaced. As noted in the earlier sections, the patent expired 2018‑05‑29 and was never asserted (0 PTAB proceedings, no litigation found), which undercuts any secondary‑consideration narrative.
• Nexus: absent a demonstrated nexus between any success and the claimed "triplet + rejection" combination, secondary considerations carry little weight.

6. Confidence, caveats, and counterarguments

• Confidence: Moderate‑to‑high that claims 1, 2, 7, 8 would be found obvious over van Bezooijen '869 alone or in view of Quine '195 / Jacob '329; moderate for claims 3 and 9 (the "consistency table / largest string / critical length" implementation is the most detailed and would need the Kosaka/NASA‑1979 combination); moderate for claims 4–6 and 10–12 (calibration is old, but the specific "measured previously → retrieve → bias estimate" loop should be pinned to a specific reference for the strongest prima facie case).
• Patent owner's best non‑obviousness arguments (and my rebuttals):
  1. "'869 selects the single largest match group; '594 deliberately keeps a plurality of attitudes and prunes over time." — Rebuttal: '869 also generates and retains multiple match groups before elimination; retaining‑then‑pruning is disclosed or an obvious variation.
  2. "'869 uses pairs, not triplets." — Rebuttal: triplet/triad ID is a known primitive (Quine/Jacob); predictable substitution.
  3. "'869 doesn't use a table of pairwise consistent/inconsistent entries." — Rebuttal: '869's IDGRP matrix + "validating match groups" disclose the substance; the table is a routine recording choice.
• § 102(e) subtlety I want to flag explicitly: US 5,935,195 (Quine) issued 1999‑08‑10, after the '594 filing date, so it is not § 102(a)/(b) art. It qualifies only as pre‑AIA § 102(e) art as of its US filing date of 1997‑03‑27 (assuming the '594 invention date is not earlier than that — there is no evidence of earlier conception). It happens to be a reference of record on the '594 face, which strongly supports its use; but if the priority‑art table in the earlier sections implies '195 is § 102(b) art, that would be an overstatement — the correct date‑qualifier is its 1997‑03‑27 US filing.
• Claim‑construction caveat: "performing a rejection method" is a broad functional limitation; a narrow construction (requiring the specific Figure 7 gyro‑consistency table for claim 1) would make the claim harder to invalidate but is unlikely given the dependent claims separately recite the two rejection variants — under the doctrine of claim differentiation, claim 1's "rejection method" should read on both variants, keeping claim 1 broad.
• Literal‑reading note: claims 6 and 12 recite a "gryoscope." Per the operating rules I am not auto‑correcting it; it plainly means "gyroscope," and this is a correctable typo, not a substantive limitation. It does not affect the obviousness analysis (the reference art uses gyroscopes).
• Caveat on my own process: I could not complete a full independent search of every secondary reference (I hit my search‑step limit mid‑verification of Harigae '587, Wertz '346, and Pritt '746). The claim charts above are grounded in the references I did verify (van Bezooijen '869, Quine '195, Jacob '329, Bender '574, the two NASA reports, and the WO 97/11882 search report). Before relying on any single ground in a filing, the Harigae/Wertz/Pritt disclosures should be confirmed at the source.

Net assessment: US 6,108,594 reads as a narrow, incremental improvement over van Bezooijen's own match‑group/gyro‑propagation framework. Its principal additions — (i) using triplets rather than pairs and (ii) an explicit temporal‑consistency rejection table — are each traceable to known star‑ID and gyro‑propagation techniques, and the motivation to combine them (acquiring attitude with narrow‑FOV trackers and few visible stars) was expressly recognized in the field. The claims are vulnerable to § 103, with claim 1 the most exposed and claim 3/9 requiring the stronger (Kosaka/NASA‑1979) combination.