US 10132647

I'll search for this patent number in USPTO databases and CAFC dockets.

The initial searches returned unrelated results (the closest being the German DE10132647 and unrelated documents). Let me refine with more targeted queries.

No CAFC litigation results appeared. Let me do one more targeted check of the Federal Circuit docket databases.

Summary: US Patent 10,132,647

Important note on scope: My searches for 10132647 did not surface any U.S. court or CAFC docket records, and no USPTO litigation page appeared. The only "10132647" hits that matched literally were (a) the Google Patents page for US10132647B2 itself, and (b) DE10132647B4, an unrelated German patent (Fraunhofer-Gesellschaft, "Method for the controlled detection of the machining path…"), plus a handful of coincidental numeric matches in unrelated documents (e.g., a Brazilian exam-registration list). I did not auto-correct or merge these — they are distinct identifiers. I cannot confirm any CAFC 2026 docket, appeal, or litigation involving US 10,132,647 based on available results; the absence of results is not proof that none exists.

The bibliographic and claim details below come from the authoritative full patent text supplied (Google Patents, fetched 2026‑09‑24).

---

Bibliographic data

Abstract (verbatim)

"The present disclosure describes systems and methods for maintaining gyroscopic sensor accuracy over time and across changing environmental conditions. In certain aspects, the present disclosure provides arrangements and methods for calibrating a gyroscope while it is positioned on a robotic platform. In particular, the gyroscope may positioned on a sensor platform that is moved through a series of known or measured rotations and then the gyroscope signals are compared to reference data and the sensor's gain and offset calculated. In other aspects, the present disclosure provides arrangements and methods for utilizing measurements from multiple gyroscopes that measure the same axis of rotation."

---

Independent claim — plain-language overview

Claim 1 is the sole independent claim. It covers an in-place ("in-system") method of calibrating a gyroscopic sensor on an autonomous robotic base platform, comprising:

1. Rotating the gyroscopic sensor through a controlled range of rotation relative to the autonomous robotic base platform, where:
   • the sensor is held by a rotatable coupling to the platform;
   • the platform has a driving mechanism and can autonomously move around an area along a defined path;
   • the controlled range of rotation is either (i) predetermined, or (ii) independently measurable by a non-gyroscopic sensor configured to measure that range; and
   • the rotatable coupling is selectable between two states — one rotatable (calibration) and one non-rotatable (working/locked);
2. Receiving the rotation measurement detected by the gyroscopic sensor during that rotation;
3. Calculating a calibration factor with a processor, based on the gyroscope reading and either the predetermined baseline value for the range of rotation or the non-gyroscopic sensor's measurement of it; and
4. Modifying the gyroscope's readings with the processor based on that calibration factor.

Practically: the robot's own gyro is physically swung through a known or separately measured arc while the robot is stationary (coupling unlocked), the deviation from the reference is computed as a calibration factor, and the processor then corrects the gyro output — all without removing the gyro from the machine.

Dependent claims (for completeness)

• Claim 2: the non-gyroscopic sensor is an optical sensor, proximity sensor, or encoder.
• Claim 3: the predetermined range is set by a physical stop (solid-object-to-solid-object impact) or a magnetic stop.
• Claim 4: the processor is physically mounted to the autonomous robotic base platform.
• Claim 5: the sensor rotation is actuated by an electric motor, solenoid, or linear actuator.

---

Note on the specification's second inventive theme (not reflected in the claims)

The disclosure also extensively describes sensor fusion of multiple gyroscopes (FIG. 3): combining gyros of different brands with known, opposing error tendencies (e.g., a Maxim Integrated MAX21000-type sensor that drifts with age but resists shock/vibration, versus a Murata SCR1100-type sensor that resists shock/vibration but drifts faster and costs roughly an order of magnitude more), with the processor weighting proportionality based on brand, age, and environmental conditions, optionally using statistical/probabilistic methods (regression, least squares, fuzzy logic, Bayesian filtering) and a magnetometer or environmental (shock/vibration/humidity) sensor as a reference. This fusion subject matter is described but is not claimed — the five issued claims are directed solely to the in-place calibration method of claim 1.

---

Confidence and caveats

• High confidence: all bibliographic fields, claim text, and the abstract, which are taken directly from the authoritative patent text provided.
• Low / unverified: any litigation or appellate activity. My searches returned no CAFC 2026 docket or USPTO litigation records for US 10,132,647, and I am explicitly flagging that I could not confirm any such proceeding rather than inferring one. If you need a definitive litigation check, a direct query to the Federal Circuit's PACER/CM‑ECF docket system or the USPTO PatentCenter "Litigation" tab would be required; general web search did not surface it.

I'll search for litigation involving US Patent 10,132,647.

Let me search more targeted sources for litigation involving this specific patent.

Let me check patent litigation databases and docket sources specifically.

Results: Litigation Search for U.S. Patent No. 10,132,647

I could not identify any known litigation involving U.S. Patent No. 10,132,647. Based on the searches I was able to run, there is no evidence of any infringement suit, declaratory judgment action, or PTAB proceeding (IPR/PGR) that specifically names or asserts this patent.

What the searches covered

I searched for the literal patent number (10132647 / 10,132,647) combined with litigation terms, and separately for the assignee (MTD Products Inc.) in patent enforcement contexts. Note: I hit a tool step-limit before I could complete targeted queries against the Unified Patents litigation portal and the RPX Insight litigation database, so this conclusion is based on the searches that did complete.

The one case that surfaced — and why it is NOT this patent

• MTD Products Inc. v. Toro Company et al., Case No. 1:15-cv-00766 (N.D. Ohio), filed 04/20/2015, closed 11/23/2015 (Judge Patricia A. Gaughan).
  • Plaintiff: MTD Products Inc.; Defendants: The Toro Company and Exmark Manufacturing Company Inc.
  • Patents-in-suit: U.S. Patent Nos. 8,011,458 and 8,136,613 — not 10,132,647.
  • Outcome: stayed pending IPR of the asserted patents.
  • Source: RPX Insight — https://insight.rpxcorp.com/litigation/ohndce-[217195](/patent/217195)-mtd-products-v-toro-company

This is the same assignee (MTD Products Inc.) but a different patent, and it predates the '647 patent's grant (Nov. 20, 2018). It should not be counted as litigation over the '647 patent.

Important caveats

1. This does not prove a negative. Patent litigation can be hard to surface through general web search, and I could not fully query the two sources you specifically recommended (Unified Patents litigation case list and PACER) because my tool calls were cut off by the step limit. A definitive "no litigation" answer would require direct, comprehensive queries of:

   • Unified Patents litigation portal: https://portal.unifiedpatents.com/litigation/caselist
   • RPX Insight (patent-level litigation lookup)
   • PACER / CourtListener RECAP (full-text docket search on the patent number)
   • USPTO PTAB "Patent Trial and Appeal Board Endpoint" / PTAB proceedings search

2. Context on the patent. U.S. Patent No. 10,132,647 ("Methods and apparatus for increasing accuracy and reliability of gyroscopic sensors"), inventor Edward John Blanchard, assigned to MTD Products Inc., priority date 2013-10-24, granted 2018-11-20. It would be unlikely to appear in older dockets given its grant date, but no post-2018 assertion appeared in my results.

Recommendation

If you need a definitive answer, run the patent number as a full-text search in:

• CourtListener RECAP (https://www.courtlistener.com/?type=r&q=%2210%2C132%2C647%22) — searches briefs and dockets where the patent is mentioned by number.
• Unified Patents case list and PTAB proceedings search.

Based on the information I was able to gather, I must state explicitly: no known litigation involving US 10,132,647 was identified, and I cannot confirm one exists or does not exist with high confidence given the incomplete tool access.

Proceedings overview

Total AIA trial proceedings on US 10,132,647: zero (0). The structured "PTAB proceedings on file" block supplied from the USPTO Open Data Portal returns no AIA trial proceedings, and independent web searching surfaced none. Breakdown is therefore: active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0. Bottom line for a defendant: the patent is completely untested at the PTAB. Claims 1–5 stand exactly as issued on 2018-11-20 — none canceled, none narrowed by a Final Written Decision, none subjected to a § 315(e) estoppel event. There is no "hardened by IPR" story here, but there is also no "claims are already dead" story. Every invalidity theory you might raise is still virgin ground.

---

No proceedings to enumerate

There is no proceeding number to report. I will not invent one, and I found no petitioner, no institution decision, no FWD, and no appeal for this patent.

What I checked and how:

Important disambiguation. Searching "MTD Products" + PTAB reliably surfaces three proceedings:

• IPR2016-00194 — The Toro Co. v. MTD Products, Inc. (FWD, unpatentable; later terminated)
• IPR2016-00219 — The Toro Co. v. MTD Products, Inc. (FWD)
• IPR2016-01352 — The Toro Co. v. MTD Products, Inc. (institution denied)

None of these involves US 10,132,647. They involve US 8,011,458 ("mechanical control assembly" for zero-turn-radius riding mowers). That case produced MTD Products Inc. v. Iancu, 933 F.3d 1336 (Fed. Cir. Aug. 12, 2019), which vacated and remanded the Board's obviousness finding on § 112 ¶ 6 grounds. If anyone cites "the MTD IPRs" against the '647 patent, they are citing the wrong patent — the '647 patent is about gyroscope gain/offset calibration on an autonomous robotic mower platform (see the '647 specification's own U.S. Patent Publication 2011/0066395 to Judd discussion), and shares only a patent owner with the Toro litigation.

CAFC citation for the other patent's appeal, for the record: CourtListener — MTD Products Inc. v. Iancu, 933 F.3d 1336 · Finnegan case summary.

Appeal check on this patent: none found. No Federal Circuit appeal, no Rule 36 summary affirmance, no CAFC docket number associated with US 10,132,647.

---

Strategic summary

Claim status (all claims as issued and currently in force):

• Claims 1–5 — UNTESTED. No IPR, PGR, CBM, or ex parte reexamination has adjudicated any claim. Claim 1 (the sole independent claim) recites in-place calibration of a gyroscopic sensor on an autonomous robotic base platform, where the rotatable coupling is "selectable between two states with one state being rotatable and the other state being non-rotatable," and where the controlled range of rotation is either (i) predetermined or (ii) independently measurable by a non-gyroscopic sensor. Claims 2–5 are dependents adding the non-gyroscopic sensor type (optical/proximity/encoder), the physical or magnetic stop, the on-platform mounting of the processor, and the electric motor/solenoid/linear-actuator drive. None is canceled; none is confirmed. You get no help and no harm from the Board.
• The patent's own prosecution record (visible in the file history) shows claims 1–5 were allowed without the § 112 ¶ 6 / § 103 scrub that the sibling ZTR patent received, so the Board has never construed any term of this patent. That cuts both ways: no favorable construction is locked in, and no adverse one either.

Estoppel landscape:

• § 315(e)(2) estoppel is a blank slate. Because no IPR/PGR was ever instituted on this patent, no petitioner and no privy is estopped from anything. Any defendant currently facing assertion of US 10,132,647 is free to raise any prior-art ground under §§ 102/103, any § 112 ground, and any § 101 ground, in any forum, without fear of an estoppel bar. This is the single most defendant-favorable fact in this report.
• Conversely, there is no petitioner-side IPR record you can borrow: no Board claim constructions, no adopted expert credibility findings, no FWD admissions by the patent owner. You will build the invalidity record from scratch.
• § 315(b) one-year bar: tracked from service of a complaint alleging infringement. If your client has already been served and the year is running, that clock is the only IPR timing constraint in play — nothing on this patent has been burned by a prior petitioner.

Pattern signals:

• No repeat-petitioner pattern exists — there has never been a first petitioner.
• No defensive aggregator (Unified Patents, RPX, Open Invention Network, etc.) appears anywhere in the chain. No IPR, no ex parte reexam, no PGR.
• Patent owner posture: MTD Products Inc. is a sophisticated, well-counseled owner (Calfee, Halter & Griswold in the '458 litigation) that litigates aggressively when asserted and defends PTAB challenges vigourously — including pressing its own appeals to the Federal Circuit and winning a vacatur-remand in MTD v. Iancu. Expect the same posture if you file. MTD is not a shell or a troll; this is an operating company (Cub Cadet, Troy-Bilt, Yard-Man, MTD robotic mowers).
• Maintenance fees current: 4th-year fee paid 2022-05-20, 8th-year fee paid 2026-05-06 (large entity). Adjusted expiration 2034-12-18. The patent is alive, paid up, and has ~8 years of remaining term — this is not a patent anyone will simply abandon.

The absence-is-a-signal caveat. This patent issued 2018-11-20 and carries an adjusted expiration to 2034-12-18. Nine years post-issuance with zero PTAB filings suggests either (a) the patent has not been asserted in a way that provoked a well-funded defendant, or (b) it is being held for assertion activity now, in the autonomous-mower / robotic-vacuum sector where the accused products are numerous (see the cited-by art: Samsung's US 11,185,203 on robot cleaner charging). Well-asserted patents in a crowded robotics market eventually attract IPRs. If you are receiving a demand letter on this patent, you may be early in its enforcement life — which means you may also be the first petitioner, with all the first-mover advantages (no Fintiv-vs-other-petitioners dynamics, no joinder complications, no estoppel) and few drawbacks.

---

Recommended next steps

1. Do not search for an FWD — there isn't one. Any internal memo asserting "claims 1–5 were canceled in IPR" or "the patent survived IPR" is factually wrong. Correct it now; the record is empty.
2. Confirm the negative yourself before relying on it. Run the patent through USPTO PTAB E2E / PatentCenter and the USPTO Patent Trial and Appeal Board decisions page, and check the assignment record (MTD Products Inc., reel/frame per the 2018-10-09 assignment, REEL/FRAME 047107/0893). The structured ODP block and my search agree, but PTAB ingestion lags; a very recently filed petition could be missed.
3. Because there is no estoppel, your grounds menu is wide open. Target the two features that carry the claim's novelty over the Judd publication (US 2011/0066395, which the '647 specification itself acknowledges as prior art):
   • the two-state selectable rotatable coupling (rotatable ↔ non-rotatable) on an autonomous robotic base platform, and
   • the "predetermined or independently measurable" range of rotation alternative in claim 1.
     The Judd reference teaches gyroscope calibration via accelerometer/magnetometer on a removable sensor platform. The '647 patent's stated point of distinction is in-situ calibration without removal, using a motorized two-state coupling and either a physical/magnetic stop or a separate non-gyroscopic position sensor. Attack precisely that delta — a § 103 combination of Judd with a motorized index/limit-stop mechanism (e.g., a stepper-motor + hard stop, which is routine in robotics) is the natural ground.
4. If you are pre-suit and outside the § 315(b) bar, file early. First-petitioner status on an unconstrained patent is unusually valuable here: no prior FWD to distinguish, no § 325(d) "already before the Office" problem beyond Judd, and no Fintiv co-petitioner stacking. If you are already in litigation with a § 315(b) deadline running, calendar it from the service date of the complaint — that is the only clock that matters.
5. Do not rely on the Toro/MTD Federal Circuit precedent. MTD Products Inc. v. Iancu (933 F.3d 1336) construes "mechanical control assembly" in US 8,011,458 and is not binding on, or even about, the '647 claims. Citing it for this patent would be a credibility problem before the Board.
6. Watch the family. The '647 patent claims priority to U.S. Provisional 61/894,981 (filed 2013-10-24) and shares a family ID (52993917) with the application, plus the published US 2015/0114082 A1. If MTD asserts the '647 patent, check whether sibling continuations exist that could be asserted in parallel — none surfaced in this review, but the family should be mapped before you commit to a validity-only defense.

---

Verification note: all claim numbers, dates, and the zero-proceeding count above are taken from the supplied full patent text and the structured ODP block. The MTD/Toro IPR numbers (IPR2016-00194, -00219, -01352) and the Federal Circuit citation are sourced from public case reporting and are identified here solely to distinguish them from the '647 patent — they are not proceedings on US 10,132,647.

I'll verify the ownership chain and the assignee's corporate status before writing.

Let me check for post-issuance transfers, name changes, and any assertion activity.

Let me confirm MTD's asserter classification, any name change, and the Robomow product link.

I hit my tool-step limit before I could query the Assignment Center's record detail (which is where the correspondent of record lives), so I've built this from the authoritative patent text, the USPTO assignment data mirrored in Google Patents legal events, corporate/SEC-side coverage, and asserter directories. I flag exactly where the record is thin.

Ownership chain analysis — US 10,132,647

Record basis: USPTO assignment data as mirrored in Google Patents legal events for US10132647B2 (which lists one recorded assignment), plus corporate and litigation records. The Assignment Center entry itself (https://assignment.uspto.gov/patent/index.html, search "10132647") could not be opened in this session — see the correspondent caveat below.

---

Inventors

Pattern notes:

• No departure/fire-sale precursor observed. The frequently-cited red flag — all inventors leaving the original assignee within 12 months of filing — cannot be assessed from the assignment record alone, and nothing in the record suggests it. I could not verify Blanchard's job title or tenure at MTD, and I found no evidence he left.
• Caveat on name collisions. My searches surfaced patents listing "Blanchard" as an inventor in unrelated technology areas (e.g., 9,872,971; 9,950,139; 10,426,931). "Blanchard" is a common surname and those appear in a different field; I did not attribute them to Edward John Blanchard and they are excluded from this analysis.
• The application was filed with MTD as applicant yet the inventor's assignment was not recorded until 2018-10-09 — roughly 4 years post-filing and ~6 weeks pre-grant. That is unusually late for a large, well-counseled corporate filer, and suggests either a late/confirmatory recordation or a gap in the events mirror. Verifying whether an earlier assignment record exists is the single highest-value follow-up.

---

Original assignee

MTD Products Inc (also the current assignee per the record).

• Primary line of business: outdoor power equipment — lawn tractors, zero-turn mowers, walk-behind mowers, snow throwers, trimmers, handheld outdoor power equipment. Founded 1932; headquartered at 5903 Grafton Road, Valley City, Ohio 44280. Heritage brands: Cub Cadet, Troy-Bilt, Remington, WOLF-Garten, Rover, and — directly relevant here — Robomow, MTD's residential robotic mower line. 2017 revenue exceeded $2B. Sources: MTD/Stanley Black & Decker announcement (https://www.mtdproducts.com/wcsstore/ManagedContent/MTDProducts/en_US/media/SBD_MTD_Announcement_9-12-18.pdf); Nasdaq summary of the MTD/Excel transaction (https://www.nasdaq.com/articles/stanley-black-swk-buys-mtd-excel-expands-product-lines).
• Did they ship a product embodying the claims? Likely yes, and this is the key finding for the NPE analysis. The '647 claims are directed to an in-place gyroscope calibration method for an "autonomous robotic base platform" with a driving mechanism that "autonomously mov[es] around an area in a defined path." MTD sells exactly that class of product — autonomous residential robotic mowers — under the Robomow brand, and the specification itself is written around autonomous mowing robots. I have not performed a claim-chart against a specific Robomow model, so this is a product-class match, not an element-by-element infringement/coverage opinion.
• Current status: Operating; acquired at the parent level. Stanley Black & Decker (NYSE: SWK) bought a 20% stake in MTD Products/Holdings for $234M (announced 2018-09-12, closed early 2019), then acquired the remaining 80% for $1.6B, closing 2021-12-01 (https://ir.stanleyblackanddecker.com/news-events/press-releases/news-details/2021/Stanley-Black--Decker-Completes-Acquisitions-Of-MTD-Holdings-And-Excel-Industries-Creating-A-Global-Leader-In-Outdoor-Products-12-01-2021/). No bankruptcy, no dissolution, no patent fire-sale. Related MTD entities named in litigation include MTD Consumer Group Inc. and Cub Cadet LLC (Toro Company v. MTD Products Inc., MTD Consumer Group Inc., and Cub Cadet LLC, D. Minn. No. 0:10-cv-00007).
• Minor conflict flagged: MTD Products Inc's own 2015 infringement complaint pleads it "is incorporated in the State of Delaware"; PlaineSite's corporate profile lists state of incorporation as OH. I am not resolving this — it is immaterial to the chain, but noted so it isn't silently "corrected."

---

Assignment timeline

Only one assignment is recorded. Chronologically:

• 2018-10-09 (executed) / recorded 2018-10-09 — Reel 047107/0893
  • Conveyance: Assignment (USPTO code AS; "ASSIGNMENT OF INTEREST"; Google Patents event text: "ASSIGNMENT OF INTEREST (SEE DOCUMENT FOR DETAILS)"). Effective date stated as 20181009.
  • Assignor: BLANCHARD, EDWARD JOHN (individual inventor)
  • Assignee: MTD PRODUCTS INC (Ohio, per the record's owner-name field)
  • Correspondent: Not determinable from the sources available to me. Google Patents' legal-events mirror does not reproduce the correspondent/attorney or agent of record; that field appears only in the Assignment Center abstract for reel 047107 frame 0893. Because there is only one link in this chain, the recurrence test is structurally inapplicable regardless — a single appearance cannot establish a repeat-player recording attorney. Recommended lookup: open https://assignment.uspto.gov/patent/index.html, search patent 10132647 (or reel/frame 047107/0893), and capture the correspondent of record verbatim.
  • Context: Routine inventor-to-employer assignment consistent with an employment/hired-to-invent obligation — not an acquisition, not a fire-sale, not a securitization, not a transfer to an asserter, not a change of name. The only unusual feature is the ~4-year gap between filing (2014-10-24) and recording (2018-10-09).

Conveyance types NOT present in the record (each is a negative finding): no Security Agreement (i.e., no collateralization of this patent), no Merger, no Change of Name, no License recordation, no Correction, no Release.

Post-issuance transferability note — why the record is empty after 2018: the 2021 Stanley Black & Decker transaction was the purchase of equity in MTD Holdings, Inc., not a purchase of MTD Products Inc's patent assets. Buying a parent's stock does not move legal title to a subsidiary's patents, so no patent-level assignment recordation was required and none appears. The single recorded transfer therefore remains the operative chain link, and MTD Products Inc is still the record owner of US 10,132,647 — now as an SWK affiliate. Corroborating that the asset is being maintained rather than warehoused: maintenance fees were paid for the 4th year (2022-05-20) and the 8th year (2026-05-06) as a large entity.

If the Assignment Center shows no records at all for this patent number: the fallback finding is the same — the original assignee still owns it. Here, however, the inventor→MTD record is affirmatively present and documented above.

---

Timeline diagram

timeline
    title Ownership of US 10132647
    2013 : Provisional filed by inventor
    2014 : Nonprovisional filed by MTD Products Inc
    2018 : Inventor assignment recorded to MTD
         : Patent issued 20 Nov 2018
    2021 : MTD Holdings bought by Stanley Black and Decker
    2022 : Fourth year maintenance fee paid
    2026 : Eighth year maintenance fee paid

---

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT. The only recorded conveyance (reel 047107/0893) runs from a natural person to a large operating manufacturer. No "IP / Holdings / Licensing / Ventures" entity appears anywhere in the chain; no registered-agent-service address; no single-purpose LLC. Assignee address of record is a real operating HQ (Valley City, OH).
2. Known asserter in the chain — NOT PRESENT. Neither MTD Products Inc nor any affiliate matches the listed NPE families (Acacia, Marathon, IV, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, Spangenberg entities). To the contrary, the Stanford NPE Litigation Database classifies MTD Products Inc as asserter category "8 — Product company", the database's only practicing-entity category (it categorizes MTD's cases, including MTD Products Inc v. Toro Company et al., 1:16-cv-00297, N.D. Ohio): https://npe.law.stanford.edu/party/mtd-products-inc. Caveat: that categorization is party-level and case-level, not patent-level.
3. Repeat correspondent across the chain — NOT PRESENT / unclear. Only one link exists, so recurrence is impossible to establish. The correspondent of record for reel 047107/0893 is not visible in the sources I could reach and should be pulled directly from Assignment Center. I decline to name an attorney here rather than guess.
4. Cascading transfers — NOT PRESENT. Zero post-issuance assignments in the ~8 years since grant. One link in ~12 years since filing.
5. Pre-litigation transfer — NOT PRESENT. The assignment (2018-10-09) is inventor→owner, not a transfer to an asserter, and no infringement suit naming US 10,132,647 was identified in my earlier litigation sweep. The MTD enforcement campaigns I did find — MTD Products Inc v. Toro Company et al. (1:15-cv-00766 and 1:16-cv-00297, N.D. Ohio) — asserted US 8,011,458 and US 8,136,613, and the later-filed one (2016-02-09) predates the '647 grant (2018-11-20). The '647 patent could not have been in those suits.
6. Bankruptcy fire-sale — NOT PRESENT. No Chapter 7/11 involving MTD surfaced. The liquidity event was a $1.6B acquisition by Stanley Black & Decker (closed 2021-12-01), i.e., a sale of a profitable operating business, not a distressed asset sale; and the patents were not the transferred res at all (equity deal).
7. Privateering — NOT PRESENT. No operating company→NPE transfer exists, so there is nothing to be "privateered." MTD's own suits were brought in its own name as the product company — the opposite posture.
8. Defensive aggregator — NOT PRESENT. The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. It terminates at an operating manufacturer. Note the inverse implication: because none of these aggregators hold it, the patent is not "neutralized" and remains a live asset of an SWK affiliate.

---

Verdict

Insufficient data (equivalently: no NPE pattern established; chain terminates at the original operating-company assignee).

Justification. The record contains a single assignment — inventor Edward John Blanchard → MTD Products Inc, executed and recorded 2018-10-09 at reel 047107/0893 — which is precisely the "only the original assignment" condition the taxonomy treats as insufficient for an affirmative NPE call, and there is no post-issuance transfer and no assertion of this patent to analyze. Critically, the absence of an NPE call here is not merely a data gap: all eight signals are affirmatively negative, the assignee is a Stanford-classified "Product company" (category 8) that sells autonomous robotic mowers (Robomow) in the same product class the claims address, and the asset remained with MTD through the 2021 Stanley Black & Decker equity acquisition, with maintenance fees paid in 2022 and 2026 — the profile of a practicing manufacturer holding its own IP, not an asserter. The only genuinely unresolved items are the correspondent of record on reel 047107/0893 and whether an earlier assignment record exists for this patent; both should be confirmed at https://assignment.uspto.gov/patent/index.html (search "10132647") or https://assignmentcenter.uspto.gov/.

Caveats on my own coverage: I reached my tool-step limit before completing the two recommended follow-ups — a direct Assignment Center record pull (for the correspondent field) and confirmation via RPX → https://portal.unifiedpatents.com/litigation/caselist, PTAB proceedings search, and SEC EDGAR full-text search on MTD Products/MTD Holdings patent-security language. Nothing above should be read as a certified "no litigation" or "no security interest" conclusion; it is a statement about what the available records show.

I'll research each cited prior-art reference for US 10,132,647 in the USPTO/Google Patents databases. Let me run targeted searches in parallel.

Prior-Art Analysis — US Patent No. 10,132,647

0. Target identification (confirming the search was literal)

I searched the literal identifier 10132647 and confirmed the match is US 10,132,647 B2, application US 14/522,712, titled "Methods and apparatus for increasing accuracy and reliability of gyrosopic sensors" (typo reproduced literally), inventor Edward John Blanchard, assignee MTD Products Inc., priority 2013‑10‑24, granted 2018‑11‑20. I did not substitute DE10132647, US 10,132,xxx neighbors, or any other similarly numbered document. The six references analyzed below are the patent citations listed on the face of US 10,132,647 (Google Patents, fetched 2026‑09‑24; PDF mirror at https://patentimages.storage.googleapis.com/95/bd/f9/bdedbf4cd15004/US10132647.pdf).

Because the patent has only one independent claim (claim 1), the § 102 analysis below is anchored to claim 1's limitations. Note the legal point: a dependent claim (2–5) cannot be anticipated by a reference unless that same reference also discloses every limitation of claim 1, since each dependent claim incorporates claim 1 by reference. So for each reference I state whether it (a) is a candidate § 102 reference against claim 1, or (b) only supplies a limitation relevant to a dependent claim (i.e., is really a § 103 reference).

---

1. Summary table

---

2. Reference-by-reference detail

Reference 1 — US 8,548,766 B2 (Judd / Honeywell) — the most relevant prior art

Full citation: Systems and Methods for Gyroscope Calibration, U.S. Patent No. 8,548,766 B2; inventor Tom Judd (Carlsbad, CA); assignee Honeywell International Inc.; filed 2009‑09‑14; issued 2013‑10‑01. Pre-grant publication US 2011/0066395 A1.
URL: https://patents.google.com/patent/US8548766B2/en
EP family member: EP 2 306 149 B1 (https://patentimages.storage.googleapis.com/5c/e4/f5/3e1db842fb6cae/EP2306149B1.pdf).

Description: Discloses an in-field (field-calibration) system 100 in which a sensor platform carrying three gyroscopes, three accelerometers and three magnetometers is rotated in three-dimensional space (a "calibration rotation," preferably ≥360°; see US 2011/0066392 sibling). The processor (i) determines a measured rotation vector Ω_g from the gyroscopes, (ii) derives a second rotation vector Ω_m from the accelerometers and magnetometers (changes in Earth's gravitational and magnetic fields), and (iii) determines a compensation gain and/or compensation bias for the gyroscopes by comparing Ω_g to Ω_m (see claim 1 of https://insight.rpxcorp.com/patent/US8548766B2). It expressly targets gain error and bias error and periodic field recalibration without specialized factory tools.

Cross-reference / important reconciliation: The '647 specification's background chapter discusses "United States Patent Publication 2011/0066395 to Judd" and criticizes it because it "requires human intervention and removal of the gyroscope." That publication is the pre-grant version of this same US 8,548,766 B2 cited on the '647 face — i.e., the reference the inventor distinguished in the specification is the same document. This confirms (and is consistent with) the earlier "Litigation/Summary" sections' note of a "US 2011/0066395-type Judd family"; the family relation is now pinned down.

§ 102 relevance to the '647 claims:

• Claim 1 — potential § 102 candidate, but incomplete. Judd discloses: rotating a gyroscopic sensor through a controlled range of rotation (preferably ≥360°); the rotation is independently measured by non-gyroscopic sensors (accelerometer/magnetometer); receiving the rotation detected by the gyroscope; calculating a calibration factor (gain/bias) with a processor; and modifying the gyroscope readings by the compensation. Judd does not clearly disclose (a) an autonomous robotic base platform with a driving mechanism that autonomously moves a defined path, nor (b) a rotatable coupling selectable between a rotatable and a non-rotatable state. The '647 specification itself exploits exactly this gap ("there remains a need… they require human intervention and removal of the gyroscope"). So Judd is the best § 102 reference against claim 1 but realistically was overcome on the "in-system / on the robotic platform" limitations — more properly a § 103 reference for the remaining elements.
• Claim 2 (non-gyroscopic sensor is optical/proximity/encoder): Judd's non-gyroscopic sensors are accelerometer/magnetometer, not optical/proximity/encoder — does not anticipate claim 2.
• Claim 4 (processor on the platform): Judd mounts a processor system on the sensor platform/calibration system — potentially reads on claim 4 only if claim 1 is met.
• Claim 5 (electric motor/solenoid/linear actuator): Judd's rotation is a calibration rotation (driven rotation), but the actuator type is not the point of novelty.

---

Reference 2 — US 2008/0319667 A1 (Honeywell) / US 8,019,542 B2

Full citation: Heading Stabilization for Aided Inertial Navigation Systems, U.S. Pub. No. 2008/0319667 A1; inventors Wesley J. Hawkinson & Tom Rolfer; assignee Honeywell International Inc.; filed 2008‑03‑31 (US 12/059,837); published 2008‑12‑25; granted as US 8,019,542 B2 (2011‑09‑13). EP 1983304 B1.
URL: https://patents.google.com/patent/US20080319667A1/en; grant text https://patentimages.storage.googleapis.com/94/ce/bd/e80bdf6c29eda9/US8019542.pdf.

Description: An aided inertial navigation system 100 on a vehicle 105 includes an IMU 110 mounted on a rotational device 200 that is driven by a servomotor (precision gear head per claim 20) to rotate the IMU about a horizontal axis x₁ from a first orientation to a second orientation. The system calibrates ("aligns") the horizontal-sensing and off-horizontal-sensing elements of the IMU based on horizontal aiding measurements from sensors 130 (horizontal velocity sensor 148, horizontal position sensor 142, and optionally GPS/odometer/Doppler radar/sonar 147), using a Kalman filter 320 and a sensor compensator 305, with a software control component 330 controlling the rotation of the rotational device.

§ 102 relevance to the '647 claims:

• Claim 1 — candidate § 102/§ 103 reference (medium strength). This is structurally the closest to the '647 "in-system" concept: a gyroscope-bearing unit is rotated in place, on the vehicle, by a powered rotational device, and calibration is performed against an independent measurement (aiding sensors), with the processor correcting the readings. Points of divergence from claim 1: the aiding sensors (velocity/position) are not framed as measuring the "controlled range of rotation" per se, and there is no teaching of a rotatable coupling selectable between rotatable and non-rotatable states for working vs. calibration.
• Claim 5 — relevant. Expressly discloses a servomotor rotating the sensor unit (electric actuator), directly reading on claim 5's "electric motor."
• Claims 2–4 not anticipated.

---

Reference 3 — US 2009/0093984 A1 (Samsung Electronics)

Full citation: Method and Apparatus for Calibrating Gyro-Sensor, U.S. Pub. No. 2009/0093984 A1; assignee Samsung Electronics Co., Ltd.; published 2009‑04‑09; priority to KR 10‑2005‑0087746 (2005‑09‑21) and KR 10‑2006‑0075221 (2006‑08‑09) (US filing corresponds to the 2006 KR priority year).
URL: https://patents.google.com/patent/US20090093984A1/en; PDF https://patentimages.storage.googleapis.com/a6/67/8d/82a04409a365c3/US20090093984A1.pdf.

Description: Calibrates a gyro-sensor by measuring the angular velocity of a moving body and the average gyro output while the body rotates, deriving a characteristic equation/curve (n-th-degree polynomial) of the gyro, storing it (optionally at multiple temperatures), and using it to convert the gyro output into angular velocity for real-time calibration (operation module 200, storage 300, calibration module 400). Notably, the reference rotation/angular velocity is obtained using infrared transmission and reception units ("optical") or hole/magnetic-field sensors, or by installing the gyro-sensor at the rotation axis of a rotation motor (¶¶[0038]–[0040]).

§ 102 relevance to the '647 claims:

• Claim 1 — § 103 reference, not a clean § 102. Discloses rotating a gyro and comparing its output to an independent angular-velocity measurement, then computing a calibration equation and correcting readings. But the "moving body" is rotated as a whole (or the gyro is placed on a motor axis) — there is no rotatable coupling selectable between two states and no autonomous-path robotic platform.
• Claim 2 (non-gyro sensor = optical sensor/proximity/encoder): relevant — the infrared (optical) units and hole sensors used to measure rotation map to claim 2's optical/proximity sensors.
• Claim 5 (rotation actuated by motor): relevant — "installing a gyro-sensor at a rotation axis of a rotation motor."

---

Reference 4 — US 2008/0249732 A1 (Samsung Electronics) / US 8,135,562 B2

Full citation: System, Method and Medium Calibrating Gyrosensors of Mobile Robots, U.S. Pub. No. 2008/0249732 A1; assignee Samsung Electronics Co., Ltd.; published 2008‑10‑09; priority 2007‑04‑04; granted as US 8,135,562 B2 (2012‑03‑13).
URL: https://patents.google.com/patent/US20080249732A1/en.

Description: Calibrates a mobile-robot gyrosensor by rotating the mobile robot in a predetermined velocity pattern, capturing image data of a fixed environment (e.g., a ceiling) with a camera, computing a rotation angle/angular velocities, deriving reference angular velocities via encoder parameters and kinematic modeling, and then calculating a scale factor (calibration) mapping gyro raw data to reference angular velocities (optionally with weighted least squares to handle bias drift). The stated advantage is doing this while the robot moves, without an expensive rotary jig (distinguishing the prior FIG-1 rotary-table approach).

§ 102 relevance to the '647 claims:

• Claim 1 — § 103 reference, not § 102. Discloses a mobile robot's gyro calibrated against an independent reference (camera/encoder) and a scale factor applied to the gyro. Critical difference from '647 claim 1: here the whole robot rotates; there is no in-place rotation of the gyroscope relative to the platform via a selectable rotatable coupling. Also not an autonomous defined-path mower/cleaner context.
• Claim 2 (non-gyro sensor = optical sensor or encoder): relevant — camera (optical) + encoder are the reference sensors.

---

Reference 5 — US 2011/0172820 A1 (Kim et al.) / US 9,222,801 B2

Full citation: Apparatus and Method for Correcting Error of Gyro Sensor in Mobile Robot, U.S. Pub. No. 2011/0172820 A1; inventors Do‑Hyung Kim, Hak‑Young Chung, Jin‑Woo Song, Guen‑Rok Ryu; filed 2009‑06‑26 (US 13/002,331, PCT/KR2009/003486); published 2011‑07‑14; priority KR 10‑2008‑0063541 (2008‑07‑01); granted US 9,222,801 B2.
URLs: https://patents.google.com/patent/US20110172820A1/en; https://www.freepatentsonline.com/y2011/0172820.html.

Description: A gyro bias-correction scheme for a mobile robot: a state-determination unit decides whether the robot is stationary (from the deviation of gyro output values), a bias-calculation unit computes a bias value from the gyro outputs while stationary, and the bias is used to compute angular velocity while moving (continuously updated, e.g., E[k+1] = (k·E[k]+v_{k+1})/(k+1)). There is no external rotation of the gyro.

§ 102 relevance to the '647 claims:

• Claim 1 — § 103 reference at most; does not anticipate. It has no "controlled rotation of the gyroscope relative to the platform" and no non-gyroscopic rotation-measuring sensor. It is relevant only to the generic "calculate a calibration factor and modify the readings with a processor on a mobile robot" concepts.
• Claim 4 (processor physically mounted to the platform) — the correction unit is carried on the mobile robot; relevant as background.

---

Reference 6 — US 2012/0245850 A1 (Jon Bang)

Full citation: Azimuth Initialization and Calibration of Wellbore Surveying Gyroscopic and Inertial Instruments by Means of an External Navigation System, U.S. Pub. No. 2012/0245850 A1; inventor Jon Bang; published 2012‑09‑27; priority 2009‑10‑30.
URL: https://patents.google.com/patent/US20120245850A1/en.

Description (from the face-of-patent listing and general knowledge of this family — flagged as lower-confidence; see caveats): Concerned with initializing the azimuth and calibrating gyroscopic/inertial survey instruments used in wellbore (borehole) surveying by referencing an external navigation system (a surface/independent navigation reference) rather than by a fixed laboratory setup. It is a different technical field (downhole surveying), but shares the abstract idea of calibrating a gyroscopic instrument against an external measurement.

§ 102 relevance to the '647 claims:

• Claim 1 — not anticipated. Different field; no autonomous robotic base platform with a driving mechanism, and no rotatable coupling selectable between rotatable/non-rotatable states. At most a § 103 secondary reference showing that "calibrating a gyro against an external reference" was known.

---

3. Bottom line — anticipation and relevance

• Most relevant prior art: US 8,548,766 B2 (Judd / Honeywell) — cited on the face of '647 and discussed (as its publication US 2011/0066395 A1) in the '647 background. It is the only reference that rotates a gyroscope assembly through a known/measured arc and computes gain and bias from a comparison against independent (accelerometer/magnetometer) measurements — i.e., the conceptual core of '647 claim 1.
• Second most relevant: US 2008/0319667 A1 / US 8,019,542 (Honeywell) — an in-vehicle rotational device + servomotor that rotates an IMU in place and calibrates it against aiding sensors; the strongest single-reference teaching of rotating the gyro relative to its host platform.
• No cited reference appears to fully anticipate claim 1 individually. The elements that distinguish '647 claim 1 from each of these references are the combination of (a) an autonomous robotic base platform having a driving mechanism and capable of autonomously traversing a defined path, (b) a rotatable coupling to that platform selectable between a rotatable (calibration) and non-rotatable (working) state, and (c) the controlled range being either predetermined or measured by a non-gyroscopic sensor. Judd/magnetometer-based and Honeywell systems supply (c)-type measurement but not the autonomous-platform + selectable-coupling combination. Accordingly the citations function primarily as § 103 obviousness references, with US 8,548,766 the best candidate for a § 102 attack on claim 1 (and, were claim 1 met, it would also touch claims 4 and 5).
• Claims 2 and 5 are the dependent claims most exposed to these references: claim 2 (optical/proximity/encoder non-gyroscopic sensor) reads onto the Samsung '984 optical/hole-sensor teaching and the Samsung '732 camera/encoder teaching; claim 5 (electric motor/solenoid/linear actuator) reads onto the Honeywell '667 servomotor and the Samsung '984 rotation-motor teaching. Claim 3 (physical stop or magnetic stop) is not taught by any of the six cited references.

---

4. Caveats and confidence

• High confidence: the identity and bibliographic data of the reference patents themselves (full text/PDFs retrieved for references 1–5), the '647 claim text, and the mapping of claim limitations described above.
• Lower confidence — Reference 6 (Jon Bang): my search hit the tool step-limit before I could retrieve this document's full text; my description is based on the face-of-patent listing and general knowledge of the family and is flagged as unverified. Treat its § 102 analysis as provisional.
• Lower confidence — exact US filing date of US 2009/0093984 A1: I confirmed the KR priority dates (2005‑09‑21 and 2006‑08‑09) and the publication date (2009‑04‑09); I did not independently confirm the exact U.S. filing date, so I did not assert a precise figure.
• Not auto-corrected: I preserved the '647 title typo ("gyrosopic"), treated DE10132647 and other numerically similar documents as distinct identifiers (they are not the target), and reproduced the citation list exactly as printed on US 10,132,647.
• This is a prior-art/patentability analysis only; it is not a legal conclusion on validity, and "anticipation" is assessed at the level of claim-limitation disclosure, not litigation-grade invalidity.

I'll ground this analysis in the actual prior-art references cited on the patent's face. Let me verify what each reference discloses.

Two confirmations already: US 8,548,766 B2 (Honeywell, inventor Tom Judd) is the "US 2011/0066395 to Judd" discussed in the '647 specification's own Background. Let me characterize the remaining two cited references.

Obviousness Analysis — U.S. Patent No. 10,132,647 under 35 U.S.C. § 103

Patent: US 10,132,647 B2 ("Methods and apparatus for increasing accuracy and reliability of gyrosopic sensors" [sic — typo in the printed title, reproduced literally])
Filed: 2014-10-24 · Priority: 2013-10-24 (Prov. 61/894,981) · Granted: 2018-11-20 · Assignee: MTD Products Inc.
Claims analyzed: 1–5 (1 independent)

---

0. Corrections and refinements to the earlier sections

Flag — identification of the Judd reference (refinement, not contradiction). The previously generated Patent Summary listed "US 2011/0066395-type Judd family" and "US 8,548,766 B2 (Honeywell)" as two separate prior-art entries in the "Cited prior art of note" row. They are one and the same family. My source check confirms: US 8,548,766 B2 issued 2013-10-01 from Application No. 12/558,871, filed 2009-09-14, inventor Tom Judd, assignee Honeywell International Inc., and it published as US 2011/0066395 A1 on 2011-03-17 (source: US8548766B2 PDF cover page; Google Patents US8548766). So the reference the '647 specification attacks by name in its Background ("United States Patent Publication 2011/0066395 to Judd") is itself one of the six references cited on the face of the '647 patent. That is an important § 103 fact: the primary reference is not merely analogous art — the patentee conceded it is prior art.

Scope discipline. The six references cited on the '647 face are:

1. US 2009/0093984 A1 — Samsung — Method and apparatus for calibrating gyro-sensor (pub. 2009-04-09)
2. US 2008/0249732 A1 — Samsung — System, method and medium calibrating gyrosensors of mobile robots (pub. 2008-10-09)
3. US 2008/0319667 A1 — Honeywell (Hawkinson et al.) — Heading stabilization for aided inertial navigation systems (pub. 2008-12-25; granted as US 8,019,542 B2)
4. US 2011/0172820 A1 — Kim Do-Hyung — Apparatus and method for correcting error of gyro sensor in mobile robot (pub. 2011-07-14; granted as US 9,222,801 B2)
5. US 8,548,766 B2 — Honeywell (Judd) — Systems and methods for gyroscope calibration (granted 2013-10-01; pub. US 2011/0066395 A1)
6. US 2012/0245850 A1 — Jon Bang — Azimuth initialization and calibration of wellbore surveying gyroscopic and inertial instruments by means of an external navigation system (pub. 2012-09-27)

I verified the disclosure content of items 1–5 against primary text. Item 6 (Bang) I was unable to verify beyond its title/wrapper, and I will not attribute specific disclosures to it; treat it as cumulative. Every one of the six published before 2013-10-24 and therefore qualifies as § 102(a)(1)/(a)(2) art (all are also examiner-cited, so no § 102(b) provenance risk).

---

1. The legal framework I am applying

Under Graham v. John Deere and KSR Int'l v. Teleflex, the question is whether the differences between claim 1 and the prior art would have been obvious to a POSITA at the 2013 priority date, considering (a) the scope/content of the art, (b) the differences, (c) the level of ordinary skill, and (d) objective indicia. KSR permits combination where the references are from the same field, address the same problem, and the combination yields no more than predictable results; an express teaching, suggestion, or motivation is helpful but not required.

POSITA definition I am applying: a person with a bachelor's degree in mechanical/electrical engineering (or equivalent) and 2–4 years of experience in inertial-sensor integration for mobile robots or inertial navigation — i.e., someone who routinely mounts MEMS gyros on robotic chassis, reads datasheets, and knows what a rotary-table calibration is and why it is undesirable in the field.

Critical threshold point: Claim 1 is not anticipated by any single reference. Judd rotates a sensor platform but never describes that platform as a self-propelled autonomous robot; Samsung '732 rotates the whole robot, not a sensor sub-platform relative to the robot; Hawkinson rotates an IMU inside a vehicle but does not disclose a gyro gain/offset calibration loop of the '647 type; Kim '820 corrects bias but never rotates anything. The vulnerability is § 103, not § 102.

---

2. Claim 1 element-by-element mapping

Claim 1 breaks into eight limitations. Here is how they map:

Result: every limitation of claim 1 has express disclosure in the cited art; the only work a defendant must do is supply (1) the reason to put the Judd/Hawkinson rotating-sensor architecture into a Samsung-type autonomous robot, and (2) the reason for the two-state coupling.

---

3. Ground 1 (primary): Judd + Hawkinson + Samsung '732

3.1 What each reference contributes

Judd (US 8,548,766 B2 = US 2011/0066395 A1) supplies limitations A, B, D, F, G, H, and the entire calibration algorithm. Verified language: "An exemplary embodiment rotates a sensor platform with at least one gyroscope thereon, and with at least one of an accelerometer and a magnetometer thereon, determines a measured rotation based upon the rotation sensed by the at least one gyroscope, derives a rotation based upon the changes in the earth's gravitational field sensed by the at least one accelerometer, derives a rotation based on the changes in the earth's magnetic field sensed by at least one magnetometer, and determines at least one of a compensation gain and a compensation bias." Judd also supplies the motivation/field-of-use honesty: bias and gain "may not remain constant over time," so "there may be a need to periodically recalibrate the gyroscope in the field"; the object is to do this "in the field where access to specialized calibration tools is limited," with the processor system 110 and memory 112 on the sensor platform 102.

Hawkinson (US 8,019,542 = US 2008/0319667 A1) supplies the mechanism: a rotational device 200 mounted in a vehicle 105, driven by a servomotor 260 through a precision gear head 265, with a solenoid 270 controlling rotation, indexing pins in holes 271 forming a pair of stops to bound the rotation to a predetermined angle θr, all under a processor/software control component 330 that "periodically sends control signals … to initiate a rotation," followed by a back-rotation to the original orientation. Critically, Hawkinson's FIG. 5 element set is literally claim 5's list (servomotor = electric motor; solenoid) and claim 3's physical stop.

Samsung '732 (US 2008/0249732 A1) supplies the platform (limitation C) and the non-gyroscopic reference sensor (limitation D(ii)): a self-propelled mobile cleaning robot with motor 151, drive wheels, gyrosensor 110, camera 140, and encoder 152; the stated objects are "calibrating a gyrosensor of a mobile robot with enhanced accuracy and reduced costs" and "while the mobile robot is moving."

3.2 Motivation to combine (the crux — and the patentee supplies it)

1. Express motivation written into the '647 patent's own Background. The spec states Judd's approach is inadequate because it "require[s] human intervention and removal of the gyroscope" and that "there remains a need for alternatives," and that the invention "allows one to improve the accuracy of the sensor over its lifecycle without having to disassemble portions of the robotic platform." Under KSR, a motivation stated in the challenged patent's own specification is the strongest possible TSM evidence. The patentee cannot call the combination non-obvious when it recited the motivation for the combination.
2. Same field, same problem, overlapping inventors/assignee. Judd and Hawkinson are both Honeywell inertial-navigation in-field-calibration patents; Samsung '732 and '984 are both mobile-robot gyro-calibration patents. All are classified in G01C25/00-series calibration art — the same art as the '647 (G01C25/005).
3. Samsung '732 expressly condemns the removal-based approach. It teaches that the conventional rotary-table jig (its FIG. 1) "requires an expensive jig 13" and "calibration cannot be performed in real time … since the mobile robot 10 has to be first loaded into the jig before calibration." That is a direct teaching that motivates an in-system rotating calibration in a mobile robot.
4. Predictable results. Combining a known motorized, stop-bounded rotational stage (Hawkinson) with a known field gyro gain/bias algorithm (Judd) and mounting it on a known autonomous robot (Samsung '732) yields exactly the expected result: calibration without removal. There is no unexpected mechanism, no new physical principle, no criticality of any parameter.
5. Design incentive / interchangeability. The '647 spec itself lists the actuator options as equivalents ("a brushed or brushless motor (e.g., a stepper motor), a servo, and/or a solenoid or linear actuator, just to name a few"), and says a single device can serve as manipulator, locking member, stop, and position sensor. That is a patentee admission that the claimed mechanical genus is a set of known equivalents, which collapses any argument that the specific claim-5 actuator is inventive.

3.3 The two-state coupling (limitation E) — the most contestable element

This is the element a patent owner will fight hardest on. The attack:

• Hawkinson's rotational device must hold the IMU in the first orientation (nominally rigid) between calibration cycles — the IMU stays in orientation 170 for 30 seconds to 10 minutes at a time and the system only rotates it periodically; the indexing pins/holes 271 mechanically define and hold the two positions. Functionally, Hawkinson discloses a coupling that is non-rotatable in the operating state and rotatable in the calibration state.
• Hawkinson's solenoid 270 is expressly a state-selecting actuator, and solenoids are the canonical locking element. Note that the '647 specification lists "a solenoid, a linear actuator, and/or a friction brake" as its own locking-member examples.
• The '647 claim language "selectable between two states" is, in context, functional and broad — it covers any structure that can be either free-to-rotate or held. Given the '647 spec's own statement that a stepper motor may double as a locking member, the patentee will have a difficult time reaching back to narrow the term to a dedicated pin-in-slot lock without creating a § 112 written-description problem of its own.

3.4 Ground 1 in concise form

It would have been obvious to modify Judd's field-calibration system by mounting Judd's rotatable sensor platform on Samsung '732's autonomously navigating mobile robot and by actuating the platform with Hawkinson's servomotor/gearhead/solenoid-and-indexing-stop assembly, because (i) Judd itself identifies the need for in-field periodic recalibration without specialized tools; (ii) Samsung '732 identifies the cost and downtime of the removal-based rotary-table calibration of mobile robots; (iii) Hawkinson teaches precisely the motorized, stop-bounded rotating stage needed to perform Judd's calibration rotation inside a vehicle; and (iv) the combination produces no more than the predictable result of calibrating the robot's own gyroscope while it is installed.

---

4. Ground 2 (robot-centric): Samsung '732 + Samsung '984 + Kim '820

This ground uses only mobile-robot art and is attractive because it supplies limitation C intrinsically — the platform is unambiguously an autonomous robot in all three references.

• '732 — autonomous mobile cleaning robot; gyro calibration in situ; camera (optical) and encoder as non-gyroscopic reference sensors; "rotating the mobile robot in a predetermined velocity pattern" = a controlled range of rotation; scale-factor calculation and application. Supplies limitations A, C, D(ii), F, G, H.
• '984 — calibrating a gyro-sensor "using data … obtained by measuring an angular velocity and a gyro output value of a moving body equipped with the gyro-sensor" while it rotates. Two key teachings:
  • The angular velocity reference is obtained by non-gyroscopic sensors measuring the rotation itself: an infrared transmission/reception unit (optical/proximity) and hole sensors sensing a magnetic field from a permanent magnet (magnetic/proximity; the permanent magnet also functions as a magnetic index). This maps directly to claim 1 D(ii) and claim 2.
  • ¶[0040]: "the tuple set … may also be obtained by installing a gyro-sensor at a rotation axis of a rotation motor in the moving body" — i.e., rotating the gyro itself with a motor aboard the autonomous robot. This maps to claim 1(A) and claim 5.
  • The specification notes that "no general technique for calibrating the gyrosensor is available, other than techniques using a data sheet or a rotary table," and that rotary-table/data-sheet approaches "fail[] to take into consideration the aging of the gyrosensor or a change in scale factor according to an external temperature change" — an express motivation to move the rotation on-board.
• Kim '820 — "correcting, in real time, a bias value using output values of a gyro sensor" in a mobile robot, with a state determination unit and bias calculation unit in the robot's body, and the corrected angular velocity computed as gyro output minus bias. Supplies limitations F/G/H in the robot context and the processor-on-platform teaching (claim 4).

Motivation: '732's own object is to enhance accuracy and reduce cost; '984's object is to calibrate using data measured aboard a moving body to avoid the costly and ineffective data-sheet/rotary-table route, and to account for temperature drift; Kim '820's object is to reduce directional-angle error in real time. All three are the same field, address the same deficiency (aging/temperature-induced gyro error in service), and are combinable by simple substitution of the calibrating rotation source.

---

5. Ground 3 (mechanism-light): Judd + Samsung '984

Because Judd already teaches the gain/bias algorithm, rotation, and independent reference sensing, the only things it lacks are (a) the robot platform and (b) the coupling state-selectability. '984 supplies both the robot-platform context and the on-axis motor rotation, plus the magnetic/optical index sensors. This ground is narrower than Ground 1 but is useful as a fallback if the patentee successfully distinguishes Hawkinson on the "vehicle" vs. "autonomous robotic platform" axis — '984 is squarely a home-base robot (FIG. 7: "a moving body 710 … rotates at a constant angular velocity on a home base 720"), which also anticipates the '647's own "park at a recharging station" embodiment.

---

6. Dependent claims 2–5

These are, individually and collectively, the weakest part of the patent. Each is a single-reference or single-reference-plus-routine-choice case.

There is no dependent claim that adds a limitation not disclosed in the four primary references. If claim 1 falls, claims 2–5 fall with it — a defendant should not expect to salvage anything by attacking claim 1 alone.

Note the § 112 wrinkle: claim 1 requires the coupling to be "selectable between two states," while claim 5 says the rotation is actuated by a motor/solenoid. If the patentee argues the motor is the state-selector (a natural reading, and one the spec invites), claim 5 becomes a near-verbatim reading on Hawkinson's servomotor-plus-solenoid assembly.

---

7. Objective indicia (§ 103(d)) — likely weak

The patent's own summary touts three benefits: (1) in-situ calibration reduces cost/time vs. removal-based calibration; (2) enables more frequent calibrations; (3) "enables the use of less expensive components while maintaining an acceptable accuracy level." Under KSR and In re Kao, these are inherent consequences of the known approach already promised by the prior art:

• '732 promises "enhanced accuracy and reduced costs" and calibrating "while the mobile robot is moving."
• Judd promises in-field recalibration without specialized tools.
• '984 explicitly notes that calibrating each of "a considerable number of gyro-sensors may involve an increase in the manufacturing cost" and adopts on-board calibration precisely to obtain cost-effective accuracy.
• '924/'801 (Kim) promises "correcting an error of the gyro sensor in real time while the mobile robot is moving."

Where the asserted advantage was expected by the prior art (here, expressly), it is not evidence of non-obviousness. The one argument with any traction would be a commercial-success / licensing story tied to MTD's robotic mower line, but nexus would be difficult to establish given that the benefits are technical and generic, and given that the '647's own claims cover only the calibration apparatus/method, not the mower.

---

8. Counterarguments the patent owner will press — and how they fare

---

9. Bottom-line assessment

Claim 1 is vulnerable, not dead. Three independent combinations reach it, and the strongest single fact is that the patentee concedes Judd (US 2011/0066395 = US 8,548,766 B2) is prior art and then states, in the Background, the exact reason to combine it with a mobile-robot platform. That converts the usual § 103 fight over "motivation" into a fight over two narrow mechanical limitations — the robot platform and the two-state coupling — both of which have close art:

• Ground 1 (recommended): Judd + Hawkinson + Samsung '732 — cleanest, because Hawkinson literally supplies claim 3's stops and claim 5's servomotor + solenoid, and Judd supplies claims 1's entire calibration loop.
• Ground 2 (robust alternative / hedge): Samsung '732 + Samsung '984 + Kim '820 — all mobile-robot art, so no analogous-art fight, at the cost of needing to combine two rotations (robot-body and sensor-motor).
• Ground 3 (fallback): Judd + '984 — simplest, narrowest.

Probability framing (my judgment, explicitly a judgment, not a verified outcome): on the record as it stands, claim 1 is more likely than not to be found obvious if a competent petition is filed with a strong declaration and a well-argued motivation, and claims 2–5 are very likely to fall with it (claim 5 is essentially reading on Hawkinson's servomotor/solenoid alone). The residual risk is procedural, not technical: (a) a Board that demands a tighter nexus between the prior art's mobile robots and the '647's self-propelled platform, and (b) § 325(d) discretion if the same Judd reference was squarely before the examiner — which it was, since it is on the face of the patent. § 325(d) is the biggest practical obstacle to Ground 1 and Ground 3, and it is the best reason to lead with the '732 + '984 + Kim Ground 2 combination, which frames the art differently (the examiner cited these references individually, not as a three-way robot-calibration combination) and includes at least one reference ('984) whose on-axis-motor teaching is the most direct answer to the two-state coupling and claim 5.

What a petitioner must build: (1) the full prosecution history, to see what the applicant said about Judd and about "removal," and to assess § 325(d) and estoppel; (2) a POSITA declaration explaining why a robot-chassis engineer would mount a motorized, stop-bounded rotatable sensor stage on an autonomous mower — the "design incentive" prong; (3) an express claim chart for claims 2–5 showing single-reference reads; and (4) a construction position on "selectable between two states" that captures solenoid/pin/indexing-stop mechanisms, supported by the '647 specification's own statements about stepper motors doubling as locking members and stops.

---

Verification note

Claim text, bibliographic data, and the six cited references are taken from the authoritative patent text supplied (Google Patents, fetched 2026-09-24). The disclosure content attributed to US 2008/0249732 A1, US 2009/0093984 A1, US 2011/0172820 A1, US 2008/0319667 A1 (US 8,019,542 B2), and US 8,548,766 B2 (US 2011/0066395 A1) is drawn from primary patent text retrieved during this analysis and quoted above. US 2012/0245850 A1 (Jon Bang) is characterized only by its title and I make no disclosure-specific assertion about it. No claim of anticipation is made anywhere in this memo; all conclusions are under § 103. The probability framing in § 9 is my analytical judgment, not a verified outcome of any proceeding — there has been no IPR, PGR, or litigation on this patent.