US 11391596

I'll search for authoritative information on this specific patent number, including any litigation or CAFC docket activity.

Let me run a couple more targeted searches to confirm there's no litigation or CAFC activity for this specific patent.

US 11,391,596 — Verification Summary

Search scope note: I searched for the literal number "11391596" / "11,391,596." I did not find any entry on a CAFC 2026 docket, and no PTAB proceeding (IPR/PGR) naming this patent or its owner. Important disambiguation: searches returned a separate, unrelated patent numbered 11,319,596 (Natera/NeoGenomics, IPR2025‑00455, biological-sample cancer monitoring). That is a different patent number and is not US 11,391,596; I have not conflated them. I also could not query USPTO PatentCenter / CAFC docket systems directly, so the "no litigation" finding is based on public web sources and should be treated as a preliminary negative rather than an authoritative docket clearance.

---

Bibliographic Data (per the patent record)

Abstract (as issued): A system and method for converging mediated reality (MR) positioning data and geographic positioning data. The method includes receiving three or more geographic coordinate points and three or more MR coordinate points, each associated with points in time; determining a geographic trend line and an MR trend line; determining a correction angle as the angle between the two trend lines; and outputting the correction angle.

---

Plain-Language Overview of the Independent Claims

Claim 1 — Method (treatment of the two data streams):

1. Receive ≥3 MR coordinate points, each from a different point in time.
2. Receive ≥3 geographic coordinate points (e.g., GNSS), each from a different point in time.
3. Time-match the streams: for each geographic point, find which MR point is nearest in time and associate them. (Note the direction: the geographic point is matched to the nearest MR point — the reverse of how the specification's summary frames it.)
4. Compute a geographic trend line by data fitting over the ≥3 geographic points.
5. Compute an MR trend line by data fitting over the ≥3 MR points.
6. Compute a correction angle = angle between the two trend lines.
7. Output the correction angle.

Claim 9 — System: The same seven-step concept implemented as processor-executed modules — an MR module (receives ≥3 time-stamped MR points), a receiver module (receives ≥3 time-stamped geographic points and performs the nearest-in-time association for each), a positioning module (data-fits both trend lines), and a correction module (computes and outputs the angle).

Claim 17 — MR device: An augmented/mixed-reality device (processor, memory, geographic receiver, display, camera). The MR module derives ≥3 MR coordinate points from camera images (time-stamped), the receiver module takes ≥3 time-stamped geographic points and performs the nearest-in-time association, the positioning module data-fits both trend lines, and the correction module computes and outputs the correction angle. This claim is the "all-in-one device" version.

Key dependent claims: cl. 2–3 (rotate the virtual viewpoint by the correction angle; overlay virtual geospatial objects on the real-scene image); cl. 4/12/20 (MR data = photogrammetry, accelerometer, inertial, and/or gyroscope data); cl. 5/13 (iterate as new points arrive); cl. 6/14 (fixed-size rolling window of the n most recent points); cl. 7/15 (regression analysis, e.g., least squares as disclosed); cl. 8/16 (determine a geographic path and an MR path).

---

Technical Point of Novelty (as stated in the specification)

The patent frames itself against a prior "vector approach" that reconciles only the two most recent geographic and MR points. The stated deficiency is that this inherits the randomness/error of GNSS (which yields a position range, not a point — see FIG. 7), causing under- or over-correction of heading. The claimed improvement is to fit trend lines through ≥3 raw past points on each stream, which smooths/averages the noise, and to correct heading by the angle between the trend lines. The specification also notes co-location of the most recent points (translation) as an optional convenience for user comprehension, and contemplates a cloud/server intermediary so that unsupported MR headsets (e.g., HoloLens) need not be directly bridged to a GNSS receiver.

---

Notable Observations / Uncertainties

• Claim-vs.-specification divergence: The granted claim 1 recites the association step one-directionally and in the reverse orientation ("for each of the geographic coordinate points, determining which of the MR coordinate points are nearest in time"), and expressly requires "data fitting." The abstract and the Summary section describe the association in the opposite framing ("each of the MR coordinate points associated with one of the points in time..."). This is a claim-drafting/prosecution nuance worth noting if claim scope is in question — the granted claims, not the abstract, are operative.
• Terminology: the patent uses "MR" (mediated reality) broadly to cover AR, mixed reality, modulated reality, and holograms, and "geographic positioning" to cover GNSS, GPS, GLONASS, Galileo, Beidou, QZSS, WPS, cellular positioning, and RTK.
• Legal-status caveat: the "Active / expires 2040‑08‑11" entry comes from Google Patents and is expressly stated there to be an assumption, not a legal conclusion. Verifying term/adjustment would require PatentCenter or the printed patent's term-adjustment notice.
• Litigation/CAFC: no CAFC 2026 docket, district-court filing, or PTAB petition involving US 11,391,596 was located in public sources as of this search. I do not have authoritative docket confirmation, so treat this as "nothing found," not "nothing exists."

I'll search for litigation involving US patent 11391596.

Initial results aren't specific. Let me run more targeted searches against litigation-tracking sources.

Let me check additional litigation databases and the EP family member for any disputes.

Search Result: No Litigation Found for US 11,391,596

I ran targeted searches for litigation involving US Patent No. 11,391,596 ("System and method for converging mediated reality positioning data and geographic positioning data," assigned to Vgis Inc.) across litigation-oriented sources, including Unified Patents' litigation/PTAB portal, PTAB filings documents, RPX/Docket Alarm litigation documents, CourtListener, and general patent-litigation coverage.

I found no district court litigation, ITC investigation, PTAB proceeding (IPR/PGR/CBM), or Federal Circuit appeal involving US 11,391,596. No plaintiff/defendant pairs, case numbers, jurisdictions, or filing dates can be reported because no such case was located.

What the record shows (verified)

• Patent: US 11,391,596 B2, Application No. 16/860,479, filed 2020-04-28, granted 2022-07-19, priority date 2020-04-28, current assignee Vgis Inc., status Active with adjusted expiration 2040-08-11. Source: https://patents.google.com/patent/US11391596/en
• Family / foreign counterparts: WO2021217244A1 (PCT/CA2021/050502, filed 2021-04-14) and EP3956690B1 (EP21797368.4, filed 2021-04-14).
• Prosecution citations (prior art cited, not litigation): e.g., US 5,266,958 A; US 2004/0105573 A1; US 8,712,686 B2; US 2016/0097861 A1; US 2019/0197982 A1; US 2019/0251373 A1; US 10,432,888 B2; WO2020006299A1; US 10,679,509 B1; US 2020/0327732 A1; US 2021/0097760 A1; US 2021/0097282 A1; US 2021/0343087 A1.
• Later-citing documents (patents citing this family, not litigation): US 11,847,750 B2 (Niantic); US 12,211,223 B2 (Vgis Inc.); CN116399310B; CN119693590B.

Caveats and confidence

• Confidence that the assertion "no litigation exists" is complete: moderate. Search returned no litigation hits, but litigation databases (PACER, Docket Alarm, RPX) are not fully indexed by general web search, and a case could exist under a party name I did not query (e.g., a plaintiff that acquired the patent, or an NPE).
• EPO opposition check inconclusive. I attempted to check whether EP3956690B1 is under EPO opposition, but the search step limit was reached before I could confirm. That is a post-grant administrative challenge, not litigation, but it is the most plausible place a dispute over this family could have arisen.
• No ITC Section 337 investigation was found, but this was not separately and exhaustively verified.

Recommended next steps for a definitive answer

To confirm the negative with higher confidence, I would query, by patent number:

1. PACER / CourtListener RECAP — docket search for "11,391,596" and "11391596."
2. Docket Alarm / RPX Insight — patent-level litigation lookup.
3. Unified Patents Portal — PTAB case list (portal.unifiedpatents.com/litigation/caselist and /ptab).
4. USPTO Patent Trial and Appeal Board — search for the patent number in IPR/PGR.
5. EPO Register / European Patent Register — check EP3956690B1 for opposition status.

If you can grant additional search steps, I can run these specific checks and report back.

I'll verify the structured-data default against live sources before reporting.

Let me run a few more targeted checks before concluding.

PTAB / AIA Trial Proceedings — US 11,391,596 B2

Source of the proceedings list: the structured "PTAB proceedings on file" block supplied with this task (USPTO Open Data Portal ingest), which states that the ODP API returns no AIA trial proceedings for this patent. I independently searched PTAB E2E-facing materials, Google Patents litigation/trial data, and general web sources for "11391596" IPR, Vgis v. / v. Vgis IPR-PGR petitions, and petitioner-side filings; nothing surfaced. Treat the absence as a verified negative, with the caveat below about ODP indexing lag.

---

Proceedings overview

Total AIA trial proceedings on file: 0 — 0 active, 0 claims invalidated, 0 claims sustained, 0 settled, 0 institution denials. All 20 claims of US 11,391,596 stand as issued and completely untested at the PTAB, so a defendant has no ready-made invalidity judgment to point to — but equally, the patent owner has no IPR-sustained claims to wave, and every prior-art ground a defendant can find remains available because no § 315(e)(2) estoppel has attached to anyone.

---

Per-proceeding detail

None to report. There is no IPR, PGR, or CBM proceeding to analyze, and I will not invent a docket number, panel, or FWD to fill this section. Specifically:

• No petitioner (no competitor, no defensive aggregator such as Unified Patents, RPX, or a litigation defendant) has filed against this patent as far as public sources show.
• No institution decision, no Final Written Decision, no termination, and no Federal Circuit appeal exist for US 11,391,596.
• Because there is no FWD, no claim of this patent is canceled, and no claim has been adjudicated patentable by the Board. The claims should not be described as "survived an IPR" — surviving requires a trial that never happened.

One related observation from the patent's own record, which is not a PTAB proceeding but is worth flagging: the Google Patents page for US 11,391,596 lists 13 examiner/third-party citations under the Citations table (e.g., US 5,266,958; US 2004/0105573; US 8,712,686; US 2019/0197982 (Magic Leap); US 10,432,888 (Trimble); WO 2020/006299 (Niantic); US 2021/0343087 (Magic Leap)). These are prosecution-cited references, not grounds in any AIA trial, and carry no estoppel.

---

Strategic summary

Claim status of US 11,391,596. Every claim — independent claims 1, 9, and 17 and all dependent claims (2–8, 10–16, 18–20) — is UNTESTED and in force. The patent is active, with a legal status of "Active" and an adjusted expiration listed as 2040-08-11 (which reflects a substantial PTA). There is no IPR-narrowed claim set, no certificate of cancellation, and no reissue/reexam narrowing in the record I could find. Practically, that means the full claim scope as granted is what a plaintiff would assert today.

Estoppel landscape. With zero AIA trials, § 315(e)(2) estoppel is a non-issue — no petitioner, and no privy of any petitioner, is barred from raising any ground. All § 102/§ 103 prior-art grounds (printed publications, patents, and public-use/system art usable in district court but not in IPR) remain on the table for a defendant. The practical constraints a defendant still faces are procedural, not estoppel-based: the § 315(b) one-year bar (running from service of a complaint alleging infringement) if the defendant wants to file its own IPR, and the Board's § 314(a)/§ 325(d) discretionary-denial practice (Fintiv-style weighing, as seen in contemporaneous Board decisions) if a parallel district court case is advanced.

Timing note for a would-be petitioner. This is a first-inventor-to-file patent (filed 2020-04-28, more than 8 years after March 16, 2013), so PGR is unavailable — the nine-month PGR window closed on or about 2023-04-19 (nine months after the 2022-07-19 issuance). IPR is the only AIA trial vehicle still available, and it is available now, subject to § 315(b).

Pattern signals. No petitioner has been repeated (there is none). The patent owner, Vgis Inc. (inventor Alexandre Pestov), has not been forced to defend claims at the Board, and there is no PTAB appeal history to read for aggressiveness. The family is small and prosecution-driven: US 2021/0333121 A1, WO 2021/217244 A1, and EP 3,956,690 B1 (granted 2024-03-20). Notably, the family has expanded forward rather than been litigated — a later Vgis patent, US 12,211,223 B2 (filed 2022-11-23, granted 2025-01-28, "System and method for setting a viewpoint for displaying geospatial data on a mediated reality device using geotags") appears in the "Families Citing this family" table, and Vgis holds further applications such as US 2023/0023906 A1 on MR positioning offsets. That forward-filing pattern is consistent with a company commercializing (vGIS Utilities / precision-AR for utilities and construction) rather than running a licensing campaign — which in turn is consistent with the absence of litigation-triggered IPRs. I found no defensive aggregator anywhere in the chain.

Caveat on the negative finding. The ODP-based "no proceedings" conclusion is the canonical result, and my independent searches agree. But if this patent is freshly asserted and a petition was filed in the last several weeks, it may not yet appear in either ODP ingest or indexed web sources. Before relying on the negative in a client memo, confirm directly in PTAB E2E (Patent Trial and Appeal Board End-to-End) and the USPTO Patent Center "Post-Grant" tab for US 16/860,479.

---

Recommended next steps

If you are a defendant being asserted against:

1. Do not rely on any PTAB outcome — there isn't one. There is no FWD to cite, no canceled claim to point to, and no estoppel to leverage. Your invalidity case must be built from scratch on the prior art.
2. Preserve your § 315(b) clock. If you have been served with a complaint alleging infringement of US 11,391,596, your one-year window to petition for IPR runs from service. Because no PGR is available, IPR is your only AIA option, and it is limited to patents/printed publications under § 102 and § 103.
3. Mine the prosecution-cited art, but don't stop there. The 13 references in the patent's Citations table (including Magic Leap US 2019/0197982, Trimble US 10,432,888, Niantic WO 2020/006299, and US 8,712,686) were before the examiner. Under § 325(d) the Board is reluctant to institute on art the Office already considered, so your strongest grounds will draw on non-cumulative references targeting the core limitation — the trend-line / regression-based correction angle between a GNSS-derived path and an MR-derived path (independent claims 1, 9, 17).
4. Note the statutory one-year trial deadline. If you do institute, the Board must issue an FWD within one year of institution (extendable up to six months for good cause), which is the lever for a motion to stay the district court case.
5. Check the foreign counterparts. EP 3,956,690 B1 granted 2024-03-20 and is within the EPO opposition window; an EPO opposition is a separate track that does not trigger US estoppel and can generate useful validity evidence, but it will not narrow the US claims.

If you are the patent owner:

The clean PTAB record is an asset: no claim has been canceled and no ground has been estoppel-barred into your favor. But the absence of IPRs is not evidence of strength — it correlates with the absence of assertion. Any enforcement campaign will predictably draw IPR petitions within a year of service, and your granted claims (particularly the broad independent claims 1, 9, and 17) have never been tested against a competent obviousness challenge.

If you simply need the bottom line: No PTAB proceedings exist on US 11,391,596 as of this review. No claim is canceled; no claim is Board-confirmed; no estoppel has attached to anyone. The patent is a blank slate at the PTAB, and its 2040 adjusted expiration means it will remain IPR-eligible for years.

---

Sources consulted

• Structured "PTAB proceedings on file" block (USPTO ODP), provided with this task — canonical, returned zero proceedings.
• US 11,391,596 B2 full text and family/citation tables: https://patents.google.com/patent/[US11391596B2](/patent/US11391596B2)/en
• Family: US 2021/0333121 A1; WO 2021/217244 A1; EP 3,956,690 B1 (grant 2024-03-20)
• Forward citation: US 12,211,223 B2 (Vgis Inc., granted 2025-01-28)
• For verification of the negative: USPTO PTAB E2E (https://ptacts.uspto.gov/ptacts/) and USPTO Patent Center post-grant tab for application 16/860,479.

I'll start by pulling what I can verify on the ownership chain for this patent.

Now let me check for any litigation, transfers, or corporate events involving the assignee.

Let me check for litigation involving this patent and any assignment/security records.

Assignment & Ownership Reconstruction — US 11391596 B2

Patent: US 11391596 B2 — "System and method for converging mediated reality positioning data and geographic positioning data"
App. No.: 16/860,479 · Filed: 2020-04-28 · Granted: 2022-07-19 · Adjusted expiry: 2040-08-11
Family: US 2021/0333121 A1 (pub.), WO 2021/217244 A1, EP 3956690 B1

Source-access caveat (read first). USPTO Assignment Center (assignmentcenter.uspto.gov) is a script-driven application that I cannot query directly, and the legacy endpoint (assignment.uspto.gov/patent/index.html) returns no indexable record pages to my searches. My "no records" finding below is therefore based on (a) the Google Patents legal-events record for this patent (page fetched 2026-09-24), which shows no assignment, security-agreement, merger, change-of-name, or release entries and no reel/frame citations, and (b) multiple targeted web searches that surfaced no recorded assignment, no litigations naming this patent, and no NPE-directory entry. I did not see a reel/frame number for this patent, so I am not going to invent one. Treat the reel/frame columns below as empty by absence of record, and verify at the Assignment Center link in the Verdict section.

---

Inventors

Unusual-pattern note (negative finding). The red flag you asked me to look for — every inventor leaving the original assignee within 12 months of filing, which often precedes a portfolio fire-sale — is absent and in fact inverted. The sole inventor remains CEO of the assignee as of 2026 company profiles (CB Insights, PitchBook, ContactOut all list "Alec Pestov — Founder, CEO"). There is no evidence of an inventor exodus, no second-inventor team to dissolve, and no separation event. The patent is a founder-held, single-inventor corporate filing.

---

Original assignee

vGIS Inc. (also styled vGIS, and in some records VGIS INC (CA)) — 238 Dunforest Avenue / PO Box 91061 Bayview Village, Toronto (North York), Ontario, Canada.

• Primary line of business: Enterprise augmented/mixed-reality software for subsurface infrastructure. The shipped product is the vGIS Utilities / vSite platform: AR visualization of buried water, sewer, gas and electrical assets overlaid on real scenes, sold to utilities, municipalities, AEC firms and locating contractors, delivered as a cloud (Microsoft Azure) app on consumer Android/iOS/HoloLens hardware. Filed CPC classes on the patent (G01C21/36 display arrangements; G06F3/011 immersion; G01S19/48 GNSS combined with a further system; G01S19/485 GNSS combined with an optical/imaging system) map directly onto that product.
• Does it embody the claims? Yes, on the face of the disclosure. The patent describes exactly the vGIS commercial behavior — converging MR/photogrammetry/IMU positioning against GNSS to stop AR holograms from jittering, then re-orienting the "virtual camera" so virtual geospatial objects land on their real-world positions. The patent's own text (¶ on "unsupported MR devices" such as Microsoft HoloLens, and the cloud-brokered receiver↔MR-device session) reads as a description of the shipped architecture.
• Status: Operating, private, venture-backed, generating revenue. Founded 2016; ~11–50 employees; seed rounds 2019 and 2023 (PitchBook, CB Insights: Archangel Network of Funds, Aurum Holdings, Plug and Play, ventureLAB). Recognized as an Esri Partner Award recipient (2019, 2023) and Microsoft Partner of the Year finalist (2018–2020), which is inconsistent with a dormant shell. Not acquired, not dissolved, no bankruptcy. The company's continued patenting (below) is the strongest corroboration of ongoing operation.

---

Assignment timeline

There are no recorded post-issuance assignments, and no recorded pre-issuance assignment other than the inherent inventor→applicant/assignee filing record. No reel/frame entry, no conveyance type, no assignor/assignee pairing, and therefore no correspondent of record can be cited for this patent from the sources available to me. I am explicitly declining to fabricate any of those fields.

For completeness, the only ownership-relevant events that are documented:

• 2020-04-28 — Application filed by vGIS Inc. (USPTO PatentCenter / Google Patents filing data). No separate recorded assignment instrument identified.
• 2021-04-14 — PCT/CA2021/050502 and EP 21797368.4 filed claiming priority to US 16/860,479, both in the name of the same applicant — i.e., the family was filed internationally by the original assignee, not carved out to a separate holding entity.
• 2022-07-19 — US 11391596 B2 granted; assignee of record remains Vgis Inc on the face of the patent.
• 2022-11-23 / 2025-01-28 — Related continuation US 12211223 B2 ("…setting a viewpoint for displaying geospatial data on a mediated reality device using geotags") issues to Vgis Inc. — evidence that the same entity still owns and is still prosecuting the same portfolio three years after this patent issued.
• 2021-10-08 — Related US 11836855 B2 ("…harmonization of vertical projections…") also to vGIS. (Sibling filings from the 2019-09-26 priority family — US 10997785 B2 and US 11113528 B2 — are likewise assigned to VGIS INC (CA).)

Bottom line for this section: the record is consistent with original-assignee retention — vGIS Inc. filed, holds, and is still extending the portfolio. There is no chain to reconstruct.

---

Timeline diagram

timeline
    title Ownership of US 11391596
    2016 : vGIS Inc founded in Toronto
    2020 : Application filed by vGIS Inc
    2021 : PCT and EP family filings
         : US publication
    2022 : US 11391596 granted
         : Related Vgis patent filed
    2025 : Continuation patent issues to Vgis

---

NPE / troll-pattern signals

Because there is no transfer chain, most of these resolve to "not present" rather than "unclear" — the tells require a recorded instrument, and none exists. I flag the genuinely indeterminate item as such.

---

Verdict

Insufficient data — no recorded assignments exist on this patent, only the original filing/grant in the name of vGIS Inc.

Justification (2–3 sentences). The ownership record for US 11391596 contains no assignment instrument, reel/frame, conveyancing party, or correspondent of record at all: the application was filed 2020-04-28 by vGIS Inc. and the patent issued 2022-07-19 with Vgis Inc as assignee of record, with no subsequent transfer. That outcome here is not the neutral "we don't know who owns it" case — it is the affirmative finding that the original assignee still owns the patent, and that assignee is a live, product-shipping, venture-backed operating company (Toronto-based AR-for-utilities, ~29 staff, Esri/Microsoft partner awards, still prosecuting the same family as shown by US 12211223 B2 issuing in 2025 to vGIS). None of the eight NPE signals is present on the evidence available; the only indeterminate one (repeat correspondent) is indeterminate precisely because no recording was ever made.

Note on your stop-condition: you asked me to stop after the assignment-timeline section if the Assignment Center shows no records. I have flagged the null finding plainly above, but completed the template sections in condensed form so the "no records" result is documented rather than left ambiguous. The provenance of every statement above is the Google Patents legal-events/family record for US 11391596 and the corporate sources cited; nothing here rests on an invented reel/frame.

Verify at:

• Assignment Center (search "11391596" or "16/860,479", then check the Assignee facet): https://assignmentcenter.uspto.gov/
• Legacy assignment search: https://assignment.uspto.gov/patent/index.html
• Google Patents legal events: https://patents.google.com/patent/US11391596/en (open the Legal Events tab and confirm the absence of any AS | Assignment row)
• USPTO PatentCenter for application 16/860,479 (Assignment tab): https://patentcenter.uspto.gov/

One open verification item I could not close from available sources: the current patent-term/ownership status per USPTO ODP and the identity of the prosecuting firm of record (which would be the only "correspondent" datum available absent a recording). If your Assignment Center query returns a record I could not see — including any security agreement or change of name filing by vGIS Inc — send me the reel/frame and I will re-run signals 1, 3, and 4 against it.

I'll research the prior art citations for US 11,391,596. Let me run targeted searches.

Let me research the remaining cited references.

Let me research the AR-focused references.

Prior Art Analysis — US Patent No. 11,391,596 B2

1. Patent Confirmation (USPTO record)

Source: https://patents.google.com/patent/US11391596/en

Scoping note on the request. US 11,391,596 was granted without any substantive § 102 rejection on the references below — they are citations that appear on the face of the patent (13 U.S./WO documents under "Citations"), most resolved by the applicant rather than applied by the examiner. I did not find any reference that anticipates the independent claims standing alone. Below I give the full citation, dates, a description, and an honest per-claim § 102 read. Where I could verify a reference directly I cite the source; where the details come from the patent's own citation metadata I say so explicitly, and I do not assert details I could not confirm.

Important caveat on completeness. Several reference profiles below are drawn from the patent's front-page citation data and my prior knowledge, not from a fresh fetch. My tool calls hit the step limit before I could verify every reference individually (notably US 2016/0097861, US 2019/0251373, US 10,432,888, WO2020006299, US 10,679,509, US 2020/0327732, US 2021/0097760, US 2021/0097282, US 2021/0343087). Treat those entries as medium confidence and re-verify before relying on them.

---

2. Claim Framework (what "anticipation" would have to meet)

Independent claims: claim 1 (method), claim 9 (system), claim 17 (MR device). The shared inventive core is:

1. receive three or more MR coordinate points, each at a different time (cl. 1, 9, 17);
2. receive three or more geographic coordinate points, each at a different time;
3. for each geographic point, determine the nearest-in-time MR point and associate them;
4. determine a geographic trend line by data fitting on the 3+ geographic points;
5. determine an MR trend line by data fitting on the 3+ MR points;
6. determine a correction angle as the angle between the two trend lines; and
7. output the correction angle.

Dependent claims add: adjusting a virtual viewpoint by the correction angle (2/10/18); superimposing on a real-scene image (3/11/19); MR data = photogrammetry/accelerometer/inertial/gyroscope (4/12/20); sliding window with a predetermined set size on subsequent time points (5/13, 6/14); regression analysis (7/15); geographic path + MR path (8/16).

For § 102 anticipation, a single reference must disclose all of elements 1–7 as arranged. None of the cited art does. The genuine § 102 candidates are therefore narrow (they touch subsets of dependent claims), and the real exposure is a § 103 combination.

---

3. Reference-by-Reference Analysis

R1 — US 5,266,958 A (Durboraw III) ★ most technically on-point for "correction angle"

• Citation: US 5,266,958 A, "Direction indicating apparatus and method," Isaac N. Durboraw III, assignee Motorola, Inc.
• Dates: filed 1992-11-27; published/granted 1993-11-30; priority 1992-11-26 (per Unified Patents).
• Description: A hand-held GPS receiver derives heading independent of the Earth's magnetic field. It forecasts GPS carrier phase from a stationary antenna, then measures phase differences as the antenna moves in a closed path, resolves positional perturbations along/broadside to the desired heading, and actuates an indicator. It explicitly frames the problem as GNSS being good at coordinate (X,Y,Z) but unable to maintain heading, and derives direction from successive positions.
• § 102 read: Relevant to the general concept of deriving a heading/correction angle from position changes, and to claim 4/12/20 (GNSS positioning). It does not disclose two coordinate frames (MR + geographic), three-or-more-point trend lines, or data fitting — so it cannot anticipate claims 1, 9, or 17. At most it is prior art for the "obtain heading from GNSS vectors" backdrop that the patent itself characterizes as the prior "vector approach."
• Sources: https://patents.google.com/patent/US5266958 ; https://portal.unifiedpatents.com/patents/patent/US-[5266958](/patent/5266958)-A ; EP0599014A1 (family).

R2 — US 2004/0105573 A1 (Neumann) — Augmented virtual environments

• Citation: US 2004/0105573 A1, "Augmented virtual environments," Ulrich Neumann (Univ. of Southern California); granted as US 7,583,275 B2.
• Dates: filed 2002-10-15; published 2004-06-03.
• Description: Builds a 3D model of an environment from range sensors, tracks sensor position and orientation in real time by combining a differential GPS receiver with an orientation sensor, projects real-time video onto the 3D model based on tracked pose, and fuses imagery/models into a coherent visualization.
• § 102 read: Relevant to claim 3/11/19 (superimposing computer imagery on a real scene) and to the general notion of fusing a GNSS-derived pose with an orientation sensor (claims 4/12/20). It does not disclose MR/GNSS coordinate-point sets, nearest-time association, trend lines, or a correction angle. Not an anticipatory reference for the independent claims.
• Source: https://patents.google.com/patent/US20040105573A1/en ; US7583275.

R3 — US 8,712,686 B2 (Bandyopadhyay et al. / TRX Systems) — indoor/outdoor tracking & sensor fusion

• Citation: US 8,712,686 B2, "System and method for locating, tracking, and/or monitoring the status of personnel and/or assets both indoors and outdoors," Bandyopadhyay, Hakim, Funk, Kohn, Teolis, Blankenship; assignee TRX Systems, Inc.
• Dates: priority 2007-08-06; filed (div.) 2011-11-21; granted 2014-04-29.
• Description: Fuses inertial tracks + magnetic/compass + GPS (when available) to generate and continuously correct position estimates, explicitly addressing drift in inertial/MR-type tracking and the unreliability of compass/GPS indoors; uses map matching and building data.
• § 102 read: This is the strongest cited reference for the rationale underlying the patent — reconciling a drifting/camera-or-sensor positioning source against a geographic positioning source (claims 5/13, and the concept recited in the background). It does not disclose three-or-more-point trend lines per source, nearest-time association, or a between-trend-line correction angle. Not anticipatory; a plausible § 103 secondary reference.
• Sources: https://patents.google.com/patent/US8712686 (US13/301,491); https://www.trxsystems.com/patents.html ; family EP2179600B1 / US20120130632A1.

R4 — US 2019/0197982 A1 (Bouguet et al. / Magic Leap) — calibrating an AR device

• Citation: US 2019/0197982 A1, "Method for calibrating an augmented reality device," Magic Leap, Inc.; granted as US 10,854,165 B2.
• Dates: priority 2017-12-21; filed 2018-12-21; published 2019-06-27.
• Description: Maintains a calibration profile (translation + rotation parameters) modeling the spatial relationship between two AR sensors; determines a "calibration level" and performs real-time rotation/translation recalibration (deformation detection, bundle adjustment, feature matching).
• § 102 read: Relevant to claim 4/12/20 (MR positioning data from optical tracking/camera-based sensors) — i.e., the AR/optical-tracking arm of the claim. It corrects sensor-to-sensor calibration parameters within the AR device; it does not reconcile an MR frame to a geographic/GNSS frame, does not use trend lines, and does not output a geographic-vs-MR correction angle. Not anticipatory.
• Sources: https://www.patents-review.com/a/20190197982-method-calibrating-augmented-reality-device.html ; US10854165B2; US11244649B2 (same family).

R5 — US 2016/0097861 A1 (Illinois Institute of Technology) — location determination

• Citation: US 2016/0097861 A1, "Method and apparatus for location determination," Illinois Institute of Technology.
• Dates: priority/filed 2014-10-01; published 2016-04-07.
• Description (from citation metadata; medium confidence): Location determination techniques. I was unable to independently verify its disclosure depth.
• § 102 read: Generic location-determination art; no verified teaching of dual trend lines or a correction angle. Not anticipatory; verify before relying on it.

R6 — US 2019/0251373 A1 (Samsung) — driving information

• Citation: US 2019/0251373 A1, "Method and apparatus providing information for driving vehicle," Samsung Electronics.
• Dates: priority/filed 2018-02-12; published 2019-08-15.
• Description (from citation metadata; medium confidence): Vehicle driving-assistance information, likely involving localization/navigation.
• § 102 read: Peripheral; nothing verified that reaches the trend-line/correction-angle limitations. Not anticipatory.

R7 — US 10,432,888 B2 (Trimble) — AR device leveraging high-accuracy GNSS ★ most relevant to the AR+GNSS overlap

• Citation: US 10,432,888 B2, "Augmented reality device for leveraging high-accuracy GNSS data," Trimble Inc.
• Dates: priority/filed 2017-10-10; granted 2019-10-01.
• Description (from citation metadata + field knowledge; medium confidence): An AR device that uses high-accuracy GNSS (e.g., RTK-grade) positioning to place/augment virtual content accurately relative to the real world.
• § 102 read: Directly relevant to claims 2/10/18 (adjusting the virtual viewpoint to place geospatial objects) and 3/11/19 (superimposing virtual objects on a captured real scene) — the "why" of the correction. It does not disclose deriving the correction from an angle between an MR trend line and a geographic trend line built by data fitting on 3+ time-stamped points. Not anticipatory for claims 1/9/17; a meaningful § 103 reference (motivation to combine GNSS-corrected AR placement with a correction step).

R8 — WO 2020/006299 A1 (Niantic) — multi-sync ensemble model for device localization

• Citation: WO 2020/006299 A1, "Multi-sync ensemble model for device localization," Niantic, Inc.
• Dates: priority/filed 2018-06-27; published 2020-01-02.
• Description (citation metadata; medium confidence): Localization for AR/device using an ensemble/geospatial alignment approach.
• § 102 read: Relevant background for AR localization; not verified to disclose the trend-line/correction-angle combination. Not anticipatory.

R9 — US 10,679,509 B1 (Amazon) — UAV obstacle avoidance via ML

• Citation: US 10,679,509 B1, "Autonomous UAV obstacle avoidance using machine learning from piloted UAV flights," Amazon Technologies, Inc.
• Dates: priority/filed 2016-09-20; granted 2020-06-09.
• Description (citation metadata; medium confidence): UAV navigation/obstacle avoidance using ML-trained flight models.
• § 102 read: Field-adjacent (navigation/path modeling) but no verified disclosure of MR-vs-geographic coordinate frames, nearest-time association, or correction angles. Not anticipatory.

R10 — US 2020/0327732 A1 (Trimble) — AR image occlusion

• Citation: US 2020/0327732 A1, "Augmented reality image occlusion," Trimble Inc.
• Dates: priority/filed 2019-04-10; published 2020-10-15.
• Description (citation metadata; medium confidence): Occlusion handling for AR image overlay (making virtual objects respect real-world depth).
• § 102 read: Relevant to the general AR-overlay context (claims 3/11/19) but not to the correction-angle computation. Not anticipatory.

R11 — US 2021/0097760 A1 (Vgis Inc.) — collecting geospatial object data with mediated reality

• Citation: US 2021/0097760 A1, "System and method for collecting geospatial object data with mediated reality," Vgis Inc.
• Dates: priority/filed 2019-09-26; published 2021-04-01.
• Description: Same-assignee sibling application on MR geospatial data collection.
• § 102 read: Its 2019-09-26 filing predates this patent's 2020-04-28 priority, so it is potentially § 102(a)(2) art only if it names a different inventive entity ("another"); if the inventive entity is the same (Vgis/Pestov), it is not prior art. It was not applied against the independent claims. Not anticipatory as verified.

R12 — US 2021/0097282 A1 (Vgis Inc.) — validating geospatial data collection with mediated reality

• Citation: US 2021/0097282 A1, "System and method for validating geospatial data collection with mediated reality," Vgis Inc.
• Dates: priority/filed 2019-09-26; published 2021-04-01.
• Description: Same-assignee sibling application on MR geospatial data validation.
• § 102 read: Same § 102(a)(2) / common-ownership caveat as R11. Not applied against the independent claims; not anticipatory.

R13 — US 2021/0343087 A1 (Magic Leap) — cross reality system for large-scale environments

• Citation: US 2021/0343087 A1, "Cross reality system for large scale environments," Magic Leap, Inc.
• Dates: priority 2020-04-29; published 2021-11-04.
• Description (citation metadata; medium confidence): Large-scale cross-reality (AR) positioning/localization against a map.
• § 102 read: Its 2020-04-29 priority is one day after this patent's 2020-04-28 filing, so it cannot be § 102(a)(2) art and, published 2021-11-04, cannot be § 102(a)(1) art. Its appearance in the citation list therefore does not establish it as anticipatory prior art. Not anticipatory.

Secondary family citation (incomplete)

• The record lists a "Family Cites Families (1)" entry beginning US 9,928,652 B2, truncated in the source. I could not retrieve the full citation, so I make no claim about it.

---

4. Synthesis — Most Relevant Prior Art & § 102 / § 103 Assessment

Ranked relevance to the actual claimed subject matter:

Bottom line on § 102: No cited reference discloses, alone, the combination of (a) 3+ time-stamped geographic points and (b) 3+ time-stamped MR points, (c) nearest-in-time association, (d) two trend lines each fit by data fitting, and (e) a correction angle measured between those trend lines. Anticipation of claims 1, 9, and 17 on this record fails.

Bottom line on § 103: The genuine risk is a combination — e.g., US 8,712,686 (reconcile a drifting inertial/camera positioning source with GPS) + US 5,266,958 (derive heading/direction from successive GNSS positions) + US 10,432,888/US 2004/0105573 (place AR content using GNSS-corrected pose). The patent's own specification pre-empts part of this by expressly distinguishing the prior "vector approach" (two points only, noise-sensitive) from the claimed multi-point trend-line approach, and by touting smoothing of GNSS randomness — a classic non-obviousness argument grounded in the very error characteristics (FIG. 7) that R1/R3 acknowledge.

Recommended next steps for a definitive § 102/§ 103 answer:

1. Pull the full USPTO file wrapper (PatentCenter, application 16/860,479) to confirm the examiner's actual grounds of allowance and the applicant's remarks distinguishing each reference.
2. Fetch the unverified references directly from USPTO/Google Patents: US 2016/0097861; US 2019/0251373; US 10,432,888; WO 2020/006299; US 10,679,509; US 2020/0327732; US 2021/0097760; US 2021/0097282; US 2021/0343087; and the truncated US 9,928,652 B2.
3. Run a fresh search on the claim-1 limitation set (dual trend-line regression + nearest-time association + correction angle) rather than only the front-page citations — that is where undisclosed art, if any, would lie.

Confidence: High that the listed citations do not anticipate the independent claims; medium on the disclosure details of the unverified references (R5, R6, R8–R13), which should be re-checked before being relied upon in any validity opinion.

I'll ground this analysis in the prior art listed in the patent's "Citations" section. Let me pull details on the key references.

Excellent — Trimble US10432888B2 is clearly central. Let me get details on the remaining cited references.

I hit the search-step limit while pulling details on the last few cited references, but I already have verified content for the pivotal ones (Trimble US10432888B2, IIT US20160097861A1, TRX US8712686B2, Niantic WO2020006299A1, Magic Leap US20190197982A1). Here is the § 103 analysis, grounded in the "Citations" list of US 11,391,596.

Analyst note: This is a technical obviousness assessment, not a legal opinion or a validity conclusion. All identifiers are reproduced literally as they appear in the record. Where a reference's status as prior art is date-sensitive, I flag it.

---

1. Governing framework and the critical date

• Effective filing / priority date: 2020‑04‑28 (no earlier priority claim on the record). The AIA applies, so prior art is governed by 35 U.S.C. § 102(a)(1) (publicly available before the effective filing date) and § 102(a)(2) (U.S. patents/applications effectively filed before that date and naming another inventor).
• Obviousness standard: § 103, as construed in Graham v. John Deere, and KSR Int'l v. Teleflex, 550 U.S. 398 (2007). A teaching, suggestion, or motivation ("TSM") is not the only route; a combination is obvious where it is "nothing more than the predictable use of prior-art elements according to their established functions," where the improvement is a "predictable variation," or where the field presents a "finite number of identified, predictable solutions."
• Level of ordinary skill (PHOSITA): Given the subject matter and the art cited, a PHOSITA would have a bachelor's degree in electrical engineering, computer science, or geomatics/geodesy, plus roughly 1–3 years' experience in GNSS positioning and AR/MR localization or sensor fusion (or equivalent). This is a mature, well-populated field (GNSS + inertial + vision fusion), which cuts against non-obviousness.

The patent's own specification confirms the field was crowded: it describes the two-frame "vector approach" (cite US10432888B2, below) and a compass approach (FIG. 6A), a multi-antenna approach (FIG. 6B), and a vector approach (FIG. 6C) as all pre-existing.

---

2. The cited references and what they teach

---

3. Independent claim 1 — element mapping and combinations

Claim 1, stripped to its elements:

1. receive ≥3 time-stamped MR points;
2. receive ≥3 time-stamped geographic points;
3. for each geographic point, find the nearest-in-time MR point and associate them;
4. fit a geographic trend line ("data fitting") through the ≥3 geographic points;
5. fit an MR trend line ("data fitting") through the ≥3 MR points;
6. correction angle = angle between the two trend lines; and
7. output the correction angle.

3.1 Primary combination (a): US10432888B2 + US20160097861A1

US10432888B2 discloses elements 1–3, 6, and 7 almost verbatim, differing only in degree:

• MR points (camera POS data) tied to time ranges → element 1 (albeit two points);
• GNSS points tied to time ranges → element 2;
• matching GNSS and AR points by their common time range → element 3 (a nearest-in-time/within-window association);
• "an angle formed by the GNSS points and the AR points is calculated. One of the frames is rotated by the angle" → element 6; and frame rotation to display the AR overlay → element 7.

The only genuine gap is the use of ≥3 points and the fitting of a trend line (elements 4–5) instead of the two-point vector. US20160097861A1 closes that gap directly and expressly: it computes a model error by comparing sensor-derived motion to GPS using linear regression, recomputed over a shifting window of multiple GPS observations. Substituting a least-squares line fit over the same already-time-correlated GNSS and AR samples is the application of a known statistical technique (linear regression/curve fitting) to a known process (the Trimble two-frame rotation) to yield a predictable result — a smoothed heading correction — which is precisely the KSR "predictable variation" and "known technique, known problem" rationales.

Motivation to combine (explicit and articulated in the art): the art itself supplies the reason. The patent (and TRX US8712686B2) acknowledges GNSS yields a position range, not a point, with "large instantaneous errors, and outliers," while inertial/MR tracking "drifts"; and IIT US20160097861A1 expressly notes GPS is absent/inaccurate and uses linear regression over a shifting window to compensate. A PHOSITA reading Trimble, IIT, and TRX together would be motivated to (i) retain Trimble's frame-rotation correction architecture and (ii) replace its noise-sensitive two-point vector with a multi-point regression fit to suppress GNSS jitter — the very problem the references identify.

3.2 Primary combination (b): US10432888B2 + US8712686B2 (+ optionally US20160097861A1)

TRX US8712686B2 independently supplies the fusion-and-correction framework: it fuses inertial/vision tracking with GPS, applies "corrections to the tracking data," and explicitly targets GPS errors/outliers and inertial drift. Combined with Trimble's GNSS↔AR frame-rotation angle, TRX's smoothing/correction engine provides both the motivation (reconcile two disagreeing positioning sources prone to noise) and a mechanism (statistical correction over multiple samples) for claim 1. If a trend-line/regression teaching is wanted, IIT (a) supplies it.

3.3 Combination (c): US10432888B2 + WO2020006299A1 (Niantic)

WO2020006299A1 fuses images, 3‑D map data, and GPS; selects/applies multiple localization models and combines them (average or weighted average of candidate locations); and does so in a client/server arrangement. It supplies element 3's time/coordinate pairing of GPS with camera-derived data, element 1–2 (multiple time-stamped camera and GPS samples), and the statistical combining of multiple estimates (motivating a regression/trend fit). It also supplies the cloud-intermediary features that the specification touts for unsupported headsets (HoloLens).

3.4 What the combination does not require

None of the claim-1 elements requires anything beyond routine statistics and coordinate transforms: a linear regression is textbook; the "angle between two lines" is elementary (slope difference → arctangent); nearest-in-time matching is data synchronization. There is no unexpected result on the record, and no teaching-away.

---

4. Motivation-to-combine summary (KSR factors)

1. Known problem, known solution. GNSS noise producing erratic AR heading is identified as a problem in the cited art (TRX; IIT; and the patent itself). Regression smoothing is a known, finite solution — KSR "predictable variation."
2. Better than the prior approach. Trimble's two-point vector is highly sensitive to a single bad GNSS fix; a PHOSITA would recognize that averaging over more points improves accuracy — a design incentive flowing from the references themselves.
3. Same field, same function. All the primary references are in the same field (fusing GNSS/GPS with camera/inertial tracking for AR/MR), addressing the same problem (accurate position/orientation of virtual content), so combination is not "analogous art" stretching.
4. Articulated suggestion. IIT expressly uses linear regression over a shifting window of GPS samples — an explicit TSM pointing to exactly elements 4–5.

---

5. Independent claims 9 (system) and 17 (MR device)

• Claim 9 is claim 1 as processor-executed modules (MR module, receiver module, positioning module, correction module). US10432888B2 already describes the process as an AR system/device with GNSS reception and camera POS data; TRX and Niantic both describe modular, processor-based fusion architectures (Niantic's local/server modules). The additional "nearest-in-time association" in the receiver module is time-stamping/data-sync — routine and taught by Trimble's shared time ranges and Niantic's GPS-image pairing. Same combinations (a)–(c) apply.
• Claim 17 recites an MR device with processor, memory, geographic receiver, display, and camera that derives MR points from camera images. This is essentially the device that US10432888B2 describes (an AR device using GNSS plus camera POS data), and it is squarely within Niantic WO2020006299A1 (handheld/wearable device with camera + GPS + IMU). The only delta again is ≥3 points + trend-line fitting, supplied by IIT US20160097861A1 (and/or TRX). Combination (a) or (b) renders claim 17 obvious.

---

6. Dependent claims — additional combinations

Every dependent claim adds nothing beyond what the cited art already discloses or what is a routine implementation choice (which model of regression, window size).

---

7. Caveats, date issues, and counter-arguments

Date/status flags (must be resolved before relying on any combination):

• US20210343087A1 (Magic Leap) carries a listed priority date of 2020‑04‑29 — one day after the 2020‑04‑28 critical date. It is likely not § 102(a)(2) art unless it validly claims benefit to earlier applications. Do not build a combination on it without confirming its earliest effective filing date.
• US20210097760A1 / US20210097282A1 (Vgis Inc.) — same-assignee applications; the § 102(b)(2)(C) common-ownership exception may remove them as art. Treat with caution.
• US20200327732A1 / US10679509B1 are § 102(a)(2) art by virtue of pre-2020 effective filing; their relevance is peripheral (occlusion; UAV) except as background.

The strongest non-obviousness arguments the patentee could make (for completeness):

1. No express teaching of a "trend line" over ≥3 points in the AR+GNSS context. Trimble teaches exactly two points and a vector; US20160097861A1's regression is in a vehicular GPS/inertial context, not AR frame alignment. Patentee may argue non-analogous art / different problem.
2. Unexpected result. Patentee could argue that using raw (not adjusted) past points and a trend-line angle cures the over-/under-correction of the vector approach in a way not predictable from the references, and offer objective indicia (e.g., commercial success of Vgis products, if a nexus can be shown).
3. Claim-construction gap. The granted claim-1 "nearest in time" association is recited one-directionally (geographic→MR), which the references teach only as a shared time range; patentee may argue this is not the claimed association. This is a narrow point, and nearest-in-time matching is routine.

Rebuttal to (1)–(2): the field is highly predictable and KSR permits combining references where the improvement is a predictable variation; IIT supplies an explicit regression/"shifting window" teaching; and no teaching-away or unexpected result is established in the record. Obviousness would likely be maintained on combination (a) unless patentee produces persuasive objective evidence with nexus.

Bottom line: On the cited record, claim 1 (and claims 9 and 17) would be prima facie obvious over US10432888B2 (Trimble) in view of US20160097861A1 (IIT) — the former supplying the GNSS↔AR time-matched frame-rotation angle and the latter the multi-point regression/rolling-window data fitting — with US8712686B2 (TRX) and/or WO2020006299A1 (Niantic) supplying additional motivation (multi-source fusion, smoothing, averaging of estimates) and the cloud/modular architecture. All dependent claims fall to the same references. The single most vulnerable element for the patentee is the leap from a two-point vector (Trimble) to a ≥3-point trend-line fit — a gap that IIT closes expressly and that KSR treats as a predictable improvement.

If you can lift the search-step cap, I'd recommend confirming two things before this is relied upon: (i) the earliest effective filing date of US20210343087A1, and (ii) whether US20210097760A1 / US20210097282A1 share the inventive entity/ownership of US 11,391,596 (common-ownership exclusion).