US 11168997

I'll run targeted searches for this patent number and any associated litigation.

Let me check for litigation and the complete claim set.

US 11,168,997 B2 — "Reverse natural guidance"

Scope note on method: I searched for this exact number (no prefix/suffix substitutions). The bibliographic and specification data below comes from the authoritative full text of US11168997B2 supplied to me (Google Patents, https://patents.google.com/patent/US11168997/en) plus a Justia Patent full-text mirror (https://patents.justia.com/patent/[11168997](/patent/11168997)). No search result I retrieved matched this patent number to any other document; the only hits for the literal string "11168997" outside the US patent were an unrelated EPO application number (EP 11168997 A, a French wall-fastener case) — that is a different, non-US identifier and should not be conflated.

Bibliographic data

Continuation chain (per the "Reference to Related Applications" section of the patent): this patent is a continuation under 37 C.F.R. § 1.53(b) of Ser. No. 15/171,118 (now US 10,072,941), which is a continuation of Ser. No. 14/258,205 (now US 9,372,093), which is a continuation of Ser. No. 13/430,199 filed Mar. 26, 2012 (now US 8,744,771).

⚠️ Literal-reading flag (no auto-correction applied): the patent text states Ser. No. 14/258,205 was "filed Apr. 22, 2015." The 14/258,xxx serial series corresponds to 2014 filings, so this date appears internally inconsistent with the serial number. I am reporting it exactly as written and not correcting it; the priority date of 2012-03-26 rests on the 13/430,199 filing.

Abstract (verbatim)

"The disclosed embodiments relate to recognition of navigation and/or guidance related information within an expressed conversational narrative between a provider of the narrative and a receiver thereof. The recognized navigation and/or guidance related information is then converted into navigation data which may be used to present a navigation route and/or guidance information related thereto to the receiver."

Independent claims — plain-language overview

Claim 1 — computer-implemented method (the "reverse natural guidance" core). The system observes a conversation between two other people (a "first person" and a "second person") rather than being fed directions by one user. The conversation contains multiple navigation-oriented conversational elements describing a route to a destination, and those elements include at least one non-verbal element — a non-verbal physical movement of a body part of the first person (e.g., pointing, waving, drawing). The processor then: (a) identifies the navigation-oriented conversational elements; (b) converts each into an associated navigation data element; (c) compiles those data elements into a navigation route; and (d) presents at least part of the route (audibly, visually, tactilely, or a combination).

Confidence note: I did not retrieve verbatim claim 1 text. The two independent claims are claim 1 (method) and claim 10 (system), confirmed by the dependent claims that refer back to them ("The computer implemented method of claim 1…" and "The system of claim 10…"). The claim 1 gist above is reconstructed from dependent claims 4, 7, 8 and 9 — which respectively add accuracy determination, first-person/second-person contextual profiles plus implied navigation elements, audio-sensor observation, and optical/motion/touch/proximity-sensor observation of the non-verbal body movement — and from the patent's "Definitions" and Detailed Description sections. Treat the concept as reliable and the exact wording as unverified.

Claim 10 — system. A processor with coupled memory and five logic modules: (1) first logic observes at least part of a conversation between a first and second person describing a route to a destination, where the conversation includes at least one non-verbally expressed navigation-oriented narrative and at least one verbally expressed navigation-oriented narrative related to the non-verbal one, the non-verbal part being a non-verbal physical body movement of the first person; (2) second logic identifies the navigation-oriented conversational elements; (3) third logic converts each element into an associated navigation data element; (4) fourth logic compiles the data elements into a navigation route; and (5) fifth logic presents at least a portion of that route.

The requirements unique to the independent claims (as distinct from the earlier family members and the pre-grant publication) are that the conversation being sensed is between two people who are neither the device nor a user entering data into it, and that the gesture-based navigation content must be paired with related verbal navigation content.

Dependent claims (method 2–9; system 11–18) add, among other things: real-time execution of identification/conversion/compilation; destination being an interior location in a building; determining and presenting an accuracy/reliability of the route; detecting omitted navigation elements needed for a complete route and supplying them; accessing a navigation database; deriving implied navigation elements from context (geographic location, another navigation element, temporal characteristic, or contextual profiles of either conversationalist); and receiving the conversation through audio, optical, motion, touch, or proximity sensors. I could not verify whether claims exist beyond 18.

Docket search: USPTO / PTAB / CAFC / district court

Negative result — stated explicitly: I found no Federal Circuit 2026 docket, no PTAB proceeding (IPR/PGR/CBM), and no district court complaint asserting US 11,168,997. Searches for the patent paired with "CAFC," "Federal Circuit," "appeal," "IPR," "PTAB," and "litigation" returned only unrelated matters. The HERE-related cases surfaced involved different patents, e.g. Omnitek Partners LLC v. HERE Global B.V., No. 6:20-cv-00438 (W.D. Tex.) (patents-in-suit not this one) and Facet Technology Corp. v. HERE Global B.V., No. 2:24-cv-00269 (E.D. Tex.) (asserting e.g. US 9,671,328 and US 9,335,255). Neither is authority about '997.

I cannot rule out a recently filed or unreported proceeding, since my searches are not a substitute for a PACER docket query or the CAFC's public docket. If you need a definitive litigation-negative, that should be confirmed directly against PACER/CAFC CM-ECF.

Overall uncertainties

1. Verbatim claim 1 wording and total claim count are not authoritatively confirmed here (claims 10–18 are quoted from a Justia mirror; claim 1 is reconstructed from dependents).
2. The Apr. 22, 2015 date for Ser. No. 14/258,205 is reproduced as written and is inconsistent with that serial series — a probable typographical error I am not correcting.
3. Assignee and legal-status entries carry the source's own disclaimer; current ownership should be verified against USPTO Assignment records (the 2018-08-14 assignments to NAVTEQ B.V. and HERE Global B.V. are the only ones shown).
4. No CAFC 2026 activity tied to this patent could be identified in any search.

I'll search for litigation involving US patent 11168997.

Let me search additional litigation-specific sources.

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

Let me run a few more targeted checks against PTAB dockets and the patent family.

Proceedings overview

Total AIA trial proceedings on file for US 11,168,997: zero. The USPTO Open Data Portal returns no IPR, PGR, or CBM proceeding naming this patent, and my independent web checks (PTAB petition/decision text, docket aggregators, and the patent's own family members) surfaced nothing to contradict that — breakdown is therefore 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denied. The defensive posture this gives a defendant is the opposite of "hardened by surviving IPRs": no third party has ever taken a swing at this patent at the Board, so every claim is untested, no claim has been canceled, and you have no PTAB record to borrow from — but you also face no § 315(e)(2) estoppel and no procedural bars that a prior petitioner would have created. You'd be the first challenger, with a clean slate.

There are no per-proceeding entries to write, because there are no proceedings. I will not invent proceeding numbers to fill the template.

---

What I verified (and what I couldn't)

Identifier trap worth flagging: searching the bare string "11168997" returns EP 11168997 A 20110607, which is an unrelated European application number — a French wall-fastener application (EP 2392750 A1, "Tire for removably supporting and attaching a structure to a wall"). Nothing about that file relates to US 11,168,997. Any future automated docket pull should key on the U.S. patent number with the "US" prefix, not the 8-digit string alone.

What I could not confirm, and am not asserting: I did not have Lexis, Docket Navigator, or a live PTAB E2E session available, so I cannot rule out a recently filed petition filed inside the last few weeks that has not yet been indexed by ODP or crawled by search engines. The prompt's default controls: no PTAB activity on file. Treat that as accurate as of the ODP ingest date, and re-check E2E before relying on it in a formal filing.

---

Patent profile (relevant to why nobody has petitioned yet)

• Patent: US 11,168,997 B2, "Reverse natural guidance" — Google Patents
• Application: 16/101,557, filed 2018-08-13; granted 2021-11-09
• Inventor: Mark Slusar | Assignee: HERE Global B.V. (originally NAVTEQ B.V.; name change recorded 2018-08-14)
• Continuation chain (per the patent's own Reference to Related Applications):
  • 13/430,199, filed 2012-03-26 → US 8,744,771
  • 14/258,205, filed 2015-04-22 → US 9,372,093
  • 15/171,118, filed 2016-06-02 → US 10,072,941
  • 16/101,557, filed 2018-08-13 → US 11,168,997 (this patent)
• Earliest priority: 2012-03-26 (pre-AIA). IPR is available on the pre-AIA § 102/§ 103 prior-art window, and any IPR would be filed under AIA procedures.
• Two conflicting expiration dates on the public record — flagging both rather than picking one:
  • Google Patents legal status: "Active, expires 2033-05-22" (adjusted expiration)
  • USPTO-adjacent aggregator: adjusted expiration 2039-10-09
    These are inconsistent and I could not reconcile them from the sources available. Do not build a damages model or a § 102(b) / on-sale analysis on either date without pulling the patent's own terminal disclaimer and PTA certificate from Patent Center. Note that the patent is a pre-AIA continuation; the 2018-08-13 filing date is not the effective filing date.
• Note on the chain dates as printed: the patent prints 14/258,205 as "filed Apr. 22, 2015." I am reproducing that literally per the source; the serial-number/date pairing is unusual and, if it matters, should be confirmed against the file wrapper. I am not auto-correcting it.
• The claim set is not reproduced in the source text I have, so I am not stating which claims exist, how many there are, or which are independent. Any assertion-based analysis needs the claim set pulled from the grant document first.

---

Strategic summary

Canceled vs. sustained vs. untested. There is no PTAB-driven narrowing of US 11,168,997. Zero claims have been canceled (no certificate of cancellation exists, because no FWD exists). Zero claims have been confirmed as patentable over art in an IPR. Every claim is untested territory. If a demand letter or complaint asserts this patent, the patentee is asserting claims that have never been through adversarial validity review — which cuts both ways: the claims have not been narrowed by prosecution-like IPR amendment or by claim construction in a FWD, so the patentee gets to argue their full original scope, but the patentee also has no administrative finding of validity to wave at you.

Estoppel landscape — the most important point for a defendant. Because no IPR has reached a final written decision, § 315(e)(2) estoppel has attached to no one. No petitioner, real party in interest, or privy is barred from raising any ground in district court or the ITC. There is no "reasonably could have raised" shadow hanging over the prior-art space, and there is no prior petitioner's failed art combination that you would be wasting your time re-running. Your prior-art grounds are effectively fully open. The one timing constraint that can bite is § 315(b): an IPR petition is barred if filed more than one year after you are served with a complaint alleging infringement of this patent. If you have already been served, that clock matters more than anything else on this page.

Pattern signals. There is no pattern to read — no serial petitioner, no defensive aggregator (no Unified Patents / RPX-chained filing), no patent-owner appeal activity at the Federal Circuit (an appeal requires an underlying FWD, which does not exist). The relevant pattern signal is the absence itself: this is a 2012-priority HERE/NOKIA navigation patent in the "natural guidance" line with no PTAB track record at all, which usually means either the patent has not been asserted hard enough to attract a challenger, or the accused products are large-platform defendants (in-dash navigation, mapping apps) who prefer litigation venues to the Board. I could not confirm which, because I found no litigation either.

---

Recommended next steps

If you are a defendant being asserted against:

1. Do not look for a PTAB win to import. There isn't one. There is no FWD to quote, no canceled claim to point at, and no claim-level disposition to lean on. Any "the troll's patent is already dead" argument is unavailable here, and asserting otherwise would be sanction-bait.
2. Front-load the prior-art search yourself. The family's earliest priority is 2012-03-26, so the critical prior-art date is set by whether 13/430,199 provides § 112 written-description support for the claims as issued — a 2018-filed continuation claiming 2012 priority is a standard target for a priority/§ 112 attack. That is exactly the kind of argument that is not cognizable in an IPR (IPR is limited to § 102/§ 103 on patents and printed publications) but is cognizable in district court or in a PGR only if the patent were AIA-first-to-file — it is not, it is pre-AIA, so no PGR is available. Your § 112 and § 101 challenges live in court, not at the Board.
3. Preserve your § 315(b) window. If you have been served but it is less than one year ago, you still have an IPR petition available; if more than a year has passed, IPR is off the table and your validity fight is a district-court fight. Verify your service date before doing anything else.
4. Pull the file wrapper from Patent Center for 16/101,557 and its three ancestors — you want the prosecution history, any § 101 rejections, any claim amendments, and the PTA/terminal disclaimer to reconcile the 2033-05-22 vs. 2039-10-09 expiration discrepancy.
5. Check E2E yourself before filing. PTAB E2E / PTAB Center — a petition could have been filed since the ODP ingest and not yet appeared in search results. A 30-second E2E pull on the patent number and on 8,744,771 / 9,372,093 / 10,072,941 is worth doing.
6. Check CourtListener and the CAFC docket for any appeal involving this patent or its family: CourtListener, CAFC. No appeal can exist without an FWD, so I expect none — but a district court opinion (e.g., a § 101 or claim-construction ruling) could exist and would be highly informative.

If you are the patent owner:

• The absence of PTAB challenges is a modest positive signal, but it is not a validity finding. Expect a first-time petitioner to attack the 2012 priority claim and the § 101 eligibility of a "receive natural-language directions, parse them, output a route" claim — the specification in this family is written at a high level of abstraction, and the claims' reliance on "conversational elements" and "navigation oriented conversational elements" is the kind of language that draws both § 112 and Alice fire in court.

---

Bottom line: No PTAB proceedings on file. No proceedings to summarize, no FWD to quote, no appeal docket to cite. The patent stands with all claims intact and entirely untested at the Board — which means a defendant has a clean, unburdened shot at an IPR (subject to § 315(b)) and no estoppel to worry about, but also no pre-built record of invalidity to reuse.

Assignment & Ownership Analysis — US 11,168,997 B2 ("Reverse natural guidance")

Source note / confidence caveat. I was able to reach Google Patents legal‑events data, the patent's own front‑matter/continuity text, EPO/UK register data for the family, and third‑party mirrors. I was not able to open the USPTO Assignment Center result page for this patent directly (search tool only), so I could not read the recorded correspondent of record or the reel/frame of the inventor→NAVTEQ link. Where a reel/frame below is stated, it comes from a mirror of the USPTO record. Anything I could not verify is explicitly marked [not retrieved] rather than guessed.

---

Inventors

Pattern notes: A single-inventor patent with no co-inventors and no joint-development trail. No evidence of inventor departure — I found no assignment executed by Slusar away from the company, and no third-party filings by him. There is no "all inventors left within 12 months of filing" pattern here; the record is a plain employed-inventor assignment. Note also that the application is a four-generation continuation chain (13/430,199 → 14/258,205 → 15/171,118 → 16/101,557), so the chain of title attaches to the family, not to this number alone.

---

Original assignee

NAVTEQ B.V. (Veldhoven / Eindhoven, Netherlands), renamed HERE Global B.V. effective 2013‑04‑23 (recorded via reel 033830/0681).

• Primary line of business: digital map data, geographic databases and location/navigation platform services. NAVTEQ was acquired by Nokia in 2007 and rebranded HERE in 2012; the North American arm is HERE North America, LLC (f/k/a NAVTEQ North America, LLC).
• Product embodying the claims: HERE ships navigation and location products (HERE WeGo, HERE SDK, HERE guidance/navigation services), and the specification itself points to "NAVTEQ Natural Guidance" as the commercial natural-guidance product line in the same technical space. I could not verify that any specific HERE product practices claims 1–20 of this patent — treat product embodiment as plausible but unverified.
• Current status: Operating. HERE Technologies remains an operating mapping/location company. Ownership changed by share sale in 2015 (Nokia → consortium of Audi, BMW, Daimler, ~€2.8B) and additional stakes taken by Intel, Bosch, Continental, NTT, Mitsubishi, Pioneer. Third-party profile data (PitchBook) indicates HERE Technologies was acquired 01‑Aug‑2025 (PitchBook profile). Critically: none of these corporate transactions appear as recorded patent assignments — they were share purchases, so they do not generate reel/frame entries.
• Not a bankruptcy event. The 2015 Nokia→consortium divestiture and the 2025 change of control are corporate M&A, not Chapter 7/11 proceedings.

---

Assignment timeline

Two records are indexed against this patent family. Correspondent names are [not retrieved] for both — see the flag under signal #3.

• Executed date [not retrieved, on or before 2013‑09‑26] / recorded 2018‑08‑14 (as indexed by Google Patents legal events for US 11,168,997) — Reel [not retrieved]

  • Conveyance: Assignment of assignors' interest
  • Assignor: Mark Slusar (individual inventor)
  • Assignee: NAVTEQ B.V. (Netherlands)
  • Correspondent: [not retrieved] — USPTO Assignment Center result page for this patent could not be opened; no correspondent name is exposed in the mirrors I could reach.
  • Context: Ordinary employed-inventor assignment of rights to the operating employer — not a fire-sale, not a shell transfer. (The 2018‑08‑14 date is the date the record is indexed/displayed against this continuation; the instrument necessarily predates the 2013 publication naming NAVTEQ B.V. as applicant.)

• Effective 2013‑04‑23 / recorded ~2014 (reel series 033830 is a 2014-era reel; Google Patents displays the event date as 2018‑08‑14 for this continuation) — Reel 033830/0681

  • Conveyance: Change of Name
  • Assignor: NAVTEQ B.V.
  • Assignee: HERE Global B.V. (Netherlands)
  • Correspondent: [not retrieved] — same limitation as above.
  • Context: Internal reorganization / change of name only. No change in beneficial ownership; NAVTEQ B.V. and HERE Global B.V. are the same legal entity before and after.

Corroborating legal events on US 11,168,997 (Google Patents legal events panel): AS Assignment — "CHANGE OF NAME; ASSIGNOR: NAVTEQ B.V.; REEL/FRAME: 033830/0681, Effective date: 2013‑04‑23"; STCF patent grant; two MAFP maintenance-fee payments (4th year and 8th year, large entity — i.e., a large operating corporation, not a micro/small-entity NPE LLC).

There are no other recorded assignments. No security agreements, no license recordations, no corrective assignments, no releases, and no transfers to any third-party holding entity.

---

Timeline diagram

timeline
    title Ownership of US 11168997
    2012 : Slusar files parent application
    2012 : Rights assigned to NAVTEQ BV
    2013 : NAVTEQ BV renamed HERE Global BV
    2014 : Reel 033830 0681 records name change
    2018 : Continuation 16101557 filed
         : Assignment records indexed 2018-08-14
    2021 : Patent US11168997 granted
    2025 : HERE Technologies change of control

(The 2012 assignment node reflects the instrument of record; its exact execution date was not retrievable. The 2025 node is a share-level change of control and generated no recorded assignment against this patent.)

---

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. The only two recorded links are (a) inventor → NAVTEQ B.V. (operating employer) and (b) NAVTEQ B.V. → HERE Global B.V. (Change of Name, reel 033830/0681). No "IP/Holdings/Licensing/Ventures" entity, no registered-agent service address, no single-member LLC appears anywhere in the chain.

2. Known asserter in the chain — Not present. Neither NAVTEQ B.V./HERE Global B.V. nor any predecessor appears on the Acacia, Marathon, IV, IPNav, Wi‑LAN, Conversant/Mosaid, Vringo, Pendrell, Innovatio, Round Rock, or Spangenberg rosters, nor in Unified Patents / RPX high-frequency-plaintiff directories. Current owner is a large operating entity (confirmed by large-entity maintenance-fee payments, legal events MAFP).

3. Repeat correspondent across the chain — Unclear / not assessable. The correspondent of record is [not retrieved] for both reel entries, because the Assignment Center detail page could not be opened in this session. Sole related data point: the prosecution agent of record on the parent application (US 2013/0253834 A1) was Lempia Summerfield Katz LLC, a Chicago firm matching the inventor's Chicago location — but a prosecution agent is not the assignment correspondent, and one appearance is not a finding under your own rule. I will not call this either way.

4. Cascading transfers — Not present. Two links total; the second is a pure name change with a 2013 effective date. There is no <24-month chain through multiple LLCs, no shared registered-agent address, and no common-principal pattern.

5. Pre-litigation transfer — Not present. I found no infringement suit naming US 11,168,997 (or its three family members) in the sources reachable, and no assignment dated within 6 months of any suit.

6. Bankruptcy fire-sale — Not present. Nothing resembling Kodak/Nortel/Polaroid. The nearest adjacent event — Nokia's 2015 sale of HERE to Audi/BMW/Daimler — was a solvent share sale of an operating business, not a §363 sale, and did not touch the patent register.

7. Privateering — Not present. No transfer from HERE to any assertion vehicle; HERE has not monetized this family through an NPE.

8. Defensive aggregator (anti-NPE) — Not present. The chain terminates at HERE Global B.V. (an operating company), not RPX/AST/LOT/Unified/OIN.

---

Verdict

Insufficient data (no NPE indicia) — i.e., a clean operating-company chain, with the two open items being unrecoverable in this session.

The complete recorded chain is inventor Mark Slusar → NAVTEQ B.V. (assignment) → HERE Global B.V. (change of name, reel 033830/0681, effective 2013‑04‑23), with no transfers to any licensing entity, no security interests, and no assertion activity. Every affirmative signal above is "not present," and the only "unclear" calls (#3 correspondent) stem from a data-access limitation rather than from any evidence of an NPE. This is not a "NPE — insufficient data" finding in the sense of an opaque shell chain; it is an operating-company-owned patent with a one-link chain of title, and the "insufficient data" label applies strictly to the unretrieved correspondent of record.

The only way this verdict would flip is if the Assignment Center detail page revealed a correspondent who is a known NPE filer and a third, unindexed transfer exists — I found no evidence of either, so absent new records I would expect "Operating-company / non-asserting" to be the correct long-run classification.

Verify at:

• USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search by patent number 11168997)
• Legacy interface — https://assignment.uspto.gov/patent/index.html
• Google Patents legal events — https://patents.google.com/patent/US11168997/en (shows AS Change of Name, reel 033830/0681)
• UK IPO register for the EP sibling — https://www.search-for-intellectual-property.service.gov.uk/EP2831546

---

What I could not verify (stated plainly)

• The reel/frame of the Slusar → NAVTEQ B.V. assignment.
• The correspondent of record on either assignment entry.
• The exact execution date of the Slusar → NAVTEQ B.V. assignment (only the indexed date 2018‑08‑14 is available to me, and that is an indexing artifact for this continuation).
• Whether any HERE product practices claims 1–20 (product-level embodiment is inferred from HERE's product line and the specification's own reference to "NAVTEQ Natural Guidance," not from claim-charting).

I'll research US patent 11168997 and its prior-art citations. Let me start with searches.

I have the citation data. Let me verify details on the specific references and check for any additional citations.

Let me get the complete citation table for US11168997 specifically.

Prior-Art Analysis — U.S. Patent No. 11,168,997 B2 ("Reverse Natural Guidance")

1. The patent, as identified in the record

Important scoping caveat up front. The references below are the documents listed in the "Patent Citations" / "References Cited" sections of US 11,168,997 (and of its immediate parent, US 10,072,941, which shares the identical specification). They are references cited against the family — not a set of formal § 102 anticipation findings by an examiner. I could not, in this session, retrieve a USPTO Office Action or the verbatim issued-claim set for US 11,168,997 itself, so the "which claim" mapping below is an analytical mapping (grouped by the claim types this family uses), not a quotation of an examiner's rejection. Treat it as a screening opinion, not a legal conclusion.

---

2. The claims being tested (structure)

The patent family uses three parallel claim types, in three independent claims:

• Independent claim 1 — computer-implemented method (dependents 2–6)
• Independent claim 7 — system with first–fifth logic in a memory (dependents 8–13)
• Independent claim 14 — mobile device with input / converter / parser / navigation processor / route generator / output (dependents 15–20)

The shared core of all three independents is: (a) receiving at least a portion of a conversational narrative descriptive of a route, expressed by a provider to a receiver; (b) the narrative has navigation-oriented conversational elements plus descriptive element(s) characterizing another element; (c) identifying those elements; (d) converting each navigation-oriented element into a navigation data element based on the descriptive elements; (e) compiling a navigation route; (f) presenting the route. The narrower versions additionally recite non-verbal physical movement of a human body and a receiver not connected to the processor.

---

3. U.S. patent documents cited (the substantive prior art)

Dates are as listed on the face of the references/Google Patents. "Pub." = publication date; "Pri." = priority/earliest effective date.

Foreign patent document

• JP 2010-145262 A (July 2010) — listed as a foreign citation in the parent. Title not verified in this session; on its face a navigation/interface document. Not anticipatory absent a full translation review.

---

4. Non-patent literature cited

---

5. Bottom line on § 102

No single cited reference appears to anticipate any independent claim (1, 7, or 14) of US 11,168,997. The reason is structural: every independent claim requires a conversational narrative descriptive of a route, expressed by a provider to a receiver, parsed into navigation-oriented elements and descriptive elements characterizing another element, then converted into navigation data and compiled. The cited references divide into three groups, none of which contains that whole combination:

• Voice/navigation dialog systems (Schalk #1, Baig #2, Sugiyama #4, Lee #5, Van Seggelen #8, Kennewick/Voicebox #10) — natural-language voice input to a machine, i.e. a user↔device dialog; no human provider→receiver narrative.
• Gesture/input interfaces (Yan #6, Helot #7, Ricci #9) — gestures as device commands, not as a route narrative delivered to a human receiver; relevant mainly to the "non-verbal physical movement" limitations of claims 6, 13 and 19.
• Display/output and hardware (Kramer #3, Khorashadi #11, Endo #12, Tanaka #13, Rossio #14, JP '262) — presentation or platform art.

Accordingly:

• § 102 potential exists only at the level of isolated claim elements, not whole claims — e.g., gestures-as-input (§ 102 of the sub-elements of claims 6/13/19 by Yan and Ricci), output modalities (claim 20 by Kramer).
• The realistic rejection theory for claims 1, 7, 14 would be § 103, combining a voice-dialog navigation reference (Kennewick #10 and/or Schalk #1) with a gesture-input reference (#6/#7/#9), and possibly a display reference (#3), to reach the "conversational narrative + descriptive elements + gesture" combination.

---

6. Limits of this analysis (stated explicitly rather than guessed)

1. I aggregated the citation list from Google Patents' "Patent Citations" view for the US 11,168,997 family and from uspto.report's "References Cited" for the parent, US 10,072,941. One Google Patents view renders the count as "(9)" and lists nine references (Schalk, Baig, Kramer, Sugiyama, Lee, Yan, Helot, Van Seggelen, Khorashadi); the parent's full list adds Endo, Tanaka, Rossio, and the Ricci/Flextronics and Kennewick/Voicebox publications. I have presented the union and flagged the discrepancy rather than silently picking one.
2. I could not retrieve the verbatim granted claims of US 11,168,997 in this session. The claim numbering used above (1–20, method/system/mobile-device) follows the family's structure as reflected in the immediate parent, US 10,072,941, whose specification is identical. Verify against the printed claims before relying on the mapping.
3. Full titles for US 7,292,978 (Endo), US 7,392,194 (Tanaka), US 2011/0054772 (Rossio) and JP 2010-145262 A were not verified here; I have not guessed at their subject matter.
4. Cited-reference dates for the pre-2013 filings (#1–#9) predate the 2012-03-26 priority date directly. For the later-published documents (#7, #8, #9, #10 — published 2013) the effective filing/priority dates (2010–2011 and 2007, respectively) precede 2012-03-26, so they can still qualify as prior art under the pre-AIA § 102(e) / AIA § 102(a)(2) theory notwithstanding their later publication.
5. Nothing here is a legal conclusion on validity, and the cited-reference list is not the same as an examiner's § 102/§ 103 rejection record.

I'll start by trying to locate the prior-art references associated with this patent, since the task directs me to the Prior Art section.

I now have the face-of-patent "Referenced Cited" list. Let me characterize the key references so the combinations are grounded in what they actually disclose.

Obviousness Analysis — US 11,168,997 B2 ("Reverse natural guidance") under 35 U.S.C. § 103

0. Provenance of the prior art I'm using — and one important discrepancy to flag

Discrepancy flag (per your instruction to flag contradictions): the previously generated sections of this analysis (the "Patent summary" and "Litigation summary") contain no prior-art section at all. Neither the full patent text supplied to me nor those summaries list any "References Cited." So there was nothing to "build on." I therefore retrieved the face-of-patent prior-art list this session from a Justia full-text mirror of US 11,168,997:

"Referenced Cited — U.S. Patent Documents: 7292978 (Nov 6, 2007) Endo; 7392194 (Jun 24, 2008) Tanaka; 8538760 (Sep 17, 2013) Khorashadi; 20060058947 (Mar 16, 2006) Schalk; 20070067088 (Mar 22, 2007) Baig; 20100082232 (Apr 1, 2010) Lee; 20100114471 (May 6, 2010) Sugiyama; 20110050589 (Mar 3, 2011) Yan; 20110054772 (Mar 3, 2011) Rossio et al.; 20110202862 (Aug 18, 2011) Kramer; 20130091971 (Apr 18, 2013) Helot; 20130131986 (May 23, 2013) Van Seggelen; 20130151031 (Jun 13, 2013) Ricci; 20130211710 (Aug 15, 2013) Kennewick; JP 2010145262 (July 2010)."
— https://patents.justia.com/patent/[11168997](/patent/11168997)

Plus non-patent literature: Hölscher et al., "Would you Follow Your Own Route Description? Cognitive Strategies in Urban Route Planning," 2011, Cognition vol. 121, pp. 228–247; Marge et al., "Comparing Spoken Language Route Instructions for Robots Across Environment Representations," SIGDIAL 2010, pp. 157–164; and the EPO Search Report/Office Action and ISR/WO for PCT/EP2013/053749.

Confidence caveat on this list: the Justia page is a secondary mirror. I did not re-verify this list against the USPTO Patent Public Search "References Cited" field. Treat it as highly likely correct but not examiner-certified.

---

1. The §103 framework and the element chart

The governing standard is Graham v. John Deere, 383 U.S. 1 (1966), as refined by KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007). Under KSR, a claim is obvious where the claimed combination is of known elements performing known functions to yield predictable results, and any "articulated reasoning with some rational underpinning" (MPEP §2141–2143) supports the combination — including the seven classic rationales (A–G).

Critical precondition: I must repeat a limitation already flagged in the earlier summary. The verbatim language of claims 1 and 10 was not retrieved in either session. The element chart below is reconstructed from dependent claims 4, 7, 8, 9, 11–18 (quoted via the Justia mirror) plus the specification. Every conclusion below is therefore contingent on that reconstruction. Before this analysis is used for anything consequential, claims 1 and 10 must be pulled verbatim from USPTO Patent Public Search.

The family's commercial framing (from the patent's own Definitions section) is "NAVTEQ Natural Guidance …
improves upon linear navigation instructions … by guiding the way humans instruct each other, through descriptions of orientation points such as distinctive points of interest and landmarks." The '997 is the inverse of that product: rather than generating natural guidance from map data, it ingests a human's natural guidance and outputs machine navigation data.

---

2. Reference catalog — what each cited reference contributes

2a. References whose content I verified this session

US 7,292,978 B2 — Endo (Nov. 6, 2007), "Shortcut names for use in a speech recognition system."
https://patents.justia.com/patent/7292978
Verified content: a vehicle-navigation speech recognition system that receives a destination requested by the user in natural spoken form ("Take me to 4009 Miranda Avenue in Palo Alto, California"), recognizes it with conventional speech recognition, and outputs directions, with the shortcut stored to a "personal profile of the user" (claim 1) and added to a grammar database. Teaches E6 (converting spoken destination utterances into navigation data), E13 (grammar/map database), and — importantly for E14 — the concept of a user profile that biases recognition. Endo's own text: "the navigation system responds to the user's request for directions … by displaying the directions … along with voice instructions or visual instructions."

US 8,538,760 B2 — Khorashadi (Sep. 17, 2013), "Methods and apparatuses for identifying audible samples for use in a speech recognition capability of a mobile device."
https://patents.google.com/patent/US8538760
Verified content (Google Patents): speech recognition tuned to a navigational region based on "context information," where "a navigational region may be associated with an estimated location of the mobile device"; map/CAD data may be received from a location-based service and context extracted from the map information; and the mobile device "may identify a set of audible samples … based, at least in part, on the context information." This is a direct teaching of E14 (context-derived processing: geographic location + navigation data) and E13.

US 8,234,059 B2 — Japanese-origin navigation device (Jul. 31, 2012), "Navigation device."
https://patentimages.storage.googleapis.com/42/c1/68/c04bfe57ed4948/US8234059.pdf
Verified content: a navigation device that lets an operator modify a searched route by touch/drag-and-drop, by pinching a route segment, or by hand-drawing a line on a touch panel, with re-searching. This surfaced in my search for the "Sugiyama" reference (US 2010/0114471) and is at minimum strong evidence of the state of the art in non-verbal, gesture-based interaction with a navigation route (E3-adjacent). I could not confirm that this is the same document/family as US 2010/0114471 (Sugiyama), so I am citing it as independent state-of-the-art evidence, not as the Sugiyama reference itself.

US 10,072,941 B2 — same family (2022-issued sibling of '997 in the chain).
https://www.freepatentsonline.com/[10072941](/patent/10072941).html
Verified: the '941 carries the identical title ("Reverse natural guidance") and identical specification text to the '997. This matters: the specification contains no discussion of the "reverse" concept, no "first person/second person" conversation model, and no claim-differentiating narrative. The "reverse" limitation appears to live entirely in the claims. (This is a §112 written-description/enablement question, not a §103 one, but it bears directly on how broadly the "reverse" hook can be read.)

2b. References I could only identify at title/number level (UNVERIFIED content)

I did not obtain and read the disclosures of: US 7,392,194 (Tanaka); US 2006/0058947 (Schalk); US 2007/0067088 (Baig); US 2010/0082232 (Lee); US 2010/0114471 (Sugiyama); US 2011/0050589 (Yan); US 2011/0054772 (Rossio); US 2011/0202862 (Kramer); US 2013/0091971 (Helot); US 2013/0131986 (Van Seggelen); US 2013/0151031 (Ricci); US 2013/0211710 (Kennewick); JP 2010-145262. I will not assert what any of these disclose. Based on number/date/assignee patterns they are plausibly speech-recognition, multimodal-dialog, gesture, or map-interface art, but that is inference, not evidence. Any §103 position built on them requires pulling the actual documents first.

2c. Non-patent literature (NPL) — the most dangerous art for the "reverse" hook

• Marge et al., SIGDIAL 2010, "Comparing Spoken Language Route Instructions for Robots Across Environment Representations." On its face this is exactly the "reverse" problem: collecting human spoken route instructions and converting them to machine-usable route representations, compared across environment representations. If Marge (or a system built on it) parses a spoken human route description into a machine route, it goes a long way toward E1–E2, E5–E7.
• Hölscher et al., Cognition 2011, "Would you Follow Your Own Route Description? Cognitive Strategies in Urban Route Planning." This is a human-cognition study of how people describe urban routes — directly relevant to E2/E3 (the expectation that human route descriptions contain landmarks, imprecise magnitudes, and deictic/spatial descriptions). This is the kind of reference a petitioner uses to show that the character of a human's spoken route description was well understood before 2012.

---

3. §103 combinations

Combination I — Primary: Endo '978 (+ Marge NPL) for speech-to-route; Secondary: gesture/multimodal nav art (US 8,234,059 or the Sugiyama/Kramer/Ricci family) for the non-verbal element; Tertiary: Khorashadi '760 for context

Mapping: Endo supplies E2/E5/E6/E7/E13/E14(profile); Marge supplies the E1/E5–E7 direction (human spoken route → machine route); the gesture art supplies E3; Khorashadi supplies E14 (context/geographic-location-based speech processing) and E13.

Motivation (KSR rationales A, C, F):

1. Both speech-only and gesture-only navigation input were known; combining input modalities was a predictable, well-known design choice (KSR rationale A — known elements performing known functions). The patent itself concedes the point by describing gesture input as merely another sensor channel: "the input 122 may include one or more sensors adapted to receive and convert the gestural narrative … an optical sensor, e.g. a camera, a motion sensor … a touch or tactile sensor … or a combination thereof."
2. The patent's own stated problem — that speech-only capture is unreliable and that gestures disambiguate speech — is precisely the classic deictic-gesture-plus-speech problem ("head that way" + pointing, which the patent expressly uses as its own example). A POSITA would combine a pointing/gesture recognizer with a speech recognizer to improve route-input accuracy; that is rationale C (known technique to improve a similar device in the same way) with a reasonable expectation of success given the maturity of both.
3. Khorashadi shows the field was already biasing speech recognition by navigational context — providing the motivation for E14 for the same reason (accuracy), rationale F.

Where this combination is weak: it is a single-user model. Endo's user speaks to the device; the gesture art's operator gestures to the device. Neither teaches E1's "conversation between a first person and a second person" where the device is an eavesdropper on a third-party conversation.

Combination II — Combination I + a conversation/meeting-capture or "directions overheard/shared" teaching, to reach E1 and E4

The task: find art where a device passively observes a dialogue between two humans, one instructing the other, and extracts structured data. Candidate families (unverified by me, must be checked): meeting/conversation transcription and summarization (speaker diarization + semantic extraction), and "sharing spoken directions" / hands-free "record my passenger's directions" features. Multimodal-dialog references such as Schalk 2006/0058947 and Kramer 2011/0202862 are plausible places to look, given their dates and the state of IBM/Microsoft multimodal-dialog work; I have not verified their content and will not assert it.

Motivation (KSR rationales D, E, F; In re Kahn / KSR "design incentive"):

1. Use of a known technique (automatic speech/speaker processing) in a known device (a phone in the receiver's pocket) to yield a predictable result — rationale D. The patent itself emphasizes this context: "if the device on which the system 100 is implemented in located in the pocket of the receiver 106 … only the audio input may be capable of receiving that portion of the conversational narrative 102."
2. Predictable variation / obvious to try — rationale E. Once the device is a phone carried by the person receiving directions, capturing the provider's speech is not a new inventive concept; it is the necessary and obvious consequence of where the device is. The patent's claim that the device "does not perform a legal analysis…", i.e. its own framing that the device is in the receiver's possession, undercuts any argument that "listening to two other people" was counter-intuitive.
3. Market/design incentive — rationale F: hands-free, eyes-on-the-road operation. The patent's own Definitions section quotes the industry rationale verbatim from NAVTEQ Natural Guidance marketing: "Research shows consumers desire more intuitive and practical directions because it is easier to follow and allows the user to keep their eyes on the road." A POSITA had every incentive to eliminate the step of the receiver manually re-entering the provider's directions.

Combination III — For dependent claims E11 (accuracy/reliability) and E12 (supply omitted elements)

E12 (omitted elements): Endo/Marge route parsing + Khorashadi navigational-region context + a standard route-completion technique. The patent's own description of how it does this — "the starting and/or ending locations may be identified and used to independently determine one or more navigation routes there between… compared with the navigation route compiled based on the conversational narrative and used to fill in any missing information" — is textbook route-computation-plus-reconciliation, i.e. applying the universally known shortest-path/route-search engine to a known input. Rationale: A (known elements, predictable result). In re Kubin, 561 F.3d 1351 (Fed. Cir. 2009) and KSR both support obviousness where a known method is applied as a matter of ordinary skill.

E11 (accuracy/reliability): speech-recognition confidence scores were standard, and Endo's sibling art expressly computes confidence scores and modifies them with weights (US 7,328,155, Endo et al., "…grammar weighted based upon location information…," per the Endo patent family listing — https://www.patents-review.com/a/20050080632-method-system-speech-recognition-grammar-weighted-based.html). Mapping a recognition confidence to a route-reliability metric is a predictable use of a known technique (rationale C).

E10 (indoor destination): Endo's own specification contemplates non-navigation contexts, and Khorashadi's context layers are explicitly building/shopping-mall/floor/product hierarchies ("a first informational layer may be associated with a shopping mall, a second … a retail store, a third … a particular floor"). That is essentially indoor navigation-data modeling, which makes the "interior destination" dependent claim directly anticipated by or obvious over Khorashadi on its own.

---

4. Rebuttals, secondary considerations, and the honest weaknesses of this obviousness case

The single most important weakness — do not gloss over it. The cited references, to the extent I verified them, are all single-user, user-to-device systems. E1 ("observe a conversation between a first person and a second person") and E4 (requiring the verbal + non-verbal pair in the same observed conversation) are the true novelty hooks. If the prosecution history shows these were added by amendment over the examiner's art (as the number of 2013-dated references in the cited list suggests — i.e., art that the examiner was later reaching for), then the applicant's own file is evidence that the examiner could not find E1/E4 in the art. Global Dossier / the '771→'093→'941→'997 prosecution histories must be pulled before asserting that E1/E4 are obvious. 🔎 That is the decisive verification step.

Counter-arguments a patent owner will raise, and how they fare:

• "The device must recognize a conversation, not a command." — Weak to moderate. Recognizing conversation vs. command is a difference in speaker role and channel, not in the underlying signal processing (ASR + semantic parsing). A POSITA arguably needed only a trigger/segmentation change. But it is a real difference in claim scope and should be met head-on with art, not assumed away.
• "Passive/third-party capture is counter-intuitive." — Weak. The patent's own "device in the pocket of the receiver" passage makes the passive-capture architecture the natural deployment, not a departure.
• "E4 requires the non-verbal element to be related to the verbal one." — This is a multimodal fusion requirement, and multimodal fusion of deictic gesture + referring speech was among the oldest known techniques in the field (the foundational "put-that-there" line of work). Rationale A.

Secondary considerations (Graham factor 4) — none evidenced in what I have. I found no evidence of unexpected results, industry praise, licensing due to the '997 specifically, or long-felt unmet need attributable to E1/E4 as opposed to natural guidance generally (which the patent concedes was already commercialized by NAVTEQ). The 2013–2018 continuation chain is not itself evidence of non-obviousness; it is a common practice for keeping a family pending.

---

5. A threshold issue that changes everything: which art is even available

The '997 claims priority to 2012-03-26 (Ser. No. 13/430,199). Four cited references published after that date:

Under pre-AIA §102(e) (assuming the AIA did not change the applicable regime — the '997's priority chain predates 2013-03-16), these four are prior art only if their filing dates precede 2012-03-26. Similarly, US 8,538,760 (Khorashadi) issued 2013 but appears to claim 2012 priority (a "201213351355" priority claim appears in the related EP/Google family record at https://economie.fgov.be/.../recueil-brevets-2018-28.pdf) — meaning it may or may not qualify. This must be date-checked reference by reference before any §103 position is asserted.

The flip side — a genuine risk to the patent owner. If any claim of the '997 is not supported by the 2012 priority document, its effective filing date slides to 2016 (parent 15/171,118) or 2018 (the '997 itself). Given that the specification text is identical to the sibling '941 and contains no "reverse"/"first person–second person" narrative (see §2a), a defense that all claims get the 2012 date is not self-proving. Moving the date forward would sweep in a large volume of 2013–2017 speech-plus-gesture navigation art. This cuts in the opposite direction from obviousness and should be investigated in parallel.

---

6. What is NOT established here (explicitly)

1. Verbatim claim 1 and claim 10 text, and total claim count. Not retrieved in either session. Everything above is contingent on the reconstruction in §1.
2. Disclosure content of 13 of the 15 cited references. I read Endo '978, Khorashadi '760, and (as independent state-of-the-art) US 8,234,059. I did not read Tanaka, Schalk, Baig, Lee, Sugiyama, Yan, Rossio, Kramer, Helot, Van Seggelen, Ricci, Kennewick, or JP 2010-145262. I have made no assertion about their content.
3. Prosecution history for the '771/'093/'941/'997 chain. This is the single highest-value missing input; it will show exactly which limitations the examiner did and did not find in the art.
4. The "Prior Art section of this page." As flagged in §0, no prior-art section existed in the previously generated material. If the intended source was a different section (e.g., Google Patents "Similar Documents," "Cited By," or the "Landscapes" block), please point me to it and I will re-run the mapping against that specific corpus. I did not use Google Patents' machine-generated "Similar Documents" or "Cited By" lists in this analysis, because those are algorithmic, not examiner-sourced, and are not reliable as §103 art.
5. Priority/§102(e) dates for the four 2013 publications and for Khorashadi. Unresolved (see §5).

---

7. Bottom line

On the record available: the broad architecture of the '997 (natural-language route description → navigation data elements → compiled route) is plainly obvious over Endo '978 in view of the Marge and Hölscher NPL, with gesture/multimodal navigation art (Khorashadi '760 for context; US 8,234,059 and the 2010–2011 multimodal-dialog publications for the non-verbal channel; US 7,328,155-family confidence scoring for the accuracy claim). Every dependent claim I could map (E9–E15) is individually vulnerable as a known-technique application.

The whole §103 case turns on two elements: E1 (the device observes a conversation between two other people) and E4 (the non-verbal element is paired with a related verbal element in that observed conversation) in claim 10. I could not establish on the verified record that any cited reference teaches E1/E4. Without prosecution-history confirmation or an independent eavesdropping-conversation reference, an obviousness challenge to claims 1 and 10 as issued rests on the argument that passive third-party capture is a predictable deployment variation of the known multimodal-input system — an argument that is available under KSR rationales D/E/F but is not yet evidenced. Recommend, in priority order: (1) pull verbatim claims 1/10; (2) pull the four prosecution histories; (3) read Marge 2010 and Hölscher 2011 in full (highest-leverage NPL); (4) read Schalk, Baig, Kramer, Sugiyama, Ricci, Kennewick; (5) run a targeted search for conversation-capture/meeting-summarization art with speaker diarization.