US 10739785

I'll search for authoritative information on this patent number.

Let me run additional targeted searches on the docket/litigation angle and confirm the patent record.

Let me check specifically whether US 10,739,785 itself appears in any 2026 PTAB or Federal Circuit proceeding.

US 10,739,785 B2 — Summary

Important caveat up front: I could not find any 2026 Federal Circuit docket that names U.S. Patent No. 10,739,785 specifically. Despite multiple targeted searches, no CAFC appeal, IPR, ITC action, or district-court complaint that asserts the '785 patent itself surfaced. What I did find is that the '785 patent's prosecution history was filed as an exhibit in a 2026 PTAB proceeding involving a sibling patent. Details below, with the basis for each statement.

Bibliographic data (from the patent document itself)

The '785 patent is a continuation of Ser. No. 15/973,983 (now US 10,338,599), which is a continuation of Ser. No. 15/207,065 (now US 9,995,051), which is a continuation of Ser. No. 13/545,918 (now US 9,388,595), filed 2012-07-10. All priority traces to the 2012-07-10 original filing.

Source: https://patents.google.com/patent/US10739785/en

Abstract (verbatim)

"A pool cleaning system for cleaning debris from a submerged surface of a swimming pool includes a self-propelled pool cleaner having rotatably-mounted supports for supporting and guiding the cleaner on the pool surface; an electric motor for enabling the rotation of the rotatably-mounted supports on the pool surface; at least one camera to capture imagery of the pool surface; a controller, in electronic communication with the at least one camera, to determine a cleanliness characteristic of the pool surface on which the cleaner has passed based on the camera imagery and generate a control signal to direct movement of the cleaner based on the cleanliness characteristic of the pool surface, and a portable electronic device configured to present a graphic on a display, the graphic depicting the submerged surface of the pool and those portions of the surface that remain uncleaned as the cleaner traverses the pool surface."

Note a drafting mismatch worth flagging: the abstract recites a portable electronic device and a "cleanliness characteristic," but the issued independent claim does not. The abstract tracks the specification/earlier-family disclosure rather than the claim set that actually issued.

Independent claim 1 — plain-language overview

There is only one independent claim (claim 1). In plain terms, it covers a self-propelled pool cleaner comprising:

1. A vehicle body;
2. "Means for moving" the body within the swimming pool (means-plus-function style language);
3. A filter configured to retain debris collected from the submerged surface;
4. At least one camera positioned on or in the body, capturing an image of at least a first portion of the submerged surface; and
5. A controller positioned on or in the body, in electronic communication with the camera, configured to generate at least one control signal that causes the body to move within the pool to that first portion of the submerged surface when the captured image fails to satisfy a predefined threshold.

Key practical takeaway: As issued, claim 1 is narrower than the specification's headline embodiments. The specification's principal disclosure describes cameras mounted externally to the pool (including repurposed home/pool security cameras) and a controller located in a dry station, a remote server, or a portable electronic device (FIGS. 2–5). Claim 1, however, requires both the camera and the controller to be positioned on or in the vehicle body — i.e., a fully onboard vision/control architecture. The claim also frames the trigger in terms of an image "fail[ing] to satisfy a predefined threshold," and recites moving the cleaner to the imaged portion (re-cleaning/return-to-location behavior), rather than abstract "cleanliness characteristic" language.

Dependent claims 2–6

• Claim 2 — The "means for moving" is a propulsion mechanism selected from the group consisting of at least one jet or at least one rotatably-mounted support (a Markush group).
• Claim 3 — Where the propulsion mechanism is a rotatably-mounted support, it is selected from the group consisting of a wheel, a roller, a track, or a scrubbing member.
• Claim 4 — Adds an electric motor for driving the means for moving the vehicle body.
• Claim 5 — The controller controls electric power provided to the electric motor.
• Claim 6 — The controller is operable to execute a cleaning program for cleaning the submerged surface.

Because of the "consisting of" language in claims 2–3, those dependent claims are closed groups (e.g., claim 3 does not cover, say, a paddle or a fin propeller).

Litigation / PTAB status (2026)

The Google Patents record carries a "Family has litigation" flag, and that is confirmed by 2025–2026 activity in the pool-cleaner space involving the current assignee, Zodiac Pool Systems LLC (parent: Fluidra North America LLC).

What I found (with sources):

• IPR2026-00202, Beatbot Technology (USA) Co. Ltd. v. Zodiac Pool Systems LLC — filed 2026-01-26, discretionarily denied 2026-05-19. This petition challenges U.S. Patent No. 11,003,191 (a sibling/child in the '785 family), not the '785 patent. Notably, Patent Owner Exhibit 2027 in that proceeding is titled "Prosecution History of U.S. Patent No. 10,739,785." (https://sandbox.docketalarm.com/cases/PTAB/IPR2026-00202/...)
• IPR2026-00158 / -00159 / -00160, filed by Wybotics, Co. Ltd., challenging US 11,003,191; US 11,262,766; and US 11,880,207 respectively — again all children of the '785 family, not the '785 patent. (Petitioner's Stipulation dated 2026-03-31, https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1558741](/patent/1558741)/...)
• District court litigation: Zodiac Pool Systems LLC v. Wybotics, Co. Ltd., No. 2:25-cv-00582-JRG-RSP (E.D. Tex., filed 2025-05-23); Wybotics Inc. v. Zodiac Pool Systems LLC, No. 2:25-cv-08526 CAS (AGRx) (C.D. Cal., filed 2025-09-09); Beatbot Technology (USA) Co. Ltd. v. Zodiac Pool Systems LLC, No. 3:25-cv-01946 (S.D. Cal.). The asserted patents in these matters, as reflected in the PTAB exhibits, are the camera/vision-family children (the '191, '766 and '207 patents).
• ITC Investigation (2022): Zodiac's § 337 complaint against Aiper asserted US 8,393,029 and US 8,393,031 — unrelated to the '785 patent.

Bottom line on dockets: I have no authoritative evidence that US 10,739,785 is currently asserted in litigation, is the subject of an instituting IPR, or is on appeal at the Federal Circuit in 2026. Its only confirmed 2026 appearance is as a prosecution-history exhibit in IPR2026-00202. If you need certainty, the definitive checks would be the USPTO Patent Trial and Appeal Board's Patent Trial and Appeal Case Tracking System (P-TACTS), PACER/EDTX and CDCA/SDCA dockets, and the Federal Circuit's docket — none of which I was able to search exhaustively here.

Uncertainty flags

• I could not independently verify the current assignee beyond the patent record's recorded assignment (Aqua Products → Zodiac Pool Systems LLC, 2020-08-28) and the 2025–2026 litigation records identifying Zodiac Pool Systems LLC with Fluidra North America LLC as corporate parent.
• The "Family has litigation" indicator on the Google Patents page likely reflects litigation over family members (e.g., US 9,388,595's history in Aqua Products v. Zodiac, IPR2013-00159 over the unrelated 8,273,183 patent), rather than over the '785 patent itself.
• The specification's detailed description appears to describe externally mounted cameras and off-board controllers as the primary embodiments, which do not read on issued claim 1's onboard-camera/onboard-controller requirement. This is a genuine scope quirk of the granted claims, not a transcription error — the full claim text is reproduced above.

I'll search for litigation involving US patent 10739785 specifically.

The initial results are unrelated ('397 patents in other tech areas). Let me refine with patent-specific and assignee-based queries for this exact patent.

Let me search more specifically for this patent number in litigation contexts.

Summary of findings — U.S. Patent No. 10,739,785

Bottom line: I was unable to confirm any litigation that expressly asserts U.S. Patent No. 10,739,785 ("the '785 patent") as an asserted patent. There is an active, well-documented worldwide dispute involving its owner (Zodiac Pool Systems LLC / Fluidra) over the same patent family, but the complaints I located name sibling patents in the family (e.g., 11,003,191, 11,262,766, 11,808,207) rather than the '785 patent itself. I am flagging this explicitly rather than inferring that the '785 patent was asserted.

Critical caveat on the number itself

Per the strict no-auto-correction rule, I note that searches on "10739785" / "10,739,785" repeatedly returned unrelated patents whose numbers merely end in "785" — specifically U.S. Pat. Nos. 7,949,785 (Intellectual Ventures cloud patent, asserted against American Airlines) and 9,750,785 (Par/Amneal vasopressin formulation patent). Those are different patents and are not the pool-cleaning '785 patent. I am not treating them as results for 10,739,785.

Family relationship (from the patent page itself)

US 10,739,785 (app. 16/416,840, filed 2019-05-20, priority 2012-07-10, now assigned to Zodiac Pool Systems LLC) is one member of a continuation family sharing the title "Pool cleaning system and method to automatically clean surfaces of a pool using images from a camera." The family includes US 9,388,595; 9,995,051; 10,338,599; 10,739,785; 11,003,191; 11,262,766; 11,880,207; and pending US 2024/0176355A1.

Litigation and contested proceedings located (family-level)

Additionally, a Zodiac regulatory filing references a U.S. International Trade Commission (ITC) investigation alongside the "U.S. Complaint" and "Amended U.S. Complaint" concerning pool cleaners with "camera/vision-related functionality" — the same technology space as the '785 patent. The specific ITC investigation number was not captured in my sources.

Interpretation and recommendation

• The family-level dispute is real and active: Zodiac/Fluidra (patent owner) vs. Wybotics and Beatbot in the U.S., the ITC, and France. Based on the complaint analyses I retrieved, Zodiac asserted the '191, '766, and '207 family members — not the '785 patent — against Wybotics' "Vision Tech" products (Wybot M2, C1/C2 lines, etc.).
• It is plausible that the '785 patent is part of the asserted family in one of these complaints (its claim set requires the controller to be "positioned on or in the body," which narrows it relative to siblings), but I have no source confirming that, so I will not state it as fact.

To confirm definitively, check directly:

• Unified Patents litigation portal (portal.unifiedpatents.com/litigation) — search exact string "10739785."
• PacerMonitor / CourtListener / Justia Dockets — pull the complaint and its exhibit list in Zodiac v. Wybotics, E.D. Tex. 2:25-cv-00582, and the ITC §337 complaint, and check the "Asserted Patents" section for 10,739,785.
• USPTO Patent Center / Global Dossier for the patent's "litigation" tab.

If you can pull the E.D. Tex. complaint exhibit list, I can tell you with certainty whether 10,739,785 is among the asserted patents.

I'll start by checking the canonical structured data (no PTAB proceedings returned by ODP) against public sources.

The ODP block says "no proceedings," but the search surfaced a Docket Alarm entry for IPR2026-00202 involving this patent. Let me verify and check for related filings.

Key finding: the Wybotics/Beatbot IPRs I'm finding target the continuation patents (11,003,191; 11,262,766; 11,880,207), not the '785. Let me confirm whether any petition actually targets 10,739,785.

Proceedings overview

Zero. The USPTO Open Data Portal structured block returns no AIA trial proceedings for US 10,739,785, and targeted web searching of PTAB docket aggregators surfaced no IPR, PGR, or CBM petition ever filed against the '785 patent itself — so the bottom line for a defendant is neither "hardened" nor "dead claims," but untested: all six claims stand unadjudicated, no petitioner is estopped, and every prior-art ground against the '785 remains fully available. The important caveat is that the '785 is the parent of a continuation family that Zodiac is now asserting, and those children are under active IPR attack with art that overlaps the '785's disclosure — which gives a defendant a ready-made, already-briefed prior-art toolkit.

Confidence statement: I could not reach PTAB E2E / Patent Trial and Appeal Board End-to-End directly in this session. My negative finding on the '785 rests on (a) the canonical ODP block and (b) third-party aggregators (Docket Alarm, Ex Parte AI-Lab, GreyB IP Verse). I did not find a single docket entry naming 10,739,785 as the challenged patent. Treat the negative as high-confidence but re-verify at PTAB E2E before relying on it in a filing.

Family-adjacent PTAB activity — ⚠️ NOT proceedings on the '785

I am separating these out deliberately. The user instruction is to flag anything web search surfaces, but none of the proceedings below challenges 10,739,785. Presenting them as '785 proceedings would be inaccurate. They matter because they involve near-verbatim claim language from the same 2012-07-10 priority chain (Google Patents Family ID 49850474).

IPR2026-00158 — Wybotics, Co. Ltd. v. Zodiac Pool Systems LLC

• Type: Inter Partes Review
• Filed: 2025-11-25
• Patent challenged: U.S. Patent No. 11,003,191 — claims 1-3 (not the '785)
• Status: Trial Instituted (institution 2026-05-15, per Ex Parte AI-Lab; GreyB IP Verse corroborates "Trial Instituted" as of 2026-08-24)
• Judge panel: Not public in the sources retrieved — shown as "Pending Judge Assignment"
• Petition grounds (as summarized by Ex Parte AI-Lab, third-party — verify against the petition):
  • Ground 1 — claims 1-3 anticipated by Fu (CN 101139007 B) under § 102
  • Ground 2 — claims 1-3 obvious over Fu in view of Benzler (US 2010/0299016) under § 103
  • Ground 3 — claims 1-3 obvious over Schnittman (US 2012/0169497) in view of Benzler under § 103
• Institution decision: Instituted 2026-05-15
• Final Written Decision: Not issued. Statutory § 316(a)(11) deadline ≈ 2027-05-15
• Settlement / termination: None reported
• Appeal: None — no FWD yet
• Defensive value: Fu (CN 101139007) is the load-bearing reference. If Fu alone or with Benzler takes out the '191 claim 1 — which recites materially the same body/means-for-moving/filter/camera/onboard-controller architecture — that art is the obvious starting point for a '785 petition.
• Sources: https://ai-lab-cl-prod.azurewebsites.net/case/ptab/IPR2026-00158/wybotics-co-ltd-v-zodiac-pool-systems-llc ; https://ipverse.greyb.com/competitive-analysis/company/zodiac-pool-systems

IPR2026-00159 — Wybotics, Co. Ltd. v. Zodiac Pool Systems LLC

• Type: Inter Partes Review
• Filed: 2025-11-25 (same-day companion filing)
• Patent challenged: U.S. Patent No. 11,262,766 — claims 1-5
• Status: Reported as instituted / proceeding on the merits; an institution-decision document dated 2026-04-14 lists IPR2026-00158, -00159 and -00160 together as proceedings "reviewed for merits and non-discretionary considerations." I could not independently confirm each one's institution outcome — treat as unsettled.
• Judge panel: Not public in retrieved sources
• Petition grounds: Not retrieved in this session
• Defensive value: The '766 is a camera/vision continuation of the same '785 priority chain. A win here narrows the family's camera-control claim space.

IPR2026-00160 — Wybotics, Co. Ltd. v. Zodiac Pool Systems LLC

• Type: Inter Partes Review
• Filed: 2025-11-25 (same-day companion filing)
• Patent challenged: U.S. Patent No. 11,880,207 — claims 1-8, 11-14, and 19
• Status: Same caveat as -00159
• Judge panel: Not public in retrieved sources
• Petition grounds: Not retrieved in this session

Petitioner's stipulation (highly relevant to estoppel/§ 315(e)(2)): On 2026-03-31, Wybotics filed a stipulation in all three IPRs agreeing that if the Board institutes, Wybotics and its RPIs/privies will not pursue §§ 102/103 or indefiniteness invalidity grounds in Zodiac Pool Systems LLC v. Wybotics, Co. Ltd., 2:25-cv-582-JRG-RSP (E.D. Tex., filed 2025-05-23) or Wybotics Inc. v. Zodiac Pool Systems LLC, 2:25-cv-08526 CAS (AGRx) (C.D. Cal., filed 2025-09-09). Source: https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1558741](/patent/1558741)/download-documents (PTACTS petition 1558741).

IPR2026-00202 — Beatbot Technology (USA) Co. Ltd. v. Zodiac Pool Systems LLC

• Type: Inter Partes Review
• Filed: 2026-01-26
• Patent challenged: U.S. Patent No. 11,003,191 — claims 1-3 (same patent as IPR2026-00158 — a parallel/competing challenge, not a joinder)
• Status: Discretionary Denial — terminated 2026-05-19 via "Board Director Discretionary Decision: Deny"
• Judge panel: Not public in retrieved sources; "Pending Judge Assignment" per Ex Parte AI-Lab
• Petition grounds (third-party summary):
  • Ground 1 — claims 1-2 obvious over Durvasula (US 2014/0015959) + Fu-007 (CN 101139007A)
  • Ground 2 — claims 1-2 obvious over Cui (CN 1540119A) + Fu-007
  • Ground 3 — claims 1-3 obvious over Knoll (DE 102007053310 A1), standalone
  • Ground 4 — claims 1-3 obvious over Renaud (US 2014/0263087)
• Institution decision: Denied 2026-05-19 on discretionary grounds (not on the merits — the Ex Parte AI-Lab record shows "Outcome: --" with no merits ruling)
• Final Written Decision: None
• Settlement / termination: Terminated by discretionary denial, not settlement
• Appeal: None reported
• Defensive value: The denial means Beatbot is not estopped. Its expert declarations (Li, Yuan) and its art set (Fu-007, Cui, Knoll, Renaud) remain live ammunition. Notably, Patent Owner's POPR relied on the prosecution history of US 10,739,785 (PO Ex. 2027) — so the '785's own prosecution record is already being litigated-by-proxy in the family.
• Sources: https://sandbox.docketalarm.com/cases/PTAB/IPR2026-00202/Beatbot_Technology_%28USA%29_Co._Ltd._v._Zodiac_Pool_Systems_LLC/ ; https://ai-lab.exparte.com/case/ptab/IPR2026-00202/beatbot-technology-usa-co-ltd-v-zodiac-pool-systems-llc

Historical party context (NOT a '785 proceeding)

IPR2013-00159, Zodiac Pool Systems, Inc. v. Aqua Products, Inc. challenged U.S. Pat. No. 8,273,183 (water-jet vector pool cleaner). FWD 2014-08-22 found all instituted claims unpatentable and denied Aqua's motion to amend. That proceeding produced the en banc Aqua Products, Inc. v. Matal, 872 F.3d 1290 (Fed. Cir. 2017) (burden of persuasion on motions to amend shifts to petitioner). Different patent, same corporate parties — cited only for party-history color, not as precedence over the '785.

Strategic summary

Claim status of the '785: 100% untested. Claims 1-6 — independent claim 1 (vehicle body / means for moving / filter / camera on or in the body / onboard controller generating a control signal when the captured image "fails to satisfy a predefined threshold"), plus dependents 2-6 — have never been before the Board. No claim has been canceled; no claim has been sustained; no certificate has issued under § 318(b). Anyone who tells you the '785's claims are "hardened by surviving IPR" is wrong, and anyone who tells you they are "canceled" is also wrong. The correct statement is: no PTAB validity adjudication exists.

Estoppel landscape: wide open. § 315(e)(2) estoppel attaches only to petitioners who are parties to an instituted IPR that reaches FWD. Because no IPR on the '785 was ever instituted, no entity on earth is estopped as to the '785, and there is no § 325(d) "same or substantially the same art" wedge from a prior '785 proceeding. Practically, however, two subtleties matter. First, a future '785 petitioner will have to reckon with art already before the Office in the sibling IPRs — Fu (CN 101139007), Knoll (DE 102007053310 A1), Cui (CN 1540119A), Benzler (US 2010/0299016), Durvasula (US 2014/0015959), Renaud (US 2014/0263087), Schnittman (US 2012/0169497) — which cuts both ways: it is a free, fully-briefed toolkit, but it also raises § 325(d) / Advanced Bionics risk if the art was previously presented. Second, if Wybotics' three instituted IPRs reach FWD, Wybotics and its RPIs/privies will be estopped in the E.D. Tex. and C.D. Cal. cases per its 2026-03-31 stipulation — but that estoppel runs to the '191/'766/'207 claims, not to the '785.

Pattern signals. Two independent Chinese pool-robot manufacturers (Wybotics and Beatbot) filed against Zodiac's camera-control continuation family within ~2 months of each other (2025-11-25 and 2026-01-26), and Zodiac's own infringement contentions (2025-12-09, cited as Pet. Ex. 1023 in IPR2026-00202) drove that. Both were defending Zodiac suits: Zodiac Pool Systems LLC v. Wybotics, 2:25-cv-582-JRG-RSP (E.D. Tex., filed 2025-05-23, amended 2025-09-22 to add the camera patents) and Beatbot Technology (USA) Co. Ltd. et al v. Zodiac Pool Systems LLC, 3:25-cv-01946 (S.D. Cal., DJ action, claim construction set 2026-06-05, pretrial 2027-03-12). Notably, Zodiac asserted the '191, '766, and '207 — but not the '785 — against Wybotics, even though the '785 is their parent. Beatbot's petition was killed at the discretionary-denial stage, and Wybotics chose "institute-and-stipulate" rather than fight. There is no defensive aggregator (no Unified Patents, no RPX) in the chain; this is classic competitor-vs-competitor litigation. Importantly, the '785's priority chain means the same 2012-07-10 specification underlies both asserted children and the unasserted parent — so a successful § 102/§ 103 attack on the children is transferable, but not automatically.

Recommended next steps

1. If you have a demand letter or complaint citing US 10,739,785: there is no FWD to point to, so do not represent to the court or opposing counsel that any claim is canceled. Your leverage is different: the '785 has never been IPR-tested while its children are under fire, which is a strong signal the patent has not been picked as the strongest family member by sophisticated adversaries (both Wybotics and Beatbot chose the '191 instead). Investigate whether that is because the '785's claim 1 "predefined threshold" limitation is harder to invalidate, or because of the pre-AIA/post-AIA or prosecution-history quirks in the chain.

2. Build your petition off the sibling IPRs. The Fu / Knoll / Cui / Benzler / Durvasula / Renaud mappings in IPR2026-00158 and IPR2026-00202 (and the Janét, Li, and Yuan expert declarations) are already drafted against the same specification; the '191 claim 1 recited in the IPR2026-00202 exhibit list is textually near-identical to the '785 claim 1 preamble-plus-elements structure. Re-map them against the '785's actual limitations — particularly "when the captured image fails to satisfy a predefined threshold" — before filing.

3. Watch the 2027-05-15 date. IPR2026-00158 was instituted 2026-05-15, so the Board's FWD is due on or about 2027-05-15 under 35 U.S.C. § 316(a)(11). That FWD will be the first authoritative PTAB reading of this claim family's "camera + onboard controller" concept and will be persuasive (though not binding) in any later '785 proceeding. Confirm institution status for IPR2026-00159 and IPR2026-00160 at PTAB E2E — my sources conflict on whether all three instituted.

4. Verify the negative. Before relying on "no PTAB proceeding on the '785," run a fresh PTAB E2E and USPTO Patent Center search on Patent Number 10,739,785 and on Application Number 16/416,840, and re-check for any petition filed after the ODP ingest date. Third-party aggregators lag the Office by weeks.

5. No appeal to track. No FWD has issued on the '785, so there is no Federal Circuit appeal to monitor on this patent. The only appellate activity in the corporate family is the 2017 en banc Aqua Products decision (different patent, 872 F.3d 1290).

Verification note: items drawn from Ex Parte AI-Lab, Docket Alarm and GreyB IP Verse are third-party summaries and may misstate exhibit numbers, institution outcomes, or elimination of claims. The Wybotics stipulation text was read from the PTACTS-hosted document at https://ptacts.uspto.gov/ptacts/public-informations/petitions/1558741/download-documents and is reliable as to the three challenged patents and the two district-court cases it names.

I'll research the assignment record for US 10,739,785 through USPTO Assignment Center and cross-reference sources.

I hit the tool-step limit before I could pull the underlying USPTO Assignment Center records directly. Below I reconstruct what the patent record and cross-referenced sources establish, and I flag precisely where the reel/frame and correspondent data could not be retrieved rather than guessing at them.

Inventors

• Both inventors appear as assignors in the 2019-08-08 recorded assignment to "AQUA PRODUCTS, INC.," which is the standard inventor→company assignment for a company-owned filing. Their employer at filing is therefore Aqua Products, Inc., the original applicant.
• No departure pattern detectable. Because the '785 patent is a continuation in a family whose original non-provisional (Ser. No. 13/545,918) was filed 2012-07-10, the same two inventors of record span 2012–2019. I found no evidence of either inventor leaving Aqua Products within 12 months of filing, and no post-filing inventor-side assignment beyond the initial company assignment. I did not verify current inventor employment status independently.
• The record does not show a separate inventor assignment for each continuation — Google Patents lists a single 2019-08-08 reassignment event naming both inventors as assignors, which is consistent with a fresh assignment taken for the 2019-filed continuation.

Original assignee

• Aqua Products, Inc. — named on the issued patent as original assignee (and the applicant on the 2019-05-20 filing).
• Primary line of business: manufacturer of robotic/automatic swimming-pool cleaners. The specification's FIG. 9–10 vehicle ("POOL ROVER"-class robotic cleaner with toothed track drive, rotating brushes, power-wash jets, onboard filtration) is Aqua Products' commercial product line. Aqua Products is an operating company that shipped products embodying the claims.
• Current status: acquired. On/around 2020 the business was absorbed into Zodiac Pool Systems LLC (a Fluidra Group company), and the patent assignment to Zodiac was recorded 2020-08-28. I did not find a bankruptcy filing or dissolution; the Aqua Products brand/product line persisted under Zodiac/Fluidra.
• Historical wrinkle worth noting: Aqua Products' 2012 patent (US 8,273,183) was the subject of Zodiac Pool Systems, Inc. v. Aqua Products, Inc. (IPR2013-00159) and district-court litigation (Aqua Products, Inc. v. Zodiac Pool Systems, Inc., 1:12-cv-09342, S.D.N.Y.). In other words, the later assignee (Zodiac) was an adverse party to the original assignee (Aqua Products) before acquiring the portfolio — an unusual but not NPE-indicative fact pattern.

Assignment timeline

Retrieval caveat (must read): I could not retrieve the Reel/Frame numbers or the correspondent of record for either assignment. Multiple targeted searches against the USPTO Assignment Center index and third-party mirrors returned no record page exposing those fields. The two assignment events, their execution/recording dates, conveyance type, and parties are confirmed by the patent record, but the reel/frame and attorney-of-record fields are unknown to me and are marked accordingly. Do not treat the "—" entries below as evidence of absence.

• 2019-08-08 (recorded) / execution date not retrievable — Reel not retrieved / not retrieved

  • Conveyance: Assignment (Google Patents labels it "ASSIGNMENT OF ASSIGNORS INTEREST")
  • Assignor: Kameshwar Durvasula; William Londono Correa (joint inventors)
  • Assignee: Aqua Products, Inc.
  • Correspondent: not retrieved — cannot assess recurrence.
  • Context: Routine original inventor-to-company assignment securing title in the 2019 continuation application (16/416,840) for the applicant-employer.

• 2020-08-28 (recorded) / execution date not retrievable — Reel not retrieved / not retrieved

  • Conveyance: Assignment (Google Patents labels it "ASSIGNMENT OF ASSIGNORS INTEREST, SEE DOCUMENT FOR DETAILS")
  • Assignor: Aqua Products, Inc.
  • Assignee: Zodiac Pool Systems LLC
  • Correspondent: not retrieved — cannot assess recurrence.
  • Context: Corporate acquisition / portfolio transfer — Aqua Products' assets (including this patent family) conveyed to Zodiac Pool Systems LLC (Fluidra Group). This is an operating-company-to-operating-company transfer, not a transfer to an asserter vehicle.

Recorded assignment events per the patent record: https://patents.google.com/patent/US10739785/en (Reassignment entries dated 2019-08-08 and 2020-08-28). Verification endpoint for the missing fields: https://assignmentcenter.uspto.gov/ (search patent number 10739785).

Timeline diagram

timeline
    title Ownership of US 10739785
    2012 : Priority application filed by Aqua Products
    2019 : Continuation filed 20 May
         : Inventors assign to Aqua Products 8 Aug
    2020 : Issued 11 Aug
         : Aqua Products assigns to Zodiac 28 Aug
    2025 : Zodiac sues Wybotics and Beatbot

NPE / troll-pattern signals

1. Shell-entity transfer — not present. Both links are to/from named, operating, product-shipping manufacturers (Aqua Products, Inc.; Zodiac Pool Systems LLC). No "IP / Holdings / Ventures / Licensing" entity appears. Conveyance context is acquisition, not transfer to a licensing vehicle.

2. Known asserter in the chain — not present. Neither assignee matches any listed NPE (Acacia, Marathon, IV, IPNav, Wi-LAN, Conversant/Mosaid, Vringo, Pendrell, Round Rock, Spangenberg entities, etc.). Zodiac Pool Systems LLC is a large pool-equipment manufacturer (Fluidra), not an NPE.

3. Repeat correspondent across the chain — unclear / not assessable. The correspondent-of-record field was not retrievable for either recording. There is no basis to allege a repeat-player recording attorney, and equally no basis to rule one out. This signal must be re-run directly against the Assignment Center before any conclusion.

4. Cascading transfers — not present. Two recorded events over ~12 months, but they are (a) an inventor→company assignment and (b) a single corporate acquisition — not a chain of LLC-to-LLC hops. No shared-anonymous-address or common-principal indicators.

5. Pre-litigation transfer — not present. The transfer to Zodiac was recorded 2020-08-28; the first suits involving the family's asserted children (Zodiac v. Wybotics, etc.) were filed in 2025 — roughly five years later. No 6-month proximity to a first suit naming this patent.

6. Bankruptcy fire-sale — not present. No Chapter 7/11 proceeding for Aqua Products appears in the record; the transfer to Zodiac reads as a going-concern acquisition.

7. Privateering — not present. No evidence that an operating company parked the patent with a litigating NPE. The current assignee (Zodiac) is itself the litigant and itself the product maker — the opposite of privateering.

8. Defensive aggregator — not present. Chain terminates at Zodiac Pool Systems LLC, an asserting operating company, not at RPX / AST / LOT / Unified / OIN.

Verdict

Operating-company assertion.

Justification: the chain is exactly two links — the 2019-08-08 inventor→Aqua Products assignment and the 2020-08-28 Aqua Products→Zodiac Pool Systems LLC acquisition — with no shell entity, no NPE assignee, no cascading LLC transfers, and no pre-litigation transfer. The current owner, Zodiac Pool Systems LLC (Fluidra Group), is a major pool-equipment manufacturer that ships cleaners embodying the claimed subject matter and is actively suing competitors (Wybotics, Beatbot) in the same family; the '785 patent's claims are onboard-camera/onboard-controller pool-cleaner claims squarely consistent with a product-commercializing owner. The only unverified variables are the reel/frame numbers and correspondents, which I could not retrieve from the Assignment Center and which should be pulled directly before this ownership chain is treated as final: https://assignmentcenter.uspto.gov/.

Cross-check against prior section: This verdict is consistent with the earlier litigation analysis — the '785 patent itself is a sibling (not the asserted child) in the family, but the assignee identity that drives the call (Zodiac Pool Systems LLC asserting the camera/vision children) is the same, so the ownership-chain conclusion and the litigation-status conclusion align. No contradiction found.

I'll start by verifying the patent record and then research the key prior-art references cited on the face of the patent.

I have strong material on several key references. Let me research the remaining cited references that bear most directly on the '785 claims.

Prior-Art Analysis — U.S. Patent No. 10,739,785 B2

0. Scope control and source note

I searched for and analyzed U.S. Patent No. 10,739,785 specifically (title: "Pool cleaning system and method to automatically clean surfaces of a pool using images from a camera"; App. No. 16/416,840; filed 2019‑05‑20; issued 2020‑08‑11; original assignee Aqua Products, Inc.; current assignee Zodiac Pool Systems LLC). I did not substitute the closely numbered siblings (e.g., 10,338,599; 10,739,785's own children 11,003,191 / 11,262,766 / 11,880,207), except where I flag them explicitly as family references.

The authoritative citation lists I worked from are the "Patent Citations (35)" and "Non-Patent Citations (9)" blocks printed on the face of the '785 record (https://patents.google.com/patent/[US10739785](/patent/US10739785)/en). Because I hit a tool‑call limit mid‑analysis, a handful of the more remote references could not be pulled in full text; for those I say so rather than inventing disclosures.

⚠️ Cross-reference flag / correction to the prior summary. The previously generated summary said the Google Patents "Citations (29)" list carries three examiner‑starred references. The fuller "Patent Citations (35)" block on this page actually stars four: US 7,362,351 B2 (Vision Iq), US 2011/0049023 A1 (Hui), US 2011/0290707 A1 (Porat), and US 2013/0152970 A1 (Aquatron). I treat all four as the examiner‑cited core art. Everything else is third‑party/background art or family art.

1. The legal test being applied (why so few references can anticipate)

The '785 patent's effective filing date is 2012‑07‑10 → pre‑AIA 35 U.S.C. §§ 102/103 govern. Anticipation under § 102 requires a single reference disclosing every element of the claim, arranged as claimed; anything short of that is at best § 103 fodder.

Only one independent claim issued. Its limitations (verbatim, abbreviated) are:

The two gating limitations are 1.d and 1.e: both the camera and the controller must be onboard the cleaner body, and the trigger is image‑vs‑threshold. Any reference with an off‑board camera or off‑board controller cannot, standing alone, anticipate claim 1 — this is the same scope quirk flagged in the earlier section, and it is decisive here.

Dependent claims: 2 (means = jet or rotatably‑mounted support), 3 (support = wheel/roller/track/scrubbing member), 4 (electric motor), 5 (controller controls motor power), 6 (controller executes a cleaning program). Claims 2–3 use closed "consisting of" language.

2. Tier 1 — References that can actually reach § 102 against claim 1

2.1 CN 101139007 A ("Fu") — Underwater cleaning robot

• Full citation: CN 101139007 A (also cited as CN 101139007 B), Beijing Institute of Technology.
• Priority 2007‑07‑31; published 2008‑03‑12.
• Description (corroborated): A self‑propelled underwater/pool cleaning robot with monitoring cameras that constitute the "environmental perception layer"; an image‑acquisition card feeds a central processor that "autonomously plans paths and actions … to drive the robot['s] execution layer." Critically, Fu sets a "pool cleanliness image threshold": the robot "uses its monitoring cameras … to capture images of the pool floor and compares them with the threshold to determine the level of cleanliness," then plans cleaning movements / adjusts wheel speeds accordingly (https://ai-lab-cl-prod.azurewebsites.net/case/[ptab](/ptab)/IPR2026-00158/doc/1001).
• § 102 assessment: HIGH relevance to claim 1 (and by extension claims 2, 4, 5, 6). Fu appears to disclose 1.a–1.e in a single reference: vehicle body, drive means (wheels/motors), debris collection, onboard cameras, and an onboard central processor generating drive signals in response to an image threshold. This is the single most dangerous § 102 reference for the '785 family and has in fact been pleaded as an anticipatory reference against sibling claim 1 in IPR2026‑00158 (Exhibit EX1007 "Fu").
• Caveat: Fu is a Chinese application; its availability as prior art turns on the § 102(e)/§ 102(a)/(b) path (publication date 2008‑03‑12 is >1 yr before 2012, so it also qualifies as a printed publication). I did not independently re‑read the full CN text in this session.

2.2 US 2010/0299016 A1 ("Benzler") — Activation system for a robotic vehicle

• Full citation: U.S. Patent Application Publication No. 2010/0299016 A1, Robert Bosch GmbH.
• Priority 2007‑06‑21; published 2010‑11‑25.
• Description (corroborated): Discloses an activation system for a robotic vehicle "2" including a digital camera 8 that captures an image of the work area, and a logic unit 10 that evaluates the image data to generate driving instructions. In a "particularly preferred" embodiment the robotic vehicle is a pool‑robotic vehicle that "drive[s] on the bottom of a pool"; the logic unit (which may be internal) "automatically calculates a trajectory … based on the image data" and the unit may sit in a moisture‑sealed housing (https://ai-lab-cl-prod.azurewebsites.net/case/ptab/IPR2026-00158/doc/1001; EX1013).
• § 102 assessment: MEDIUM‑HIGH for claim 1. Benzler appears to disclose a robotic (pool) cleaner with an onboard camera + onboard logic unit generating drive commands from the image — i.e., limitations 1.d and 1.e. Its clean‑anticipation strength depends on whether a filter (1.c) and a "means for moving" in the claimed sense are disclosed for the pool embodiment. If any element is missing, Benzler is a § 103 reference (it was in fact relied on in an obviousness combination against the sibling '191 patent). Not a clean anticipate‑all on the record I have.

2.3 US 2013/0151061 A1 / WO 2012/023676 A1 ("Hong") — Cleaning robot and underwater sediment cleaning apparatus and method

• Full citation: US 2013/0151061 A1 and WO 2012/023676 A1, Korea Institute of Robot & Convergence (Pohang Institute of Intelligent Robotics).
• Priority 2010‑08‑20; published 2012‑02‑23 (WO) / 2013‑06‑13 (US).
• Description: An unmanned water‑tank cleaning robot with a body, driver/screws, a suction part, a camera (168) that photographs the cleaning operation/environment, sensors, and a controller that is located OUTSIDE the water tank feeding an operator console with a display (https://patentimages.storage.googleapis.com/fc/7e/60/b66efe8276395c/WO2012023676A1.pdf).
• § 102 assessment: LOW for claim 1 — fails limitation 1.e. The controller is off‑board (external console), so it does not read on the "controller positioned on or in the body" requirement. It remains useful as background/§ 103 art for pooled‑camera pool cleaning and for claims 2–4.

3. Tier 2 — Highly relevant but not clean § 102 anticipation (mostly § 103 / element‑level art)

US 5,613,261 A ("Cleaner") — Minolta Co., Ltd.; priority 1994‑04‑14; issued 1997‑03‑25 (Kawakami et al.). Self‑propelled floor cleaner with a dirt/reflectivity sensor 19, onboard CPU/controller 20, and a driving controller that alters travel/cleaning in response to the degree of dirt; it also compares dirt before/after cleaning and returns to a spot that is still dirty (http://www.everypatent.com/comp/pat5613261.html). Assessment: discloses 1.a, 1.b (rotatable supports/motor), an onboard controller, and threshold‑based return behavior — but the sensor is a reflectivity photodiode array, not a "camera" capturing an image, and there is no pool/debris filter combination. So: not a § 102 anticipation of claim 1; strong § 103 material (it was in fact proposed as a § 103 ground in third‑party analysis: https://www.zyllaw.com/download/US_Patent_Litigation_and_Infringement_Risk_Analysis_Report_EN.pdf).

US 2004/0040581 A1 ("Azoteq" / Bruwer) — Azoteq (Pty) Ltd; priority 2001‑02‑28; published 2004‑03‑04. Suction‑type pool cleaner with onboard control means, sensors that "may be light sensitive, or detect contrasting colors which are indicative of areas to be cleaned or avoided," and the controller can "detect dirt in order to move away from surfaces in the pool which are already clean" (https://patents.google.com/patent/US20040040581A1/en). Assessment: relevant to 1.c/1.e behavior and to dirt‑detection steering, but the sensing is color/reflectivity‑based and the propulsion is suction/water‑flow, not the claimed electric‑motor self‑propulsion; element‑level § 103 art, not § 102.

US 2011/0049023 A1 (Hui) — Pool cleaning vehicle having improved logic; priority 2009‑08‑31; published 2011‑03‑03. Pool cleaner with event/timing logic (mercury‑switch wall detection, pump shut‑off on approach to the water surface, randomized climb times) to avoid turtling (U.S. 8,623,201) (https://patents.google.com/patent/US20110049023). Assessment: relevant to claim 6 (executing a cleaning program) and generally to autonomous pool‑cleaner control, but it has no camera/image‑threshold element; § 103 only.

US 2013/0152970 A1 (Aquatron Robotic Technology Ltd.) — Automatic pool cleaner for cleaning a pool with minimum power consumption and method thereof; priority 2011‑12‑19; published 2013‑06‑20. Pool cleaner with power‑optimizing control logic. Assessment: background/§ 103 for autonomous cleaner control; not a § 102 reference against claim 1.

US 7,362,351 B2 (Vision Iq) — Method, system and device for detecting an object proximate to a water/air type interface; priority 2000‑12‑06; issued 2008‑04‑22. Vision‑based detection of objects relative to a water/air interface. Assessment: relevant to the camera/vision element (1.d) and to waterline detection, but it is a detection system, not a cleaner; not § 102 against claim 1.

US 2011/0290707 A1 (Joseph Porat) and US 8,007,653 B2 (Aquatron) — pool cleaner with integral chlorine generator / related Porat pool‑cleaner disclosures; priority 2004‑09‑15; published/issued 2011. Assessment: relevant to claims 4–5 (electric motor; controller controlling motor power) and to onboard electronics in pool cleaners; § 103 only for claim 1.

US 2014/0009748 A1 (Leonessa) — Pool Cleaner with Laser Range Finder System and Method; priority 2012‑06‑27; published 2014‑01‑09. Only ~2 weeks before the '785 priority date; it is pre‑AIA § 102(e) art as of its 2012‑06‑27 filing (published later). Assessment: laser‑range‑finder navigation for a pool cleaner; relevant to autonomous pool‑cleaner navigation (claims 1.b/1.e) but not camera‑image‑threshold control; not a § 102 reference to claim 1, potentially § 102(e)/§ 103 for narrower interrogations.

US 2010/0307545 A1 (submersible cleaning robot with camera/monitor, referenced in the WO '676 search report) and US 5,205,174 A (Silverman, "Scavenger submersible visual and acoustical tank inspection") — priority 1991‑05‑24; issued 1993‑04‑27. Silverman is a submersible visual inspection apparatus (camera‑bearing ROV) for tanks. Assessment: relevant to the "camera on a submerged vehicle" concept (1.d) but it is an inspection tool with no cleaning filter/threshold‑based pool navigation; § 103 only (Silverman is itself cited as EX1016 in the sibling IPR).

4. Full citation-by-citation table (all 35 "Patent Citations")

Dates are Priority / Publication(or issue) exactly as printed on the '785 record. "§ 102?" column = could this reference alone anticipate a '785 claim as issued?

★ = examiner‑starred on the record. Items #32–34 are the '785's own ancestor applications (continuation chain) and cannot be prior art against it. Item #35 published after the '785 priority date and therefore is not prior art at all.

5. Non‑Patent Citations (9) — prosecution documents, not prior art

All nine are the applicant's/examiner's own prosecution papers and are not § 102 prior art: the Spanish Patent Office search report (ES 201331047, 2015‑02‑03) and the Office Actions / Notices of Allowance for U.S. App. Nos. 13/545,918, 15/207,065, and 15/973,983 (2015–2019). They matter only as a roadmap to which art the Examiner actually relied on — and the plain takeaway is that the applications issued largely without substantive prior‑art rejections (corroborated by the sibling‑patent IPR petitions, which characterize the examiner as having performed "essentially no substantive review," issuing only a double‑patenting rejection overcome by terminal disclaimer).

6. Bottom line

• Only two cited references can plausibly anticipate claim 1 alone: CN 101139007 A ("Fu") (strongest — onboard cameras + onboard central processor + explicit "pool cleanliness image threshold") and US 2010/0299016 A1 ("Benzler") (onboard digital camera + onboard logic unit generating drive instructions, with an express pool‑robot embodiment). Fu in particular has already been asserted as an anticipatory reference against a sibling claim in IPR2026‑00158, and it is the reference I would expect to be run against the '785's claim 1.
• The examiner‑starred references (Vision Iq US 7,362,351; Hui US 2011/0049023; Porat US 2011/0290707; Aquatron US 2013/0152970) are, on their face, § 103 / element‑level art, not anticipatory art.
• Hong (WO 2012/023676 / US 2013/0151061) and US 5,613,261 are the next‑most‑relevant but each fails at least one claim‑1 limitation (Hong: off‑board controller; Minolta: sensor is not a camera).
• The family members and JP6162905B2 are not prior art (same family / later date).

7. Uncertainty and limitations (stated explicitly)

1. Tool limit. I could not retrieve full specimens of several references (e.g., US 5,205,174 Silverman, the Aquatron '970 disclosure, JP S61‑62905, WO 2009/031361) before hitting the step cap. For those I relied on title, assignee, date, and corroborating secondary excerpts only — I did not fabricate disclosure content.
2. § 102 conclusions are provisional. A true § 102 finding requires element‑by‑element mapping against the reference's full text. My "Yes/Possibly" ratings for Fu and Benzler rest largely on PTAB‑petition excerpts describing those documents (EX1007 "Fu," EX1013 "Benzler") rather than my own re‑reading of the originals; treat them as high‑confidence but not certified.
3. Pre‑AIA assumption. I assumed pre‑AIA §§ 102/103 because the '785 claims priority to a 2012‑07‑10 filing. If any issued claim were held not to be entitled to that priority date, the analysis (and the applicable art) would shift to AIA §§ 102/103 as of the 2019‑05‑20 filing, changing which publications qualify.
4. Date inconsistency noted, not resolved. Your task header says "Current Date: April 26, 2026," while the environment date is 2026‑10‑01; this does not affect the substantive prior‑art analysis but may matter for any live‑docket assertions.

I'll ground this obviousness analysis in the actual prior-art references cited on the '785 patent's face. Let me pull specifics on the key references.

Obviousness Analysis — U.S. Patent No. 10,739,785 B2 under 35 U.S.C. § 103

Note on statutory framework: The '785 patent's effective filing date is July 10, 2012 (continuation chain to Ser. No. 13/545,918), which predates the March 16, 2013 AIA first-inventor-to-file cutoff. Accordingly, pre-AIA 35 U.S.C. § 103(a) and pre-AIA § 102 govern. This matters because references such as US 2014/0009748 A1 (Leonessa) and WO 2012/023676 (Pohang) are available as § 102(e) art based on their pre-2012-07-10 filing/provisional dates, even though they published after the '785 priority date.

Per the earlier summary, treat as authoritative: claim 1 is the sole independent claim and requires (i) a vehicle body, (ii) means for moving, (iii) a filter, (iv) a camera on or in the body, and (v) a controller on or in the body that causes the body to move to the imaged portion when the captured image fails a predefined threshold.

1. Person of Ordinary Skill in the Art (POSITA)

A POSITA here would hold a bachelor's degree in mechanical, electrical, or robotics engineering (or equivalent) and have roughly 2–3 years of experience designing or programming automated pool cleaners or comparable autonomous surface-cleaning robots. That skill level is important: image-based condition detection, threshold comparisons, and closed-loop steering were all routine engineering by 2012 (see § 5 below).

2. Graham factors — scope of the claimed invention

The claims are narrow in one dimension and broad in another:

• Narrow: claim 1 requires an onboard camera and an onboard controller that closes a control loop back to the imaged location. The specification's headline embodiments (FIGS. 2–5) instead place the controller in a dry station, remote server, or portable device with externally mounted cameras. As flagged in the prior section, the issued claim does not read on those embodiments.
• Broad: the "means for moving" limitation is a § 112(f) means-plus-function term, construed to cover the corresponding structures disclosed (wheels, rollers, tracks, jets) and equivalents. The "predefined threshold" is likewise recited at a high level of generality — no specific metric, algorithm, or value is required.

3. Element-by-element availability in the cited prior art

No single cited reference discloses every element, so this is a § 103 (not § 102) case. But each element is squarely taught:

The single element that is not clearly in the cited art in image-based form is the "capture image → compare to threshold → return to that location." Minolta supplies the threshold-plus-return logic (with an optical reflectivity sensor, not a full camera), and Silverman/Leonessa/Pohang supply the onboard camera and camera-driven movement. Replacing Minolta's reflectivity sensor with the camera already on the vehicle — or adding Minolta's comparison logic to the camera-equipped vehicles — is the crux of the obviousness case.

4. Concrete combinations rendering claim 1 obvious

Combination 1 (strongest): Silverman '174 + Minolta '261, optionally with a swimming-pool cleaner reference (H-Tech '031 / Polaris '817 / Aqua Products WO 2004/019295)

• Silverman discloses a submersible, self-propelled, tracked vehicle that simultaneously cleans (scrubs/vacuums) and inspects a submerged surface; it has an onboard navigation camera 6, onboard sealed electronics 13, a vacuum + filter returning clarified liquid, and expressly states many functions "could be performed on-board the vehicle, using microprocessors … and customized programming." Source: https://patents.google.com/patent/[US5205174A](/patent/US5205174A)/en
• Minolta '261 discloses a self-propelled cleaner with (a) non-contact optical sensing of surface condition, (b) a stored reference value for a clean surface, (c) a comparison unit computing degree of dirt from the difference, (d) a map of condition values by position, and (e) a controller that modulates travel/cleaning based on the measured degree of dirt and can revisit/judge areas. Source: https://patents.google.com/patent/[US5613261A](/patent/US5613261A)/en
• Motivation: both are autonomous surface-cleaning vehicles; both address the same problem (wasting time on clean areas / inadequately cleaning soiled areas). Applying Minolta's dirt-threshold-and-response logic to Silverman's camera-equipped submersible cleaner is the predictable combination of two known elements to achieve a known benefit — precisely the KSR "predictable variation" and MPEP 2144.1(a) "combination of prior art elements according to known methods to yield predictable results" rationales.
• Adding a pool-cleaner reference (H-Tech US 5,435,031 automatic pool cleaning apparatus; Polaris US 2007/0094817; or the assignee's own Aqua Products WO 2004/019295, "Pool cleaner with on-board water analysis, data recording and transmission device") supplies the "swimming pool" context and a conventional filter/scrubber/steering architecture, to the extent Silverman's "liquid reservoir" (nuclear/industrial tank) is argued not to be a swimming pool.

Combination 2: Leonessa '748 + Pohang WO 2012/023676 (or US 2013/0151061 / US 9,074,386) + Minolta '261

• Leonessa (priority 2012-06-27; a § 102(e) reference 13 days before the '785 priority date) is the most direct teaching of the claim's onboard architecture: a swimming-pool cleaner with a chassis mounted camera, an onboard controller "in communication with the camera" that "control[s] movement of the pool cleaner based on output from the camera," a filter assembly, and a stated ability to "identify specific locations of debris within the environment" and "optimize cleaning routes." Source: https://patents.google.com/patent/US20140009748
• Pohang teaches an underwater cleaning robot with a body-mounted camera 168, image processing, and an automatic cleaning unit that drives the robot along a route and handles the "cleaning area"; it also detects sediment quantity to control "the moving direction of the body … and whether the water tank cleaning operation continues." Source: https://patents.google.com/patent/WO2012023676A1/en
• Minolta again supplies the explicit threshold + re-clean algorithm.
• Motivation: Leonessa itself frames the goal as detecting debris locations and optimizing routes; a POSITA seeking to act on that detected debris would naturally incorporate a threshold criterion (Minolta) and an automatic return/route correction (Pohang) to prioritize soiled areas — an obvious design choice backed by MPEP 2144.1(f) (design incentive/market pressure) and the "finite number of identified, predictable solutions" rationale.

Combination 3 (fallback): H-Tech '031 (or Polaris '817 / Hui US 2011/0049023) + Silverman '174 + Minolta '261 + Aqua Products WO 2004/019295

Uses a conventional self-propelled swimming-pool cleaner as the base, Silverman for the onboard camera and onboard microprocessor suggestion, Minolta for the threshold/re-clean logic, and Aqua Products' own WO 2004/019295 for onboard instrumentation and data transmission. This is a broader, more defensible combination if the patent owner argues that Silverman's industrial-tank vehicle is non-analogous art (an argument that should fail under In re Klein / In re Bigio because both are submerged-surface autonomous cleaning vehicles — the same field of endeavor).

5. Why a POSITA would have been motivated — rationales on the record

Grounding motivation in the references themselves (not hindsight):

1. Same field / same problem. Silverman, Minolta, Leonessa, Pohang, H-Tech, and Polaris all address autonomous cleaning of submerged or floor surfaces and the inefficiency of uniform or random cleaning.
2. Express onboard-processing suggestion. Silverman: functions "could be performed on-board the vehicle, using microprocessors, storage media, and customized programming," and an "on-board microprocessor would allow … pre-program[ming] … for automatic operation." This alone supplies motivation to move the controller onto the vehicle.
3. Image analysis of underwater/pool scenes was known. The patent's own citation set includes US 7,123,746 (Poseidon, detecting an object in relation to a surface) and US 7,363,251 (Vision IQ, detecting an object proximate to a water/air interface), and its family citations include US 2004/0052418 (probabilistic image analysis) — showing that using image-processing to evaluate a scene (including underwater) and detect/respond to conditions was routine well before 2012.
4. Known technique applied to a known structure. Image thresholding and closed-loop steering are known techniques; applying them to a camera already mounted on a pool cleaner is the predictable use of a prior-art technique to improve a similar device (MPEP 2144.1(b)).
5. Design incentive / market pressure. Pool cleaners of this generation competed on cleaning efficiency, coverage, and runtime — motivating concentrated cleaning of dirty regions rather than uniform passes (KSR).

6. Dependent claims 2–6

All dependents are readily obvious on the same combinations:

• Claim 2 (jet or rotatably-mounted support): Silverman treads 5; Leonessa propulsive scrubber assemblies 34/36 and water-jet outlet nozzle assemblies 42; conventional wheeled/roller pool cleaners (H-Tech, Polaris). Both alternatives are literally disclosed in the cited art.
• Claim 3 (wheel, roller, track, scrubbing member): Silverman "treads 5" (track) and Leonessa "brush 56" (scrubbing member); rollers are ubiquitous in the cited pool-cleaner art. Closed Markush group is met.
• Claim 4 (electric motor): Silverman "reversible, stall-protected motor 4"; Leonessa "vibration motor 60." Trivial addition.
• Claim 5 (controller controls power to motor): Pohang's controller "power supply 310 … controlling the power supplied to the cleaning robot" and its wheel/pump drivers; Silverman's remote-controlled speed/direction regulation. Obvious.
• Claim 6 (controller executes a cleaning program): Pohang "automatic cleaning unit 354 … controls the cleaning robot 100 to move along a predetermined route"; Leonessa's SLAM-based route operation; Silverman's pre-programmed automatic operation; Hui US 2011/0049023 "improved logic." Obvious.

7. Weaknesses, counter-arguments, and honest caveats

A robust analysis must flag where the record is thin or contested:

1. The "image vs. threshold" link is the pressure point. Minolta's threshold logic uses a reflectivity sensor, not a camera image. A patent owner will argue that none of the cited references teaches using a camera image to judge cleanliness (Silverman's camera is a navigation camera; Pohang's camera "photographs the cleaning operation … the environment … the overall operating state"; Leonessa's camera identifies objects/obstacles for mapping). The obviousness case is strongest if the petitioner pairs the camera-based references with an explicit image-analysis-for-condition reference (e.g., Vision IQ '251 or Poseidon '746, both in the citation set) to bridge the gap. Without that bridge, the combination is vulnerable to a hindsight objection — a concern the earlier search results also flagged for analogous cases in this family (noting "evident hindsight risk" where the claim-read is assembled only after reading the patent).
2. Non-analogous art (Silverman). Expect an argument that Silverman's industrial/nuclear tank vehicle is non-analogous to a swimming-pool cleaner. Rebut with the shared field (submerged-surface autonomous cleaning) and by leading with Leonessa/Pohang (both expressly underwater-cleaning robots) rather than Silverman.
3. Bodily-incorporation fallacy. The references need only be combined in the claim sense (In re Keller; In re Mouttet); they need not be physically retrofitted. Make this explicit.
4. Secondary considerations. No evidence of unexpected results, commercial success with nexus, licensing, or industry praise was identified in the prior sections. Absent such evidence, the strong prima facie case under KSR should control.
5. Prosecution history suggests the examiner never saw the key art. Each of the four family applications (13/545,918; 15/207,065; 15/973,983; 16/416,840) was allowed after a single non-final rejection (Non-Patent Citations list). This is consistent with the examiner not having combined, e.g., Leonessa + Minolta, or Silverman + Minolta — which is precisely the combination a petitioner would present under § 103.
6. Unverified references. I could not, within this session, independently verify the full text of H-Tech US 5,435,031 / US 5,507,058 and Polaris US 2007/0094817 (searches returned the citation data but not their specifications). Treat their substantive descriptions in § 3–4 above as expected from the references' titles and field rather than confirmed verbatim. The Silverman, Minolta, Leonessa, and Pohang disclosures are confirmed from the sources cited.
7. AIA/§ 325(d) and § 311 note (procedural): if this analysis feeds an IPR, expect Patent Owner to invoke § 325(d) to argue the asserted art or substantially the same art was already before the Office (much of it is of record). A petitioner should therefore design the ground around references not previously applied in combination (e.g., Leonessa + Minolta, or Pohang + Minolta), which strengthens both the merits and the § 325(d) posture.

8. Bottom line

• Claim 1 is obvious under pre-AIA § 103(a) over the primary combination Silverman US 5,205,174 + Minolta US 5,613,261 (plus a swimming-pool cleaner base such as H-Tech US 5,435,031, Polaris US 2007/0094817, or Aqua Products WO 2004/019295), and independently over Leonessa US 2014/0009748 + Pohang WO 2012/023676 + Minolta US 5,613,261. Both combinations supply every element, and motivation is grounded in the references' own express statements (Silverman's onboard-microprocessor/pre-program teaching; Leonessa's camera-driven movement and debris-location detection; Minolta's threshold-and-rec lean logic).
• Claims 2–6 are obvious over the same combinations, with each added feature expressly disclosed in the cited art (treads/wheels/jets/brushes; motors; controller power regulation; cleaning programs).
• The pivotal vulnerability of the obviousness case is the leap from "camera" to "image failing a cleanliness threshold." A petition or opinion must bridge that gap with the image-analysis references already in the citation set (Poseidon US 7,123,746; Vision IQ US 7,363,251) or with an analogous image-based condition-detection reference, and should lead with the underwater-cleaning references (Leonessa, Pohang) to blunt any non-analogous-art argument against Silverman.

Confidence: High on the element mapping for Silverman, Minolta, Leonessa, and Pohang (sources fetched). Medium on the H-Tech and Polaris content (not independently verified here). The combination conclusions are preliminary and would need a full text review of each reference and confirmation of the '785's effective priority date before being relied upon for a filing or an invalidity opinion.