US 11378396

I'll search for authoritative information on this patent, including any 2026 litigation activity.

Let me verify with targeted searches for litigation and the patent's parties.

I have the specification but need the actual claim text. Let me retrieve it.

Important scope note up front

I could not retrieve the granted claim text for US 11,378,396 from an authoritative source in this session, and the full patent text supplied to me is truncated inside the Detailed Description (it ends mid-sentence at "…may be used to estimate translational"). The bibliographic data, abstract, and specification below are grounded; the claim overview is inferred and explicitly flagged as unverified. Also, no CAFC 2026 docket entry or USPTO proceeding listing this patent number was found (see "Litigation / docket search" below) — the searches for "11378396" returned only unrelated hits (e.g., a UK Companies House number, unrelated patents).

1. Bibliographic data (grounded)

Continuity chain (from the specification's Cross-Reference section):

• 16/586,933 (this patent) — continuation of →
• 15/291,576, filed Oct. 12, 2016, now US 10,429,507 — continuation of →
• 14/819,478, filed Aug. 6, 2015, now US 9,489,746 — continuation of →
• 14/230,099, filed Mar. 31, 2014, now US 9,103,907 — continuation of →
• 12/710,057, filed Feb. 22, 2010, now US 8,717,545 — claims priority to →
• Provisional 61/154,207, filed Feb. 20, 2009.

A later continuation, 17/857,103, issued as US 12,025,441 B2 (2024-07-02); a further continuation 18/760,745 published as US 2025/0137781 A1.

Assignment/security history noted on the front page: 2021-12-15 reassignment to MVI (ABC), LLC; 2021-12-17 release by secured party Horizon Technology Finance Corporation to Stereovision Imaging, Inc.; 2021-12-17 assignment to Aeva, Inc. (from MVI (ABC), LLC). Legal status listed as Active.

2. Abstract (verbatim)

"A system uses range and Doppler velocity measurements from a lidar system and images from a video system to estimate a six degree-of-freedom trajectory of a target. The system estimates this trajectory in two stages: a first stage in which the range and Doppler measurements from the lidar system along with various feature measurements obtained from the images from the video system are used to estimate first stage motion aspects of the target (i.e., the trajectory of the target); and a second stage in which the images from the video system and the first stage motion aspects of the target are used to estimate second stage motion aspects of the target. Once the second stage motion aspects of the target are estimated, a three-dimensional image of the target may be generated."

3. Plain-language technical overview (grounded in the specification)

The patent addresses the problem that conventional lidar+video fusion required a pre-specified model of the target and assumed the target was essentially motionless. The disclosed approach instead solves for the target's own motion (6 degrees of freedom) without a target model, then uses that motion solution to "de-jitter" and register the data into a stabilized 3-D image.

Core mechanics described:

• Lidar subsystem: two, three, four, or five (or more) beams. Each beam yields, at each point, range + Doppler velocity simultaneously (dual-frequency chirped coherent laser radar, per the incorporated "Chirped Lidar Specification," U.S. App. Ser. No. 11/353,123). Points are parameterized by azimuth, elevation, range, Doppler velocity, and time.
• Video subsystem: a camera producing frames (2-D images) at a frame rate generally lower than the lidar point rate.
• Six-DOF decomposition: the lidar constrains ωx, ωy, and v_z^trans (rotational rates about x and y, and translational velocity along z) using spatially separated beam spots and their Doppler/range returns (see FIGS. 5–6). The video constrains v_x^trans, v_y^trans, and ωz via frame-to-frame feature tracking (see FIG. 7). This split of the six DOF between the two sensors is the conceptual heart of the disclosure.
• Two-stage estimation:
  • Stage 1 (FIGS. 11–13): iterative phases using least-squares estimators (1110 for lidar-derived ωx, ωy, v_z; 1120 for the full set) to produce transformation matrices T^(0), then T^(1), then T^(2).
  • Stage 2 (FIGS. 14–15): refinement using pixel-level differences between frames (not just extracted features) to produce T^(3), then a final refinement T^(4) using lidar measurements in overlap regions where the same point is revisited (including "overscan" beam revisits).
• Time stabilization: because scan points are acquired sequentially, measurements are adjusted to reference times (sub-point clouds → point clouds → video frame times) via transformation matrices describing the target's piece-wise motion.
• Application extensions described: physiological/behavioral monitoring (eyeball motion, facial features, neck/pulse/respiration, lip vibration for speech reproduction), with the "gross motion" removed to isolate "fine motion."

4. Independent claims — ⚠️ NOT VERIFIED

I do not have the granted claim text for US 11,378,396 and will not fabricate it. What can be said with confidence:

• The title is "System and method for generating motion-stabilized images…," and the specification's Summary uses parallel "In some implementations of the invention…" phrasing typical of a claim set that includes at least one system claim and one method claim (and typically a non-transitory computer-readable-medium claim in this family).
• Based on the Abstract/Summary and the sibling continuations, the independent claims are very likely directed to: (i) a lidar system providing range and Doppler velocity for points on the target; (ii) a video system providing frames; and (iii) a processor configured to estimate the 6-DOF trajectory in the described two stages, and to generate a motion-stabilized 3-D image.
• I could not confirm the exact wording, the number of independent claims, or the precise limitations (e.g., whether preambles recite "without a prior model," whether specific equations such as the v_z relation Eq. (2) are claimed, or how the two stages are positively recited).

To close this gap authoritatively, pull the claim set from: USPTO Patent Center / Patent Public Search (app. 16/586,933), the granted-PDF at https://patentimages.storage.googleapis.com/... linked from https://patents.google.com/patent/US11378396/en (the "Download PDF" link), or Espacenet/Global Dossier. I recommend treating any claim characterization not sourced from those documents as untrustworthy.

5. Litigation / CAFC 2026 docket search — no matching results

• Searching "11378396" combined with CAFC 2026 docket returned no appellate proceeding naming this patent. Results were unrelated (a UK company registration number 11378396; other patents).
• The Federal Circuit September 2026 oral-argument calendar and 2026 weekly summaries reviewed did not list US 11,378,396, Richard L. Sebastian, Anatoley T. Zheleznyak, Aeva, or Digital Signal Corporation as parties.
• Aeva's public filings reference securities litigation arising from its 2020–2021 SPAC (InterPrivate) transactions (complaints filed Dec. 23, 2020 and Jan. 20, 2021) — these are not patent suits involving this patent.
• No IPR/PGR petition for US 11,378,396 was surfaced. (Note: PTAB documents returned in search involved Google/GoPro and patent 10,958,840 — an unrelated patent.)

Uncertainty statement: Absence of a hit is not proof of absence. District-court and PTAB dockets for this patent may exist under a case caption not indexed by the queries run, and 2026 docket data may be incompletely indexed. A definitive check requires querying USPTO Patent Public Search (which links litigation/IPR "external links"), PACER/Docket Alarm, and USPTO PTAB search by patent number.

6. Bottom line

US 11,378,396 B2 is an Aeva (originally Digital Signal Corporation) continuation patent (app. 16/586,933, filed 2019-09-28, issued 2022-07-05) by Sebastian and Zheleznyak, claiming priority to Feb. 20, 2009 (prov. 61/154,207), with adjusted expiration 2030-08-11. It covers a model-free, two-stage six-degree-of-freedom trajectory solution fusing coherent-lidar range/Doppler (ωx, ωy, v_z) with video feature/pixel motion (v_x, v_y, ωz) to produce a motion-stabilized 3-D image. Its verified bibliographic and abstract data are solid; its claim scope is unverified here and should be read from the official grant document; and no CAFC 2026 or PTAB activity for this number was found.

I'll search for litigation involving US Patent 11378396 across patent litigation sources.

Let me dig deeper into litigation sources specific to this patent and its owner Aeva.

Let me search targeted litigation databases for this specific patent number.

I was unable to complete all the searches I wanted (I hit a tool-step limit), but here is what I found. I want to be precise about the distinction between genuinely responsive results and look-alike patent numbers.

Bottom line

I did not find any litigation specifically asserting U.S. Patent No. 11,378,396. Based on the sources I was able to query, no district court complaint, ITC action, CAFC appeal, or PTAB proceeding naming this patent was identified. I cannot rule out that litigation exists; I can only report that my searches did not surface it.

What the patent is (from the authoritative Google Patents text)

• Number: US 11,378,396 B2 — "System and method for generating motion-stabilized images of a target using lidar and video measurements"
• Granted: July 5, 2022 (application 16/586,933, filed Sept. 28, 2019)
• Inventors: Richard L. Sebastian; Anatoley T. Zheleznyak
• Original assignee: Aeva Inc; current assignee listed as Aeva Technologies Inc. The Google Patents assignment history also shows a 2021 chain involving StereoVision Imaging, Inc. and MVI (ABC), LLC before reassignment back to Aeva, Inc.
• Family: It is a continuation of 15/291,576 (now U.S. 10,429,507) → 14/819,478 (now U.S. 9,489,746) → 14/230,099 (now U.S. 9,103,907) → 12/710,057 (now U.S. 8,717,545), claiming priority to provisional 61/154,207 (filed Feb. 20, 2009). Later continuations include 17/857,103 (now U.S. 12,025,441) and 18/760,745 (US 2025/0137781 A1).

This lineage matters for litigation searching: the asserted patent in any suit could be a family member (e.g., the '545, '907, '746, or '507 patents) rather than the '396 patent itself, and none of those were confirmed as being in suit either on the evidence I retrieved.

Numbers I checked and rejected as NOT this patent

To apply the strict "do not auto-correct identifiers" rule, note that several search hits contained superficially similar identifiers that are different patents and must not be conflated with US 11,378,396:

• U.S. 11,794,396 — asserted by i4F in i4F v. Artisan (C.D. Cal., 2:26-cv-04985) over composite SPC flooring. This is a different patent (different number, different art).
• U.S. 11,633,396 — AstraZeneca v. Cipla (D.N.J., 3:24-cv-08167).
• U.S. 10,456,396 — Oyster Point Pharma v. Apotex (D.N.J., 2:23-cv-03860).
• A UK company registration number "11378396" (Argus for Security Limited) — not a patent.

None of these involve US 11,378,396 or its owner.

Sources checked

• Google Patents page for US 11,378,396 (patents.google.com/patent/US11378396/en) — no "Litigation" section/litigation entries in the fetched text.
• General web searches for the patent number plus "litigation," "lawsuit," "complaint."
• Searches combining the patent/its family and assignee "Aeva Technologies / Aeva, Inc."
• A search aimed at the Unified Patents litigation portal and PTAB-related results.

Caveats and suggested next steps

• I could not directly query the Unified Patents litigation portal, CourtListener/RECAP, PACER, the CAFC docket, or PTAB APIs within this session. Those are the places most likely to confirm or rule out a case, and I recommend verifying there.
• Aeva is an actively litigious-adjacent company in the lidar space (its SEC filings are available at investors.aeva.com and sec.gov), and its patents trace partly to StereoVision Imaging, so a suit naming this patent could plausibly exist but was not captured by the searches I ran.
• If you want a definitive answer, run the patent number and its family members through portal.unifiedpatents.com/litigation, CourtListener's RECAP search, and PTAB (for IPRs). I can do that if you let me continue tool use.

If it's helpful, I can re-run with more search steps and check the Unified Patents caselist, CourtListener, and PTAB directly for both US 11,378,396 and its family members.

I'll verify the structured data against live sources before writing anything.

Proceedings overview

There are zero AIA trial proceedings on file against US 11,378,396 — no IPR, PGR, or CBM petition has been instituted, denied, or terminated, and no Final Written Decision or Federal Circuit appeal exists for this patent. The structured USPTO ODP block ("no AIA trial proceedings") is the canonical answer, and my independent web searches ("11,378,396" IPR, Aeva IPR PTAB, and petitioner-side searches naming Aeva/StereoVision) surfaced nothing to contradict it — no petition document, no institution decision, no FWD, no CAFC opinion. Defensive posture: the patent is untested, not hardened. Unlike a patent that has survived IPRs, this one carries no § 315(e)(2) estoppel and no PTAB record. Every claim is live, and you would be the first petitioner — which cuts both ways (fresh art, no adverse estoppel, but no roadmap and no read on how the Board construes "six degree-of-freedom trajectory" / "motion stabilized").

Verification caveat (read this before relying on it): I could not query PTAB E2E/Case Tracking directly from this session, and ODP ingest lags filings. My negative finding rests on (a) the structured ODP block and (b) general web search that did not reach the PTAB E2E docket. Confirm independently before you tell a client "there's no IPR." Suggested queries: PTAB E2E Patent Trial search for patent number 11378396, plus owner-name searches for "Aeva, Inc.," "Aeva Technologies," and "StereoVision Imaging" to catch family-wide challenges.

Proceedings to list

None. There is no proceeding number, petitioner, panel, or disposition to report, and I will not invent one. For completeness, the fields you asked for would be empty as follows:

• Type: N/A — no petition filed
• Judge panel: none assigned
• Petition grounds: none — no § 102/§ 103/§ 112 ground has been presented to the Board
• Institution decision: none issued
• Final Written Decision: none — no claim of 11,378,396 has been canceled, and none has been sustained after trial
• Settlement / termination: N/A
• Appeal: none — no FWD exists to appeal
• Defensive value: The absence means no claim is dead and no claim is confirmed. You cannot point a court at an FWD canceling claim 1; conversely, you are not estopped and no adverse claim-construction record exists.

Strategic summary

Claim status: all claims UNTESTED. Because no AIA trial has reached an FWD, the entirety of the patent — every independent and dependent claim as issued 2022-07-05 — stands unadjudicated by the PTAB. There is nothing to "list as surviving claims," because nothing was canceled. Note the claim set you'd be attacking issued from application 16/586,933 (filed 2019-09-28), a continuation in a long chain back to provisional 61/154,207 (filed 2009-02-20), with intermediate patents 8,717,545 → 9,103,907 → 9,489,746 → 10,429,507. Later continuations are still issuing: 12,025,441 (from 17/857,103) granted 2024-07-02, and 18/760,745 (filed 2024-07-01) published as US2025/0137781. That ongoing prosecution output means a validity challenge aimed only at '396 may be met with a sibling patent asserting near-identical claims — plan for the family, not the single number.

Estoppel landscape: clean slate, and that is the key asset. With no instituted IPR and no FWD, no petitioner or privy is subject to § 315(e)(2) estoppel on this patent, and neither are you. Every prior-art ground — patents, printed publications, and the full § 103 combination space — remains available in an IPR, and all grounds (including system/device art, § 112 written-description/enablement, and § 101) remain available in district court. If you do file, the ordinary IPR limits apply: § 311(b) confines you to patents and printed publications under § 102/§ 103, and your § 315(b) one-year clock runs from service of a complaint asserting this specific patent, not from any earlier family member.

Pattern signals: none — because there is no pattern. No serial petitioner, no Unified Patents or other defensive aggregator in the chain, and no PTAB appeal activity by the patent owner. The title chain shows the asset passed hand-to-hand rather than being litigated: Digital Signal Corporation → Horizon Technology Finance (security interest, released) → DSC ABC, LLC → StereoVision Imaging, Inc. → MVI (ABC), LLC (recorded 2021-12-15, an ABC/assignment-for-benefit-of-creditors vehicle) → Aeva, Inc. (recorded 2021-12-17). That is a distressed-asset transfer into Aeva, not a monetization campaign — which is a plausible reason no competitor has bothered to petition. Google Patents lists the status as Active, adjusted expiration 2030-08-11, consistent with a 20-year term from the 2010-02-22 non-provisional filing plus a lengthy PTA tail. You therefore have roughly four more years of exposure if you are accused.

Recommended next steps

• Do not rely on an invalidated claim. No FWD exists, so there is no opinion to cite and no canceled claim to tell a plaintiff is dead. Any letter asserting that "claims were canceled in IPR" would be false; conversely, do not assume the patent is fragile merely because it is old — the priority date is 2009-02-20, which makes prior art harder to find and gives the owner a strong priority shield.
• Verify the docket yourself first. Run PTAB E2E Case Tracking on U.S. Patent No. 11,378,396 and on owner names "Aeva," "StereoVision Imaging," and "MVI (ABC)"; also docket-search CourtListener for any civil action (e.g., an Aeva v. competitor lidar suit) or CAFC appeal touching this patent. If a recently filed petition exists that ODP has not ingested, that changes everything above.
• Zero proceedings is itself a signal — but a weak one. Heavily asserted patents eventually draw IPRs; this family has been on the market since 2009 and holds FMCW/Doppler-lidar claims, yet no one has petitioned. Treat that as evidence of limited assertion activity, not as evidence of invalidity. Before committing to an IPR budget, do a focused prior-art search against the 2009-02-20 priority date, and expect the owner's strongest defense to be priority/§ 102(a)(2) provenance from the 2010 non-provisional.
• If you do file, expect no trial-stage milestones yet. With nothing pending, there is no institution-deadline, no oral-hearing date, and no statutory 1-year FWD due date to calendar. Those begin only if a petition is filed: institution decision within three months of the preliminary response, then FWD within 12 months of institution.

I'll research the ownership chain for US 11378396, starting with the legal events already in the patent text and then verifying via search.

Good — I now have the key events. Let me nail down the reel/frame details, correspondents, and the pre-SVI history (Digital Signal Corporation).

Assignment & Ownership Analysis — US 11,378,396 B2

Data provenance note: I could not query USPTO Assignment Center directly in this session (only a general web-search tool was available, and assignmentcenter.uspto.gov is not directly indexable). The reel/frame numbers and event dates below come from (a) the Google Patents legal-events record for US 11,378,396 (fetched 2026‑09‑24 and included in the source text), and (b) an indexed USPTO assignment-record excerpt for the same StereoVision Imaging → MVI (ABC) → Aeva transaction. Correspondent-of-record attorney names could not be retrieved and are reported as unknown — do not treat as absent. Verify at https://assignmentcenter.uspto.gov/ (search by patent number) or https://assignment.uspto.gov/patent/index.html.

Inventors

Pattern notes (with the caveat that no employment records were retrieved):

• This is a two-inventor core across the entire portfolio. That concentration is a portfolio-fragility tell, not an NPE tell.
• The relevant "departure" event here is corporate, not personal: the family's original corporate owner (DSC) went into insolvency in 2016, and the later operating owner (StereoVision Imaging) went into insolvency in 2021. I found no data on inventor departures within 12 months of any filing, so I expressly do not assert that signal.
• Notable timing quirk: StereoVision filed a still-pending continuation in this family (US 2021/0259544 A1, filed 2021‑08‑26) only ~26 days before assigning everything to an ABC fiduciary (2021‑09‑21).

Original assignee

Two different entities answer to "original assignee" depending on which record you read, and the distinction matters:

1. Entity named on the issued patent (US 11,378,396, granted 2022‑07‑05): Aeva, Inc. — because the patent issued after the December 2021 transfers, the assignee of record at print was Aeva, Inc. (Google Patents header: "Original Assignee: Aeva Inc / Current Assignee: Aeva Technologies Inc"). Aeva, Inc. is a subsidiary of Aeva Technologies, Inc. (NASDAQ: AEVA), an operating FMCW/4D LiDAR company. It ships products embodying the claims (Aeries II 4D LiDAR sensor; Aeva reported shipping to 25+ customers in Q3 2022 and "surpassed 130 granted patents"). Status: operating, public.
2. Entity that originated the 2009 priority family: Digital Signal Corporation (DSC) — a Delaware corporation that developed chirped coherent laser radar and long-range 3D facial-recognition systems. Primary business: 3D imaging / lidar + facial recognition for security and commercial use. Current status: dissolved via a Delaware assignment for the benefit of creditors ("DSC ABC"), reported in Horizon Technology Finance's 2017 annual report; DSC's IP was then sold.
3. Intermediate operating owner: StereoVision Imaging, Inc. (SVI) — a California corporation, "a market leader in 3D imaging and facial recognition technology for security and commercial applications" (Horizon Technology Finance press release, 2018‑04‑11). It bought DSC's assets (incl. IP) in February 2018 for a combined cash-and-equity price of ~$3.5M, and itself took a ~$3.2M Horizon term loan in Q1 2018. Current status: dissolved — general assignment for the benefit of creditors effective 2021‑09‑21.

Caveat: I could not confirm from the face of the 2009 provisional or the Assignment Center index whether DSC or a predecessor was formally the original assignee of record for the 2009 priority filing. Treat "DSC" as likely but unverified; the DSC 2016 ABC and the Feb‑2018 DSC→SVI sale are documented in Horizon's SEC filings.

Assignment timeline

Recorded assignments against this patent (conveyance data per USPTO assignment record; execution dates from the recording documents). The patent has no separately recorded assignment for the DSC era.

1. 2021‑09‑21 (executed) / recorded 2021‑12‑15 — Reel 058520/0078
   • Conveyance: Assignment (general assignment for the benefit of creditors)
   • Assignor: StereoVision Imaging, Inc. (California)
   • Assignee: MVI (ABC), LLC (single-purpose Delaware LLC; business address P.O. Box 439, Grandville, MI 49468)
   • Correspondent: not retrieved — required field for an NPE analysis; retrieve from Assignment Center.
   • Context: insolvency liquidation — SVI made a California general ABC of all assets, explicitly including "patents, copyrights, trademarks… all other intellectual property," with a claims bar date of 2022‑04‑01. The ABC notice states SVI "has no corporate affiliation" to MVI despite the name similarity.
2. 2021‑11‑23 (executed) / recorded 2021‑12‑17 — Reel 058533/0549
   • Conveyance: Assignment
   • Assignor: MVI (ABC), LLC
   • Assignee: Aeva, Inc. (California)
   • Correspondent: not retrieved.
   • Context: sale out of the ABC estate to an operating acquirer — Aeva purchased the SVI portfolio and immediately kept prosecuting it (continuations US 12,025,441 and US 2025/0137781 both claim priority to this family).
3. recorded 2021‑12‑17 — Reel 058533/0569
   • Conveyance: Release of Security Interest ("Release by Secured Party")
   • Assignor/releasing party: Horizon Technology Finance Corporation (NASDAQ: HRZN)
   • Assignee/beneficiary: StereoVision Imaging, Inc.
   • Correspondent: not retrieved.
   • Context: secured-lender lien release in connection with the wind-down; Horizon was SVI's venture-debt lender (term loans due 12/11/2019 and 12/11/2021 plus common stock) and participated in the DSC→SVI sale proceeds in 2018.

Reel/frame caveat: Reels 058520/0078, 058533/0549 and 058533/0569 are the recordings for the SVI→MVI→Aeva and Horizon-release transactions as they appear in indexed USPTO assignment data for the StereoVision/Aeva family. Because a single recorded assignment covers all conveyed properties, the same reel/frame normally applies to every SVI patent — but I could not open the Assignment Center abstract for US 11,378,396 itself to confirm the exact frame. Verify.

No infringement suit naming this patent was found in the records reviewed, so there is no litigation to date against the chain.

Timeline diagram

timeline
    title Ownership of US 11378396
    2009 : Priority application filed in DSC family
    2016 : Digital Signal Corp liquidated via ABC
    2018 : DSC IP sold to StereoVision Imaging
    2019 : Continuation application filed
    2021 : SVI assigns all assets to MVI ABC
         : MVI sells portfolio to Aeva Inc
         : Horizon releases security interest
    2022 : Patent issues to Aeva Inc
         : Aeva files further continuations

NPE / troll-pattern signals

1. Shell-entity transfer — Not present (as an NPE tell).
Formally, the patent did move from an operating company (SVI) to a single-purpose, non-operating Delaware LLC with a P.O. Box address and no products in commerce (Reel 058520/0078, executed 2021‑09‑21). However, MVI (ABC), LLC is not a licensing entity — it is a fiduciary assignee-for-the-benefit-of-creditors, and the ABC notice expressly disclaims any corporate affiliation between assignor and assignee. The transfer is insolvency mechanics, and the property was flipped to an operating acquirer within ~2 months (Reel 058533/0549, executed 2021‑11‑23). Weak/contextual, not a troll signal.

2. Known asserter in the chain — Not present.
The chain (Digital Signal Corporation → StereoVision Imaging → MVI (ABC), LLC → Aeva, Inc.) contains none of Acacia, Marathon, Intellectual Ventures, IPNav, Wi‑LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, or any Spangenberg entity. MVI (ABC), LLC is a professional ABC fiduciary, not a mass-filer plaintiff; no Unified Patents or RPX high-frequency-plaintiff listing was surfaced for it in this review.

3. Repeat correspondent across the chain — Unclear (insufficient data).
I could not retrieve the correspondent of record for any of Reels 058520/0078, 058533/0549, or 058533/0569. This is the single most probative field for the question asked and it is missing from my data set. It must be pulled from Assignment Center. I will not infer recurrence from firm names I never obtained.

4. Cascading transfers — Present.
Two consecutive assignments through a chained single-purpose LLC within 63 days of execution (2021‑09‑21 → 2021‑11‑23) and 2 days of recording (2021‑12‑15 → 2021‑12‑17), all sharing the same transaction package and (per the ABC notice) the same Grandville, MI fiduciary address. The speed and chaining pattern matches the signal literally — but the driver is the ABC, not assertion-readiness.

5. Pre-litigation transfer — Not present.
No infringement suit naming US 11,378,396 (or its family members) was found. The 2021 transfers are ~10 months and ~13 months before issue (2022‑07‑05) and are tied to the ABC, not to a filing deadline. No evidence the chain was arranged to enable assertion.

6. Bankruptcy fire-sale — Present.
This is the dominant, well-corroborated signal, and it occurs twice in the corporate history:

• Digital Signal Corporation made an assignment for the benefit of creditors (Delaware) in 2016 — "DSC (assignment for the benefit of creditors), LLC" — documented in Horizon Technology Finance Corporation's Form 10‑K (FY2017/2016).
• StereoVision Imaging, Inc. made a California general ABC to MVI (ABC), LLC effective 2021‑09‑21 (Reel 058520/0078), expressly conveying all patents and IP, with a bar date of 2022‑04‑01.
  Strictly, these are state-law ABCs rather than Chapter 7/11 filings, and the buyer was an operating company rather than a liquidation syndicate — but the substance (owner insolvency → patents sold out of an estate) is present.

7. Privateering — Not present.
Aeva, Inc. is the acquirer of the portfolio for product-development purposes (it kept prosecuting the family and touts the patents as supporting its Aeries II FMCW lidar line). There is no evidence SVI "lent" the patents to Aeva to assert on SVI's behalf, and SVI ceased to exist at the time of transfer.

8. Defensive aggregator — Not present.
The chain terminates at Aeva, Inc., an operating, publicly traded lidar manufacturer (NASDAQ: AEVA) — not at RPX, AST, LOT Network, Unified Patents, or OIN. The patent has not been neutralized.

Verdict

Operating-company assertion — applied on the ownership prong, with an explicit caveat on the assertion prong.

Justification: the chain ends at Aeva, Inc. (subsidiary of Aeva Technologies, Inc., NASDAQ: AEVA), an operating company that ships FMCW 4D LiDAR products and has continued prosecuting the family (priority to US 12,025,441 and US 2025/0137781), per Reel 058533/0549 (executed 2021‑11‑23, recorded 2021‑12‑17). The only "fire-sale" content in the record is insolvency-driven — DSC's 2016 ABC (Horizon 10‑K) and SVI's 2021 California ABC at Reel 058520/0078 (executed 2021‑09‑21) — and the buyer was an operating acquirer, not an asserter. No NPE signal is present; the one strong pattern (bankruptcy fire-sale) and the one weak pattern (cascading chained-LLC transfers) both trace to the ABC mechanics rather than to assertion financing. Caveat: I found no evidence Aeva is actually suing competitors on this patent, so if the "suing actual competitors" prong is treated as mandatory, the accurate characterization is "operating-company ownership, no assertion observed." The correspondent-of-record field (Signal 3) could not be retrieved and is the main open item.

Verify at: https://assignmentcenter.uspto.gov/ (search patent number 11378396); patent text at https://patents.google.com/patent/US11378396/en; SVI ABC creditor notice at http://media.hometeamsonline.com/photos/htosports/ROCKCREEKFA/MVI_ABC_Creditor_Notice.pdf; DSC→SVI sale and DSC ABC at Horizon's Q1‑2018 portfolio update http://ir.horizontechfinance.com/node/12256/pdf.

I'll research US patent 11378396 and its cited prior art. Let me search for authoritative sources.

Let me look for the specific citation tables for this patent.

I'm getting mixed results. Let me target the patent's own citation records specifically.

Important scope and confidence note

I attempted multiple targeted searches against the public patent databases. I was unable to retrieve the examiner's/“References Cited” (PTO‑892 style) citation table for US 11,378,396 B2 within the searches performed. Because the operating rules require that I not fabricate patent numbers, dates, or claim mappings, I will not invent a list of § 102 anticipatory references attributed to this patent. Below is what I can ground in the sources I did obtain, clearly separated from what I could not verify.

If you need the exact examiner-cited art, the authoritative source is the document itself: the front page of US 11,378,396 B2 lists the “U.S. PATENT DOCUMENTS” / “FOREIGN PATENT DOCUMENTS” under (56) References Cited, and the prosecution file wrapper is in USPTO PatentCenter for application 16/586,933. Those were not retrievable through my searches here.

1. The patent, interpreted literally

Continuity chain (from the specification's CROSS‑REFERENCE section): 16/586,933 is a continuation of 15/291,576 (now US 10,429,507) → continuation of 14/819,478 (now US 9,489,746) → continuation of 14/230,099 (now US 9,103,907) → continuation of 12/710,057 (now US 8,717,545) → which claims priority to provisional 61/154,207 filed 2009‑02‑20.

Critical-date consequence: Because the effective filing date is carried back to the 2009‑02‑20 provisional (assuming the chain holds), only references that qualify as prior art before that date (or § 102(b) statutory bars) can anticipate. Art published after 2009‑02‑20 is relevant only for § 103 or as family/background. This matters enormously for the § 102 analysis below.

2. Citations I can ground: references incorporated by the specification

The specification does not rely on external prior-art citations for its novelty argument; instead it incorporates by reference the following co-pending applications (these are the only “citations” I could verify from the text):

These are specification‑incorporated references, not examiner‑cited prior art, and none of them is identified by the applicant as anticipating the claims (they are meant to be read into the disclosure, not against it).

3. Family members (relevant as § 102/§ 103 “own work,” not third‑party art)

From the Espacenet “cited documents” listing and Google Patents family data, the closely related family includes:

• US 2010/0271615 A1 (2010‑10‑28) and US 8,717,545 B2 — “System and Method for Generating Three Dimensional Images Using Lidar and Video Measurements” (Sebastian; Zheleznyak)
• US 10,152,771 B1 (2018‑12‑11) — cites the above Sebastian/Zheleznyak disclosure (forward citation)
• US 11,675,085 — lists 11,378,396 (Sebastian, July 5, 2022) among its Referenced Cited U.S. patent documents (this is a forward citation of 11,378,396, not prior art against it)

Because these share the same inventors/priority, they are not prior art under § 102(a)/(b) against 11,378,396 to the extent the priority chain is valid; the provisional 61/154,207 (2009‑02‑20) is the operative date.

4. § 102 anticipation — what I can and cannot responsibly state

I could not verify any third‑party reference that the examiner cited against the claims of 11,378,396, and the full claim set was not present in the text I retrieved (the fetched copy ends mid‑description). Accordingly:

• I cannot name specific anticipatory references, their publication/filing dates, or map them to specific claims of 11,378,396 without fabricating that information.
• What I can say with confidence is the legal framework the § 102 analysis must follow:
  1. A reference anticipates only if it discloses every limitation of a given claim (e.g., the core combination of: a lidar generating simultaneous range and Doppler velocity at each of multiple points via two or more beams; a video camera producing frames; and a processor resolving a six‑DOF trajectory — 3 translational + 3 angular velocities — to generate a motion‑stabilized 3‑D image).
  2. Because the effective date is 2009‑02‑20, any asserted § 102 reference must be a patent/printed publication dated before that date (or a § 102(b) bar), or must fall under the exception rules — the recency of the Aeva-era lidar literature (mostly post‑2015) means most would be § 103 candidates, not clean § 102 anticipations.
  3. Claim 1 (and its independent counterparts, likely claim 11 in this family) is the only claim that can be anticipated; all dependent claims require the reference to also disclose their added limitations (e.g., four or more beams in a rectangular pattern; overlap regions; overscan beam; the two‑stage / multi‑phase least‑squares estimation using T₍i,i₊₁₎⁽⁰⁾…⁽⁴⁾ transformation matrices).

5. Recommendation to complete the request accurately

To produce the citation‑by‑citation § 102 table you want, retrieve these primary documents and I can analyze them:

1. US 11,378,396 B2 front page, item (56) — the examiner's “References Cited” list (US + foreign patent documents and NPL).
2. The file wrapper for application 16/586,933 (USPTO PatentCenter) — PTO‑892 notices, examiner's reasons for allowance, and any IDS submissions.
3. The parent US 8,717,545 B2 (56) list — the original examiner art, which usually carries forward.

Once you supply any of those (or allow me to retrieve them), I can give each reference a full citation, its publication/filing date, a brief description, and a precise mapping of which claim(s) it could anticipate under 35 U.S.C. § 102.

Bottom line: I confirmed the identity, family, priority date, and specification‑incorporated references of US 11,378,396, but I could not verify the examiner‑cited prior‑art list through search, and I decline to fabricate it. The controlling date for any § 102 analysis is 2009‑02‑20.

I'll research the prior art cited against US 11,378,396 before building the § 103 analysis.

Let me dig for the specific prior-art references cited in US 11,378,396.

I need the actual claim set of US 11,378,396 and the cited prior-art list. Let me pull those.

Scope note — what the page actually provides

Before the analysis, an important limit on the record I was given. The pasted page for US 11,378,396 does not contain a formal "References Cited / Prior Art" bibliography. What it contains under the prior-art heading is:

• Prior art keywords: target, implementations, lidar, measurements, beams (i.e., the broadest concepts, not a reference list)
• Prior art date: 2009-02-20
• Incorporated-by-reference prior documents named in the specification itself (these are the only concrete prior-art-like items on the page):
  • U.S. application Ser. No. 11/353,123, "Chirped Coherent Laser Radar System and Method" (the "Chirped Lidar Specification") — identified in the art as US 7,511,824 B2 / WO 2006/088822 A2, Digital Signal Corporation, filed 2006-02-14
  • U.S. application Ser. No. 11/230,546, "System and Method for Remotely Monitoring Physiological Functions"
  • U.S. application Ser. No. 11/610,867, "System and Method for Tracking Eyeball Motion"
• The priority chain to Provisional 61/154,207 (2009-02-20) and continuations US 8,717,545, US 9,103,907, US 9,489,746, US 10,429,507.

I could not retrieve the examiner-cited U.S./foreign patent list or NPL list for 11,378,396 from the page or from search, so I will not invent reference numbers I cannot verify. Where I rely on my own field knowledge rather than the page, I flag it.

One more caveat: I do not have the verbatim claim text of 11,378,396 from the page. I am reconstructing its likely scope from the shared specification and its sibling claims (US 8,717,545, 9,103,907, 9,489,746). Treat the claim language below as a paraphrase of the family, not a quote of '396.

1. The claim(s) under review

From the abstract and family, claim 1 of 11,378,396 almost certainly reads in substance:

A system comprising: (a) a lidar subsystem that directs at least two beams toward a target and, for each beam, produces a range measurement and a Doppler velocity measurement for each of a plurality of points; (b) a video subsystem capturing a plurality of two-dimensional images; and (c) a processor that (i) estimates vztrans, ωx and ωy for each point from the range and Doppler measurements (first stage); (ii) estimates vxtrans, vytrans and ωz from two-dimensional position and velocity measurements of at least one feature obtained from the images, and from the already-estimated vztrans, ωx, ωy; and (iii) generates a three-dimensional image from the six components.

The narrower siblings add: at least three or four beams displaced along two (orthogonal) axes; a fifth "overscan"/monitoring beam held on a located feature; and (in later continuations such as US 12,025,441) limitation of the target to a face and the feature to an eye, eyelid, lip, mouth corner, or nostril. Dependent claims of '396 likely track the specification's face/eye/neck/lip subject matter.

Effective filing date: 2009-02-20. Because 11,378,396 is a continuation in a chain claiming that date and the disclosure is coextensive, the claims are entitled to a pre-AIA § 102/§ 103 framework unless prosecution added claim text not supported by the 2010 parent (a fact issue I cannot resolve from this page). This matters a lot — see § 6 below.

2. What each on-page prior-art item teaches

Note the page's own admission in the Background: conventional systems merge lidar and video but "require some pre-specified initial model of the target" and "require that the target remain substantially motionless." That admission frames the problem but also concedes that combining lidar and video per se was known and that the remaining problem was motion — i.e., a known problem with a known type of solution, which is the classic KSR posture.

3. Combination A — the primary § 103 combination

R1 (chirped dual-frequency coherent lidar) in view of R4 (video feature tracking).

The claimed first stage is elementary rigid-body kinematics. For a rigid target, the measured line-of-sight (Doppler) velocity at a point is linear in the unknowns:

v_los = vz^trans + ωx·y − ωy·x

This is exactly the structure the specification itself recites (its Eq. (2): vz = vz^trans + [ω × (R − R_c − v^trans·Δt)]_z). Consequently:

• Two beams displaced along y → two equations → solve vztrans and ωx (patent's own FIG. 5).
• Two beams displaced along x → solve ωy (FIG. 6).
• Four beams in a rectangle → overdetermined system → least-squares improvement.

A POSITA who has R1 (a lidar that gives simultaneous range and Doppler at each of multiple beams) has every input needed for this algebra; nothing further is invented. Any college-level dynamics text supplies the equation.

The video half is equally routine. Measuring 2D position/velocity of tracked features across frames and least-squares fitting vxtrans, vytrans, ωz is standard optical-flow/pose estimation.

Motivation to combine (KSR factors, all present):

1. Complementary capabilities / predictable result. The lidar of R1 is fundamentally blind to motion lateral to the line of sight and to roll; video is exquisitely sensitive to exactly those. Combining them to obtain all six degrees of freedom is the obvious completion of either system, and the result (a full 6-DOF vector) is predictable, not unexpected.
2. Known problem, known solution type. The Background concedes the field was already merging lidar and video; the only stated deficiency was motion. Using the other sensor to cover the axes the first cannot see is the most predictable remedy.
3. Finite number of identified solutions. Given three unknowns from the lidar channel and three from video, there is essentially one sensible allocation.
4. Design incentive. Removing the "target must be motionless" constraint and avoiding any "pre-specified initial model" (both stated goals in the spec) is a direct, articulated design incentive.
5. Same field, overlapping assignee/technical lineage. R1–R3 are all Digital Signal Corporation/StereoVision documents — one company's own toolbox.

On this record I would expect a § 103 rejection of claim 1 (and of the two-beam/four-beam/orthogonality claims) to be prima facie sustainable, with the strongest sub-limitation being the coupling in claim 1's second step — i.e., feeding vztrans, ωx, ωy into the video-based estimate of vxtrans, vytrans, ωz — which is nonetheless the natural way to seed a least-squares fit (parameter initialization is a routine expedient).

4. Combination B — the facial-feature / physiological dependent claims

R1 + R4 + R2 (11/230,546) + R3 (11/610,867).

If any claim of '396 (or its siblings) requires the tracked feature to be a face, eye, eyelid, lip, mouth corner, or nostril, or requires monitoring a physiological function, then:

• R2 expressly teaches aiming the lidar at a person's surface features and isolating fine motion (pulse, respiration, vocal-cord vibration) by removing gross motion — precisely the gross/fine decomposition recited in the '396 specification.
• R3 expressly teaches the eye as the tracked feature.
• The motivation is explicit and strong: R2's own purpose is to monitor a subject who is not motionless, so one is directly led to a motion-compensation/trajectory estimator of the '396 type, and to use the already-tracked facial features as the video features.

I would rate these dependent claims as highly vulnerable to § 103, because R2 and R3 are the applicant's own incorporated-by-reference documents and are essentially blueprint-level teaching for the face/eye subject matter.

5. Combination C — the two-stage "refine by image subtraction" claims

If '396 carries the US 8,717,545-style refinement limitation (transform image Ii and Ij to a common time using the first-stage motion aspects, then subtract to obtain Δxij, Δyij, Δθzij, and iterate), that too is a standard technique: frame-to-frame image registration / image stabilization / motion-compensated differencing is ubiquitous (and, remotely, the concept of correcting point-cloud distortion via capture times appears in much later art such as US 2020/0041647 A1). The second-stage refinement is a routine iterative least-squares improvement on a first-stage estimate — a predictable, conventional engineering step (KSR: "obvious to try," mere automation of a known iterative technique).

Similarly, the FIG. 15 refinement using overlap regions (same point measured twice by different beams at different times) is a straightforward correspondence/closure-error minimization, and the "overscan" fifth beam is a design choice to obtain redundant measurements — a classic obviousness posture.

6. The strongest counter-arguments (where a § 103 case can be attacked)

I would be doing you a disservice not to flag these:

1. Pre-AIA § 103(c) common ownership. R1 (Ser. No. 11/353,123), R2 (Ser. No. 11/230,546) and R3 (Ser. No. 11/610,867) are all Digital Signal Corporation / StereoVision Imaging documents, and '396 is commonly assigned within that lineage. To the extent any of these qualifies as prior art only under pre-AIA § 102(e)/(f)/(g), it cannot be used in a § 103 rejection. Caveat for the attacker: the WO 2006/088822 A2 publication of R1 (published 2006) is a § 102(b) printed publication and is therefore outside the § 103(c) exception — so the R1 content is still fair game if the WO publication is used as the reference, even though the U.S. patent/application is not.
2. No express teaching to derive angular rates from Doppler. R1 teaches range and Doppler measurement; it does not, on its face, teach solving a multi-beam Doppler set for ωx and ωy. Expect the patentee to argue this is the inventive core. The rebuttal is that the linear equation v_los = vz + ωx·y − ωy·x is textbook kinematics and the specification presents it as "as would be appreciated" — an admission of routine skill.
3. The specific coupled two-step estimator (lidar-first, then video with lidar outputs as inputs) and the iterative re-estimation with interval subdivision (FIGS. 11–13) can be argued to be a specific arrangement, not merely two known systems side by side. The counter is that separating a six-parameter problem into sensor-aligned sub-problems, then iterating, is a routine numerical-methods choice.
4. Absence of a verified citation record. I cannot confirm what the examiner actually cited, whether a § 103 rejection was made and overcome, or what art was distinguished during prosecution of '396 or its 2019-filed parent — that page content was not provided. Any real opinion should be re-run against the file wrapper and the verified claim text.

7. Bottom line

On the materials available from the page — R1 (chirped dual-frequency coherent lidar giving simultaneous range + Doppler over multiple, steerable beams), plus routine video feature-tracking, plus R2/R3 for the facial/eye subject matter — claims directed to (a) estimating vztrans, ωx, ωy from lidar range/Doppler and (b) estimating vxtrans, vytrans, ωz from video features to (c) assemble a 6-DOF trajectory and a motion-stabilized 3D image are reasonably likely to be obvious under 35 U.S.C. § 103, because (i) the governing mathematics is elementary rigid-body kinematics that the specification itself dismisses as "appreciated," (ii) the two sensors are mutually complementary with an articulated incentive (remove the motionless-target and initial-model constraints), and (iii) the dependent-claim subject matter (face, eye, lip, physiological monitoring, gross/fine motion separation) is squarely disclosed in the applicant's own incorporated references R2 and R3.

The genuine battlegrounds are the pre-AIA § 103(c) common-ownership exclusion (with the WO publication of R1 as the workaround), the specific two-stage coupled estimator with iterative refinement, and the overlap/overscan correspondence technique — plus the fact that I could not verify the examiner's actual citation record or the verbatim '396 claim text from the page supplied.