US 11577776

I'll search for authoritative information on this specific patent number, including any 2026 CAFC docket activity.

Let me check the litigation docket referenced for this patent and any Federal Circuit 2026 activity.

Let me check for any Federal Circuit (CAFC) appeal activity in 2026 involving this case.

Search Summary: US Patent No. 11,577,776

Scope note on identifiers: I searched on the literal number 11577776 / 11,577,776. One search returned an unrelated hit (a chemical "CID: 11577776" compound registry entry in unrelated patent documents) — I excluded that as a non-match. Everything below pertains to the granted US patent.

Important caveat on CAFC: I did not find any Court of Appeals for the Federal Circuit (CAFC) docket referencing US 11,577,776 in 2026. The 2026 appellate activity surfaced in my searches is a Texas state-court case (Fifteenth Court of Appeals, No. 15‑25‑00001‑CV), which is not the Federal Circuit. The Ninth/regional patent litigation is in federal district court (S.D. Tex.), not yet at the CAFC. I cannot confirm CAFC activity from authoritative sources, so I flag that as unknown rather than asserting it.

Bibliographic Data (from the authoritative patent record)

Family: AU2021200976A1 (abandoned), EP3875349A1 (withdrawn), CN113353146A (active/pending), plus US continuations US 18/157,204 (issued as US 12,049,270 B2) and US 18/784,321 (published US 2024/0375708 A1).

Abstract (verbatim)

"Techniques are described for managing redundant steering system for a vehicle. A method includes sending a first control command that instructs a first motor coupled to a steering wheel in a steering system to steer a vehicle, receiving, after sending the first control command, a speed of the vehicle, a yaw rate of the vehicle, and a steering position of the steering wheel, determining, based at least on the speed and the yaw rate, an expected range of steering angles that describes values within which the first motor is expected to steer the vehicle based on the first control command, and upon determining that the steering position of a steering wheel is outside the expected range of steering angles, sending a second control command that instructs a second motor coupled to the steering wheel in the steering system to steer the vehicle."

Independent Claims — Plain-Language Overview

There are three independent claims: 1, 9, and 16.

Claim 1 (Method) — A four-step process:

1. Send a first control command to a first steering motor coupled to the steering wheel to steer the vehicle.
2. After sending that command, receive back (a) vehicle speed, (b) vehicle yaw rate, and (c) the steering wheel's steering position.
3. Determine an "expected range of steering angles" — the values within which the first motor is expected to have steered the vehicle given the first command — where that range is a function of at least the vehicle speed and the yaw rate.
4. If the actual steering position falls outside that expected range, send a second control command to a second (redundant) motor coupled to the same steering wheel to steer the vehicle.

Claim 9 (Apparatus) — A processor-based apparatus for controlling an autonomous vehicle implementing the same core steps, but with a narrowed computation: the expected range is determined as a function of the vehicle speed, the yaw rate, and a first pre-determined offset of the first motor's position within the steering system.

Claim 16 (Non-transitory computer-readable storage medium) — Code that, when executed, performs the same method as claim 1 (range as a function of at least speed and yaw rate), claimed in CRM form.

Selected Dependent Claims (context for scope)

• Claim 2 — first/second control commands are based in part on first/second pre-determined motor offsets; Claim 3 folds the first offset into the range computation.
• Claim 4 — torque-based safeguard: receive a measured torque from the first motor, determine an expected torque, and widen the expected range if measured torque is outside tolerance (this is the driver-hand-on-wheel false-positive mitigation).
• Claim 5 — if outside range, send a third command to deactivate the first motor (claimed as dependent in both method and apparatus contexts).
• Claims 7–8 / 15 — first motor in the steering column, second motor in the steering gear coupled to the steering shaft; the two motors at different locations.
• Claims 6, 13, 14, 19 — commands may be position control commands (angular displacement/position of the steering wheel) or torque control commands.
• Claim 11 — range may additionally account for a third pre-determined offset describing steering-wheel offset at the neutral position.
• Claim 20 — if the steering position is within range, continue operating the first motor.

Litigation Found (grounded)

• Federal district court (patent): CreateAI, Inc. v. Bot Auto, Inc. et al, No. 4:25-cv-03371, S.D. Tex., filed 2025-07-21. US 11,577,776 is one of the asserted patents (redundant steering). (Sources: RPX Empower litigation page; Ex Parte case analysis.) The accused products are autonomous semi-trucks; an identified point of contention is construction of "expected range of steering angles" and whether the accused products use two distinct steering motors.
• Related state-court matter: CreateAI Holdings, Inc. f/k/a TuSimple Holdings, Inc. v. Bot Auto TX Inc., Business Court No. 24-BC11A-0007 → Fifteenth Court of Appeals No. 15-25-00001-CV (a TUTSA trade-secret/injunction appeal; memorandum opinion filed 2026-09-15 affirming). This case discusses redundancy as a general AV concept but proceeds on trade-secret, not patent, theories.

Uncertainty Statement

• I have no authoritative evidence of a CAFC 2026 docket for this patent; the patent's active enforcement appears to be at the district-court stage as of my searches.
• Docket statuses change frequently; the litigation details above reflect the most recent searchable records and should be verified against the live PACER/CAFC dockets before being relied upon.
• All bibliographic and claim data above are taken from the authoritative full patent text supplied in the record rather than inferred.

I'll search for litigation involving US 11,577,776 across litigation-tracking sources.

Initial searches returned pharmaceutical chemical-ID noise. Let me target the litigation-specific sources directly.

Litigation Involving U.S. Patent No. 11,577,776

One litigation matter is known. Note that "11577776" is also a PubChem compound identifier (a thiazole-carboxamide compound) that dominates generic search results — I discarded those and confirmed the litigation on patent-specific sources. Similarly, there is an unrelated S.D. Tex. case numbered 4:21-cv-03371 (Dedmon v. Shell Exploration & Production Co., settled July 2025) — that is a different case number from the one at issue and is not related.

Case 1: CreateAI, Inc. (f/k/a TuSimple, Inc.) v. Bot Auto, Inc. et al.

Infringement theory for the '776 patent — as reflected in the complaint, plaintiff mapped each limitation of independent claim 1 to the accused trucks:

• sending a first control command to a first steering motor (Compl. ¶107);
• receiving vehicle speed, yaw rate, and steering wheel position after the command (¶108);
• determining an expected range of steering angles as a function of at least speed and yaw rate (¶109);
• sending a second command to a second (redundant) steering motor when the steering position falls outside the expected range (¶110).

Notably, the complaint quotes defendants' own promotion of vehicles with "full redundancy, from steering and braking systems to sensors and computing power" (Compl. ¶76). Identified points of contention include the construction of "expected range of steering angles" and the factual question whether the accused trucks actually contain two distinct steering motors.

Sources:

• https://portal.unifiedpatents.com/litigation/Texas%20Southern%20District%20Court/case/4%3A25-cv-03371 (litigation link listed on the Google Patents page for US 11,577,776)
• https://litigation.rpxcorp.com/litigation/txsdce-[2019667](/patent/2019667)-tusimple-v-bot-auto#overview (RPX Empower — "CreateAI, Inc. v. Bot Auto, Inc. et al," 4:25‑cv‑03371, S.D. Tex., filed 07/21/2025)
• https://ai-lab.exparte.com/case/dct/txsd/4:25-cv-03371/createai-inc-v-bot-auto-inc (Ex Parte AI Lab — status "Open"; cause "Patent Infringement")
• https://ai-lab.exparte.com/case/dct/txsd/4:25-cv-03371/doc/analysis/1 (complaint element‑by‑element mapping for the '776 patent)

Related family litigation (non‑patent, same parties)

The Google Patents record for US 11,577,776 flags "Family has litigation" and links a Darts‑IP "First worldwide family litigation" entry. That reference traces to a Texas state‑court trade‑secret action, which is related but is not an assertion of the '776 patent:

• CreateAI Holdings, Inc. f/k/a TuSimple Holdings, Inc. v. Bot Auto TX Inc.
  • Trial court: Texas Business Court, Harris County, 11th Division — No. 24‑BC11A‑0007 (Hon. Sofia Adrogué)
  • Appellate court: Texas Fifteenth Court of Appeals — No. 15‑25‑00001‑CV
  • Claim: Misappropriation of trade secrets under the Texas Uniform Trade Secrets Act (TUTSA), including alleged safety‑technology trade secrets describing a "dual 'nerve system' for auxiliary braking and the technical parameters of its steering system" (i.e., a redundant steering system).
  • Status/Outcome: Trial court denied CreateAI's temporary injunction and dissolved the TRO; the Fifteenth Court of Appeals affirmed on September 15, 2026, holding the trial court did not abuse its discretion (sensor array public/not secret; decision‑making technology materially different; safety‑technology claim overbroad and unsupported). A related suit was also filed against former CTO/CEO Xiaodi Hou and four other former executives.

Sources: https://search.txcourts.gov (15th COA Opinion, filed Sept. 15, 2026); https://www.vitallaw.com/news/... (IP Law Daily, Sept. 17, 2026); https://tcjl.com/15th-court-of-appeals-affirms-business-court-denial-of-ti-in-misappropriation-of-trade-secrets-case/

Caveats

• No other litigation asserting U.S. 11,577,776 was found — no additional district court suits, no ITC Section 337 investigation, and no CAFC appeal on the '776 patent itself. The CAFC (and Texas appellate) activity identified relates only to the state trade‑secret dispute.
• I could not retrieve the PACER docket directly; the case details above are drawn from litigation aggregators (Unified Patents, RPX Empower, Ex Parte AI Lab), the Google Patents litigation entries, and Texas appellate opinions. Docket‑level events (e.g., motion practice, stay, IPR petitions) may have occurred after the last indexed update. If you need the certified docket, I'd pull the S.D. Tex. 4:25‑cv‑03371 docket from PACER/PatentCenter directly.
• The '776 patent is part of a family that also includes continuation US 12,049,270 and pending US 2024/0375708, plus EP 3875349A1 (withdrawn), CN 113353146A (pending), and AU 2021200976A1 (abandoned). Litigation may exist against those family members that would not show up in a search limited to the '776 number.

I'll verify the canonical "no proceedings" baseline against live sources, since the structured ODP block could lag recently-filed petitions.

No PTAB hits yet. Let me check whether the accused infringer (Bot Auto) or a defensive aggregator has filed anything, and check the family members too.

Let me run two final confirmatory searches on the continuation patent and on any 2026 filings before the § 315(b) bar window.

PTAB / AIA Trial Proceedings — U.S. Patent No. 11,577,776

Proceedings overview

There are zero AIA trial proceedings on file against U.S. 11,577,776 — no IPR, no PGR, and no CBM. The structured "PTAB proceedings on file" block (USPTO Open Data Portal ingest) returns an empty set, and my independent searches for petitions targeting this patent (and its family members) by Bot Auto, Unified Patents, or any other party surfaced nothing. That means the breakdown of active / invalidated / sustained / settled / institution-denied proceedings is 0 / 0 / 0 / 0 / 0, and the bottom-line defensive posture for a defendant is the opposite of "the patent is hardened by PTAB wins": all of 11,577,776's claims are UNTESTED at the Board, and no statutory estoppel (§ 315(e)(2)) attaches to any petitioner because no trial was ever instituted. The patent carries a live district-court assertion (S.D. Tex. 4:25‑cv‑03371) with no administrative validity challenge in parallel — an unusual and strategically significant posture for a 2023-granted patent in active litigation.

No proceedings to report — the empty set, and what sits around it

Because the requested per-proceeding format presumes at least one filing, I am reporting the null result explicitly rather than inventing docket numbers. For completeness, here is the adjacent activity that a defendant will confuse for PTAB activity (it is not), with the reason it is not:

1. District court (patent) — not a PTAB proceeding

• CreateAI, Inc. (f/k/a TuSimple, Inc.) v. Bot Auto, Inc. et al, No. 4:25‑cv‑03371, S.D. Tex., filed 2025-07-21, Judge Charles Eskridge. U.S. 11,577,776 is asserted (redundant steering), alongside U.S. 12,190,609 and the "'783 Patent." No IPR was filed by these defendants on the '776.
• Docket milestones retrieved (last retrieved 2026-01-16; verify on PACER): Motion to Dismiss filed 2025-09-22 (Dkt. 11); opposition 2025-10-06 (Dkt. 19); reply 2025-10-16 (Dkt. 20); Motion to Stay Pending Ruling on Motion to Dismiss filed 2026-01-16 (Dkt. 23); Initial Pretrial Conference set 2026-02-18.
• The court has not stayed the case pending any AIA trial — there is none to stay for.
• Source: https://dockets.justia.com/docket/texas/txsdce/4:2025cv03371/[2019667](/patent/2019667)

2. Texas state trade-secret appeal — not a PTAB proceeding and not a patent-validity ruling

• CreateAI Holdings, Inc. f/k/a TuSimple Holdings, Inc. v. Bot Auto TX Inc., Fifteenth Court of Appeals No. 15‑25‑00001‑CV (trial court: Texas Business Court, Harris County, No. 24‑BC11A‑0007, Judge Sofia Adrogué). Affirmed 2026-09-15 (memorandum opinion, Chief Justice Brister). This is TUTSA, not patent law; the appellate court discussed redundant steering/safety technology only as an alleged trade secret and held CreateAI's safety-technology claim failed "both by overbreadth and by the details in evidence."
• Link: https://www.courtlistener.com/opinion/[10586473](/patent/10586473)/createai-holdings-inc-fka-tusimple-holdings-inc-v-bot-auto-tx-inc/
• Why this matters to a PTAB analysis: it creates a public record of prior-art-adjacent admissions about redundancy being a "central dogma" of every AV company, and about the accused trucks using a different brand of truck and different steering component with different interfacing protocols. That record is the kind of material an IPR petitioner would mine — which highlights how remarkable it is that no IPR was filed.

Strategic summary

Claim status of 11,577,776: every claim is UNTESTED and INTACT at the Board. Independent claims 1, 9, and 16, and dependents 2–8 and 10–15 and 17–20, have never been before an APJ panel. There is no Final Written Decision, no certificate of cancellation, and no disclaimer of record. The only post-grant modification to the document itself is an unrelated certificate of correction dated 2023-04-11 (a clerical fix, not a claim change). Practically: an infringement defendant cannot point to any canceled claim, cannot invoke issue preclusion, and cannot tell a counterparty "claim 1 is dead" — because it is not.

Estoppel landscape. Because no IPR/PGR was ever instituted against this patent, 35 U.S.C. § 315(e)(2) estoppel does not exist for anyone. There is no petitioner estopped, and equally, no prior-art ground has been "burned" by exposure to the Board. For a new defendant evaluating an IPR today, the entire prior-art universe remains available on paper — but the procedural gates may be closed:

• § 315(b) one-year bar. The '776 patent was asserted in S.D. Tex. on 2025-07-21. Section 315(b) runs one year from service of the complaint, not filing. If Bot Auto (or any co-defendant) was served in July/August 2025, the bar lapsed around July–August 2026 — i.e., before today's date of 2026-09-29. I do not have the certified service date from PACER, so I flag this as the single most important fact to verify: if service predates 2026-09-29 by more than a year, an IPR by that defendant is time-barred; the only remaining PTAB route would be joinder under § 315(c) to an existing instituted IPR — and there is no such IPR to join. A defendant not yet served, or sued on a different patent in the same family, faces no such bar.
• § 325(d) / § 315(e)(1): not triggered, since nothing was before the Board.

Pattern signals. No pattern exists to read: the same petitioner has filed no IPRs on this patent; the patent owner (CreateAI/TuSimple) has pursued no PTAB appeals on the '776 because it has never needed to defend one; and there is no defensive aggregator in the chain — the "Unified Patents" link on the Google Patents page is a litigation-tracking page, not a Unified Patents IPR filing, and I found no Unified Patents petition on this patent or its family. The family itself is a mixed bag that a challenger should map: US 12,049,270 (continuation, granted 2024-07-30), US 2024/0375708 (pending continuation), EP 3875349A1 (withdrawn), CN 113353146A (pending), AU 2021200976A1 (abandoned) — an IPR aimed only at the '776 would leave the '270 continuation standing, which is why a family-level validity strategy (and a family-level § 315(b) assessment) matters here.

One contextual note on timing. Had a petition been filed, it would have landed under the PTAB's March 26, 2025 "Interim Processes for PTAB Workload Management" regime, which expanded discretionary-denial briefing under Fintiv, General Plastic, and Advanced Bionics-type considerations. The parallel district court case, now past a motion to dismiss and into scheduling, would have been a live Fintiv-type denial risk. That dynamic — combined with the expired § 315(b) window — plausibly explains the absence of any petition better than any assumption that the art is strong.

Recommended next steps

If you are a defendant being asserted on 11,577,776:

• Do not assume the absence of PTAB activity means the claims are strong. It means they are simply unadjudicated. There is no FWD to cite, no cancellation to leverage, and no estoppel to invoke. Your validity attack lives in district court under §§ 102/103/112 unless an IPR remains procedurally available to you.
• First action item: pull the certified docket and the return of service for S.D. Tex. 4:25‑cv‑03371 from PACER/PatentCenter to fix your § 315(b) date. If you were served more than one year ago, an IPR is barred; if service was recent (or you have not been served), a petition targeting claims 1, 9, and 16 — with dependents — is still open, and the § 325(d) risk is low because the examiner's cited art was limited (see the prosecution citations below).
• Examiner-cited art to differentiate from and to avoid § 325(d) overlap: US 2012/0055730 A1 (Nippon Soken), US 2012/0303218 A1 (Honda), WO 2017/028788 A1 (BYD), US 2017/0088167 A1 (Toyota), DE 10 2015 116937 A1 (Bosch), US 2018/0029637 A1 (Ford), US 2018/0304918 A1 (Toyota), KR 101878939 B1 (Mando), and US 2022/0089211 A1 (Ford, later art). The Board's § 325(d) practice — and the recently designated informative decisions on material examiner error — makes it essential to show either art the examiner never saw or a demonstrable error in how the examiner treated these references.
• Interim PTAB posture (if you do file): under the March 26, 2025 discretionary-denial regime and Ford Motor Co. v. AutoConnect Holdings LLC (IPR2025‑01342, Paper 27, May 12, 2026, informative), a parallel district court case past a motion to dismiss invites a Fintiv-type denial. Budget for a Sotera-type stipulation and discretionary-denial briefing from the outset.
• Cross-family check: before investing, verify whether the same prior art also reads on continuation US 12,049,270 and pending US 2024/0375708. Invalidating only the '776 leaves the family's other claims in play.

If no PTAB activity exists and you are advising on risk: say it plainly — as of 2026-09-29, the '776 patent has never been challenged at the PTAB, by anyone, on any ground. That absence is a signal that the patent's assertion is recent and that the § 315(b) window has likely closed for the current defendant — not that the patent has proven itself.

Verification and caveats

• Canonical source: the structured "PTAB proceedings on file" block (USPTO ODP) — empty set. I searched to confirm and found no contradicting filing.
• PTAB E2E / Decisions portal (verify directly): https://ptacts.uspto.gov/ptacts/ (search Patent No. 11577776) and https://www.uspto.gov/patents/ptab/decisions
• No FWD exists, so no panel, grounds, institution reasoning, verdict-at-claim-level, or settlement terms can be reported — I have not manufactured any.
• No Federal Circuit appeal of a PTAB decision exists for this patent; the CAFC is not in this chain. The 2026 appellate activity in the neighborhood is the Texas Fifteenth Court of Appeals opinion (15‑25‑00001‑CV, 2026-09-15), which is not a patent-validity ruling and has no estoppel effect on the '776.
• The '776 Google Patents page flags "Family has litigation" and links the Darts‑IP "first worldwide family litigation" entry — that traces to the same Texas state trade-secret dispute, not to an AIA trial and not to a PTAB proceeding.
• Docket and Board statuses change daily. The S.D. Tex. service date, the outcome of the pending Motion to Stay (Dkt. 23), and any petition filed after the most recent ODP ingest should be re-verified before being relied on. If a petition has been filed in the weeks since the last indexed update, the § 315(b) analysis above flips its conclusion.

I'll build the assignment reconstruction. Let me pull authoritative records on the assignment chain, correspondents, and the assignee's corporate history.

The reel/frame lookups returned noise. Let me target the assignment records and corporate/IP-portfolio details more directly.

Assignment / Ownership Chain — US 11,577,776 B2

Sourcing note (important): The two reel/frame entries below are taken from the USPTO-derived legal-events record supplied in the authoritative patent text and are consistent with the USPTO Assignment Center data model. I was not able to open the Assignment Center record itself in this session — targeted queries on the literal reel/frame strings (052033/0384, 073832/0553) returned unrelated document noise, and the Assignment Center pages are not directly indexable by the search tooling available here. The "Correspondent" field for both records is therefore UNRETRIEVED, not "none." I flag this rather than guess an attorney name. Verify at the Assignment Center search page linked in the Verdict section.

Inventors

Pattern assessment: No unusual pattern. All three inventors assigned to the original assignee before the 2020-03-05 filing date as part of ordinary employment onboarding — this is a clean, single-step, pre-filing inventor-to-company assignment, which is the opposite of the "inventors depart and the portfolio is fire-sold" pattern. I found no evidence that any of the three departed TuSimple within 12 months of filing, and I found no later assignment involving any inventor as assignor (no inventor-to-shell follow-on conveyance). I also found no evidence of inventor-initiated reversion, which would have produced a third reel/frame. Note: inventor current employment is not determinable from the assignment record; absence of a record is not proof of continued employment.

Original assignee

TuSimple, Inc. (California corporation; recorded address 3636 Nobel Drive, Suite 350, San Diego, CA 92122).

• Line of business: Autonomous trucking — full-stack L4 autonomous driving for Class 8 semi-trucks. It was the first autonomous-driving startup to IPO on Nasdaq (April 2021, ~$1.35B raised).
• Did it ship a product embodying the claims? It operated an actual autonomous truck fleet and ran public "Driver Out" driverless runs; its trucks used an electric steering actuator, and its own investor materials (March 2023 and Sept. 2023 presentations) tout a portfolio of ~500–589 worldwide patents/applications as core asset value. However, I could not verify that any TuSimple-shipped vehicle implemented the specific two-motor redundant steering architecture recited in claim 1 (steering-column motor + steering-gear motor). That is precisely the disputed factual issue in the co-pending litigation (see Verdict justification). Treat "shipped an embodiment" as plausible but unverified for the specific claim scope.
• Current status: Operating, but pivoted out of the AV product market — not dissolved, not in bankruptcy. Sequence: shut down U.S. self-driving truck operations and delisted from Nasdaq (Jan/Feb 2024); Delaware certificate of amendment changing TuSimple Holdings, Inc. → CreateAI Holdings, Inc. approved 2024-12-11; public rebrand announced 2024-12-19; OTC ticker/name change effective 2025-04-24 (OTC: TSPH); the company announced a plan to monetize its existing autonomous-driving IP. It also announced "dissolution and liquidation" of its Guangzhou subsidiary and cut ~100 staff there — a subsidiary wind-down, not a Chapter 7/11 of the patent-owning entity, and I found no patent sale in any insolvency proceeding.

Assignment timeline

Two recorded conveyances. Both are on the record as pending-status-Active.

• 2020-02-20 / 2020-02-25 (executed; signing dates run across this window) / recorded 2020-03-05 — Reel 052033 / 0384

  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)"
  • Assignor: Kaixin Zheng; Xiaoling Han; Zehua Huang (all three, jointly)
  • Assignee: TuSimple, Inc. (California)
  • Correspondent: Not retrieved (see sourcing note). No recurrence can be assessed for this link.
  • Context: Ordinary-course inventor-to-employer assignment, executed pre-filing and recorded on the filing date — this establishes TuSimple as the original owner and gives the chain a clean first link.

• 2025-07-24 (effective) / recorded 2025-12-03 — Reel 073832 / 0553

  • Conveyance: Change of Name — "CHANGE OF NAME; ASSIGNOR: TUSIMPLE, INC."
  • Assignor: TuSimple, Inc.
  • Assignee: CreateAI, Inc. (California)
  • Correspondent: Not retrieved (see sourcing note).
  • Context: Change of name only — no consideration, no transfer of ownership. This is not an asset sale, not a fire-sale, and not a transfer to an asserter; the same legal person owns the patent before and after.

Correspondence / mismatch flags I can support with the record:

1. Two different effective dates for "the name change" exist, because there are two different legal persons. The parent (Delaware) changed name 2024-12-11 (per the Delaware certificate of amendment lodged in the Texas appeal). The patent-owning subsidiary (California) changed name to CreateAI, Inc. with a 2025-07-24 effective date in the assignment record. The Google Patents current-assignee field names the subsidiary ("CreateAI Inc"); public filings and the state-court appeal name the parent ("CreateAI Holdings Inc"). Anyone checking standing should confirm which entity the federal complaint names.
2. Recording lag: the change of name was recorded 2025-12-03, ~4.4 months after its 2025-07-24 effective date and ~4.5 months after the 2025-07-21 complaint was filed. Recordation is a perfection/notice step, not the transfer date, so this lag is benign — but it does mean the patent record did not reflect the new name when the suit was filed.
3. No separate assignment is recorded for the family members US 12,049,270 B2 (continuation) or US 2024/0375708 A1. Ownership of those flows from the same original assignment; they would each carry their own change-of-name record. I did not retrieve those records.

Timeline diagram

timeline
    title Ownership of US 11577776
    2020 : Inventors assign to TuSimple Inc
         : Application filed 2020-03-05
    2023 : Patent issued 2023-02-14
    2024 : TuSimple exits US trucking
         : Delaware parent renamed CreateAI
    2025 : Patent owner renamed CreateAI Inc
         : Change of name recorded 2025-12-03

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT.
No link in the chain moves the patent from an operating company to a licensing-only LLC. The only post-filing conveyance is Reel 073832/0553, a change of name where assignor and assignee are the same corporate person. No "IP / Licensing / Holdings / Ventures" LLC appears as an assignee of the patent. (The parent's "Holdings" suffix is a corporate-form artifact of a public reporting company, not a licensing shell, and it is not the patent assignee.)

2. Known asserter in the chain — NOT PRESENT.
Neither assignee (TuSimple, Inc.; CreateAI, Inc.) appears on the Acacia / Marathon / IV / IPNav / Wi-LAN / Conversant / Vringo / Pendrell / Innovatio / MPHJ / Lumen View / Round Rock / Spangenberg lists, nor is either surfaced as a high-frequency plaintiff by RPX or Unified Patents. CreateAI is asserting its own organically developed patent, not an acquired one.

3. Repeat correspondent across the chain — UNCLEAR / NOT ESTABLISHED.
This is the one signal I cannot close out, and I will not infer it. The Correspondent field for both Reel 052033/0384 and Reel 073832/0553 was not retrievable in this session. There are therefore zero observed correspondent data points — not one, and not a recurring one. I did not treat the prosecution attorney on an unrelated TuSimple patent (e.g., the "Paul Liu" firm signature appearing on US 10,953,880) as an assignment correspondent; prosecution counsel of record and the recording correspondent are different fields, and conflating them would be exactly the kind of inference the brief warns against. Action item: pull the correspondent from both reel/frame summary pages to close this signal. With only two links — and one of them a pure name change — the recurrence threshold (2+ links sharing a correspondent) is at its theoretical minimum anyway.

4. Cascading transfers — NOT PRESENT.
Two recorded events span 5 years 9 months (2020-03-05 → 2025-12-03). There is no chain of consecutive LLC-to-LLC assignments; there is no <24-month cascade; there are no shared correspondence addresses to compare.

5. Pre-litigation transfer — NOT PRESENT.
The only post-issuance record (effective 2025-07-24, recorded 2025-12-03) is a name change, not a transfer, and its effective date falls three days after the 2025-07-21 complaint — i.e., it could not have been arranged to enable assertion (the suit preceded it). There is no assignment dated within 6 months before the first suit.

6. Bankruptcy fire-sale — NOT PRESENT.
I found no Chapter 7 or Chapter 11 filing by TuSimple, Inc. or CreateAI Holdings, Inc., and no §363 sale or patent auction. Note carefully: the "dissolution and liquidation process" announced for the Guangzhou subsidiary is a subsidiary-level wind-down; it is not an insolvency of the patent owner and it produced no recorded patent assignment.

7. Privateering — NOT PRESENT.
The patent was not transferred to a third-party NPE to assert on the operating company's behalf. The original owner retained and is asserting the patent itself. No SEC 8-K/10-K disclosure of a patent transfer to an assertion vehicle was found.

8. Defensive aggregator — NOT PRESENT.
The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. No neutralizing acquisition has occurred.

Additional observation (not one of the eight, but material): the current assignee has exited the market in which the claimed technology is practiced. CreateAI's stated plan expressly includes monetizing its autonomous-driving IP, and its sole AV enforcement action targets a competitor (Bot Auto) associated with its former co-founder. This is economically adjacent to NPE conduct but is categorically different: the patent was never acquired; it is being asserted by the original organically-developing owner.

Verdict

Operating-company assertion.

Justification: The entire ownership chain is two links — the original pre-filing inventor assignment to TuSimple, Inc. (Reel 052033/0384, executed 2020-02-20/25, recorded 2020-03-05) and a pure change of name to CreateAI, Inc. (Reel 073832/0553, effective 2025-07-24, recorded 2025-12-03). No shell LLC, no listed NPE, no cascading LLC transfers, no insolvency sale, and no defensive-aggregator termination appear anywhere in the record; the asserter is the original organically-developing owner suing a direct competitor in CreateAI, Inc. v. Bot Auto, Inc. et al, No. 4:25-cv-03371 (S.D. Tex., filed 2025-07-21). Caveat that keeps this from being a clean call: the current assignee has exited autonomous trucking and no longer ships a product embodying the claims (its AV monetization path is licensing plus this litigation), so the "ships products embodying the claims" prong of the operating-company test is now weak — the fact that the patent was never purchased and the plaintiff is the original inventor-company is what drives the verdict. The correspondent field for both reel/frame entries remains unretrieved and is the one open verification item; a repeat-player recording attorney would be the only fact capable of changing this verdict, and with two links (one being a name change) a recurrence finding is unlikely.

Verify at: https://assignmentcenter.uspto.gov/ (search by patent number 11577776) — and the mirrored legacy interface at https://assignment.uspto.gov/patent/index.html. Confirm the correspondent on Reel 052033/0384 and Reel 073832/0553, and, for completeness, run the same lookups on the family members US 12,049,270 and US 2024/0375708.

Cross-reference to the previously generated sections: no contradiction found — the ownership facts here (TuSimple, Inc. of California as original assignee; CreateAI as successor by name change effective 2025-07-24) are consistent with the bibliographic and litigation summaries already produced. One point worth adding to those sections: the federal plaintiff is styled "CreateAI, Inc." in the patent record, whereas the Delaware parent that renamed itself on 2024-12-11 is "CreateAI Holdings, Inc." — a distinction worth confirming against the caption of the 4:25-cv-03371 complaint, since it bears on standing to assert the '776 patent.

I'll research the specific prior-art references cited on the face of US 11,577,776. Let me pull details on each.

Let me gather the remaining cited references.

Prior-Art Analysis — US 11,577,776 B2

Scope and method note

I ran targeted searches on the literal number 11577776 (not similar numbers) and on each reference listed in the "Patent Citations" table of the authoritative full text you supplied. Generic searches for "11577776" are polluted by a PubChem compound ID and by an unrelated S.D. Tex. case number (4:21-cv-03371, Dedmon), which I excluded — consistent with the prior litigation summary.

Two housekeeping flags before the substance:

1. Date inconsistency. This task header says "Current Date: April 26, 2026," while the fetched patent record and the prior sections reference events dated September 2026 (the Texas 15th COA opinion of 2026-09-15). The system date given to me is 2026-09-29. I treat the later date as operative for "as-of" purposes and flag the earlier date as contradictory. Nothing in this prior-art analysis turns on the difference.
2. Prefer-the-record rule applied. I used the patent's own citation list as the authoritative universe of prior art and treated the search results as corroboration, not replacement. Where a reference's claims are quoted, they come from the record.

I also confirmed that no anticipation case (§ 102) exists among the cited references — which is consistent with the claims having been allowed. This is a § 103 / background citation set, not a § 102 set. I explain that per reference below rather than manufacturing anticipation theories.

The citation set (verbatim from the record)

The '776 patent lists 10 patent citations (a mix of examiner-cited [*] and third-party), 2 "Family Cites Families" items, and 1 non-patent citation.

Key statutory observation: all genuine prior art must predate 2020-03-05. Items #9, #10, and US 2025/0333104 post-date that date and therefore cannot be § 102(a)(1)/(a)(2) art for the '776 patent; their presence in the table reflects family/prosecution cross-citations, not substantive art. I say so explicitly rather than forcing them into a § 102 slot.

Per-reference detail

1. US 2012/0055730 A1 — Nippon Soken, "Steering control apparatus"

• Dates: published 2012-03-08; priority/earliest date 2010-08-19. Well before 2020-03-05. ✔ prior art.
• Description: A variable-gear-ratio steering (VGRS) system with an input shaft, output shaft, a variable gear ratio device having a gear mechanism and a first motor, plus a power-steering device with a second motor. Two "drive control parts": the first controls the first motor based on steering-wheel angle and a speed-increase ratio; the second "controls drive of the second motor based on the steering wheel angle and the speed increase ratio when abnormality is detected in the variable gear ratio device."
• § 102 assessment: The single most semantically adjacent element is "a second motor is driven upon detection of an abnormality" — i.e., a redundancy/fallback concept, and it is the closest thing in the set to claim 1's final step. But it is missing every material element of claims 1/9/16: it computes no "expected range of steering angles," uses no yaw rate, and performs no comparison of an actual steering position against such a range. It also controls a variable-ratio gear device rather than two steering motors both "coupled to the steering wheel" for steering. → Cannot anticipate claims 1, 9, or 16, or any dependent claim. Best characterized as § 103 background supporting the general "redundant/backup motor" concept. Relevant claims if combined: 1, 5, 7–8.

2. US 2012/0303218 A1 — Honda, "Electric steering device"

• Dates: published 2012-11-29; earliest date 2010-11-29. ✔ prior art.
• Description: An electric steering device with a steering wheel carrying one or more driver-operated units, a turning motor, and a control unit that invalidates a driver-operation signal when the steering angle to the right or left exceeds a threshold; the threshold is made variable with vehicle speed. (It uses a steering-angle-vs-threshold logic and a vehicle-speed dependence.)
• § 102 assessment: It touches two '776 themes tangentially — (a) treating the steering angle as a monitored quantity and (b) making a speed-dependent threshold. But it has no second motor, no yaw rate, and no expected steering-angle range. → No anticipation of any claim. Of marginal § 103 interest only.

3. WO 2017/028788 A1 — BYD, "Auxiliary steering system and auxiliary steering method for electric vehicle and electric vehicle"

• Dates: published 2017-02-23 (PCT/CN2016/095706, filed 2016-08-17); family priority 2014-11-17 (CN 201410653379.6, per the BR family member). Either way, ✔ prior art.
• Description: A detection controller + aligning motor detects whether a steering assisting device is normal; if it is abnormal, it controls the aligning motor to drive the wheels to align and issues a steering-malfunction signal, steering-wheel torque signal, and direction signal; a vehicle controller then drives the motors to increase drive torque of the outer front wheel, brake the inner rear wheel, and stop driving the inner front/outer rear wheels (torque-vectoring fallback steering).
• § 102 assessment: This is a genuine fault-detection-then-fallback steering scheme and is among the closer references. However, the fallback is differential wheel torque / drive-torque vectoring, not a second motor coupled to the steering wheel, and it computes no expected range of steering angles from speed and yaw rate. → No anticipation of claims 1/9/16. Its "detect steering-assist malfunction → invoke alternate steering" teaching is § 103 material against the general redundancy step, and its torque signal output touches claim 4/6-type concepts. Relevant claims if combined: 1, 4, 5, 6.

4. US 2017/0088167 A1 — Toyota, "Automated driving apparatus and automated driving system" (granted as US 10,167,013)

• Dates: published 2017-03-30; earliest date 2015-09-28. ✔ prior art.
• Description: Executes automated driving along a first target steering angle while controlling rotation of the steering wheel to a corresponding angle; when the driver applies a steering operation, the vehicle temporarily follows the actual steering angle; a second target steering angle asymptotically approaches the actual angle (and later asymptotically returns). Includes a steering actuator controlling the steering angle and a steering-wheel actuator controlling torque applied to the steering wheel — i.e., two distinct actuators associated with the steering wheel.
• § 102 assessment: Notably, this reference has two actuators (one for steer angle, one for steering-wheel torque) and detects driver steering intervention — relevant to the '776 driver-hands-on-wheel false-positive theme (claims 4, and the torque-compensation discussion). But it discloses no expected steering-angle range computed from speed and yaw rate, and the two actuators are not redundant steering motors switched on fault. → No anticipation of any claim. Strong § 103 candidate alongside the Bosch/BYD references. Relevant claims if combined: 1, 4, 6, 9.

5. DE 10 2015 116 937 A1 — Robert Bosch Automotive Steering GmbH, "Method for operating a steering system of a motor vehicle"

• Dates: published 2017-04-06; earliest date 2015-10-06. ✔ prior art. (Same family as US 2018/0086379 A1.)
• Description: A partially redundant steering system with two (or more) actuator paths 22a/22b, each with its own control device and motor 28 feeding assistance torque into the steering gear; overall and partial assistance torques are computed in each path; a master/slave scheme detects an error and swaps master/slave roles so the load is maintained substantially interruption-free; in one embodiment the second assistance torque is reduced relative to the first during failure when vehicle speed exceeds a threshold and steering dynamics are low.
• § 102 assessment: This is the closest reference on redundancy architecture and fault-driven switching, and it expressly ties the fault response to a vehicle-speed threshold — much nearer to the '776 "function of at least speed…" language than any other cited art. It is still missing the two dispositive claim-1 elements: (i) receiving a yaw rate and (ii) determining an expected range of steering angles against which the actual steering position is compared. → No anticipation. Its speed-threshold fault handling is the best § 103 hook against claims 1/3 and 9. Relevant claims if combined: 1, 3, 5, 9.

6. US 2018/0029637 A1 — Ford, "Enhanced steering operation"

• Dates: published 2018-02-01; earliest date 2016-07-29. ✔ prior art.
• Description (limited retrieval): I was unable to retrieve the full specification within my search budget. Based on the cited title, era (2016 priority), and Ford's steering portfolio, it concerns automated/assisted steering operation and driver-input handling. Treat the content characterization here as provisional.
• § 102 assessment: On the record available, no basis to find anticipation of claims 1/9/16; likely a general automated-steering/driver-input reference. Relevant claims if combined (provisionally): 1, 4, 20. Uncertain — flagged.

7. US 2018/0304918 A1 — Toyota, "Automatic driving system"

• Dates: published 2018-10-25; earliest date 2017-04-19. ✔ prior art.
• Description (limited retrieval): Full text not retrieved within budget; title/assignee/date are per the record. Toyota's corresponding era discloses automatic driving with steering control and driver-intervention handling. Provisional characterization.
• § 102 assessment: No anticipation support on the record. Likely background on automated steering. Uncertain — flagged.

8. KR 101878939 B1 — Mando Corp., "Active front steering apparatus for vehicle and controlling method thereof"

• Dates: granted 2018-07-16; earliest date 2017-06-14. ✔ prior art.
• Description (limited retrieval): Full text not retrieved within budget. An active front steering (AFS) apparatus — i.e., a steering system that superimposes an additional steering angle on driver input via a motor, with a control method.
• § 102 assessment: AFS superimposition is conceptually adjacent to "position control command that indicates an amount of angular displacement" (claim 6) and to a steering-wheel-coupled motor, but there is no indication of a yaw-rate-based expected range or a second redundant motor switched on range violation. → No anticipation. Relevant claims if combined (provisionally): 6, 7. Uncertain — flagged.

9. US 2021/0276615 A1 — TuSimple, "Managing redundant steering system for autonomous vehicles"

• Dates: published 2021-09-09; earliest date 2020-03-05 (same as '776).
• Assessment: This is the '776 patent's own pre-grant publication (same application family, same inventors, same priority). It is not prior art to itself and is listed in the table solely as a self/family citation. Disregard for § 102/§ 103 purposes.

10. US 2022/0089211 A1 — Ford, "Methods and apparatus to improve electronic power assisted steering in a vehicle"

• Dates: published 2022-03-24; earliest date 2020-09-18.
• Assessment: Because its earliest date (2020-09-18) is after the '776 priority date (2020-03-05), this reference is not prior art to the '776 patent under § 102(a)(1) or § 102(a)(2). Its appearance as an examiner citation is best explained by citation during prosecution of a later family member (e.g., continuation US 12,049,270, whose file has a later effective date), or as a cross-citation in the aggregated family record. For the '776 patent it has no § 102 or § 103 effect. Flagged explicitly rather than miscategorized.

Family-cite items

• US 10,996,673 B1 — Apple, "Manual override" (earliest date 2017-09-28). Predates 2020-03-05 ✔. Concerns manual override of an automated steering/control function — the closest cited art to the '776 driver-intervention/override discussion (see the '776 specification's statement that on exceeding a torque threshold the module "stops sending commands… so that the driver can steer"). No anticipation (no expected steering-angle range vs. speed+yaw rate). Potential § 103 relevance to claim 4's hand-on-wheel logic.
• US 2025/0333104 A1 — Shanghai Jiao Tong University, steer-by-wire steering feel (earliest date 2024-04-30). Post-dates priority; not prior art.

Non-patent citation

• Extended European Search Report for EP 21160671.0, dated 2021-07-02 (9 pp.). This is a prosecution document for the European family member, not prior art. Its chief value is evidentiary: it tells us which art the EPO examiner considered relevant and how the EPO assessed claim scope for the same invention. It is not citable as § 102/§ 103 art against '776.

Synthesis: most relevant prior art and claim mapping

Ranked relevance to the '776 claims:

1. DE 10 2015 116 937 A1 (Bosch) — redundant dual-actuator steering with fault-driven switching and a vehicle-speed threshold for the fault response. Closest art to claim 1's architecture and to claims 1/3/5/9; still lacks yaw rate and the expected-angle range.
2. WO 2017/028788 A1 (BYD) — steering-fault detection → fallback steering (torque-vectoring), with torque/direction signal outputs. Best § 103 support for the "detect failure, then steer by alternate means" step (claims 1, 4, 5, 6).
3. US 2017/0088167 A1 (Toyota) — two steering-associated actuators and driver-intervention detection, relevant to claims 4 and 6 and to the "different locations" concept (claim 15).
4. US 2012/0055730 A1 (Nippon Soken) — second motor activated on detection of abnormality; general redundancy (claims 1, 5, 7–8).
5. US 10,996,673 B1 (Apple) — manual override, relevant to claim 4's torque-threshold hand-on-wheel logic.
6. Remaining cited items (#2, #6, #7, #8) — background only, and three of them I could not fully retrieve; characterized provisionally.

Central § 102 conclusion: No reference in the cited set anticipates claims 1, 9, or 16, or any dependent claim, under 35 U.S.C. § 102. The reason is uniform across the set: no cited reference discloses the combination of (a) receiving a yaw rate, (b) determining an "expected range of steering angles" as a function of at least vehicle speed and yaw rate, and (c) comparing an actual steering position against that range and switching to a second steering-wheel-coupled motor on violation. The cited art shows the generic building blocks (redundant motors; fault detection and fallback; speed-dependent thresholds; first/second control commands; driver override), which is precisely the profile of an obviousness-type-over-background citation set, not an anticipation set. This corroborates the grant and the "expected range of steering angles" term identified as a construction battleground in the pending S.D. Tex. litigation summarized earlier.

Where the art could matter under § 103 (single-reference-plus-knowledge or combinations):

• Claim 1/9 combined-art hooks: Bosch (redundant path switching on fault + speed threshold) + BYD (fault→fallback steering) + Toyota (dual steering actuators/driver intervention) + knowledge of yaw-rate feedback for steering as of 2020.
• Claim 2/3 (pre-determined motor offsets folded into the range): no cited reference discloses motor-position offsets within a redundant steering system; Bosch's master/slave path identity is the nearest structural analog.
• Claim 4 (measured vs. expected torque → widen range): closest topical art is generally torque-sensor-based steering intervention detection; among the cited items, Toyota US 2017/0088167 (steering-wheel torque actuator/intervention) and Apple US 10,996,673 (override) are the best available hooks — but neither discloses widening an expected angle range.
• Claims 6/13/14/19 (position vs. torque command): Mando KR 101878939 (AFS superimposition) and Nippon Soken (first/second drive control parts) are the closest, provisionally.
• Claims 7/8/15 (column vs. gear motors at different locations): Bosch (actuator paths into the steering gear) and Nippon Soken (column-integrated motors) are the nearest structural disclosures.

Caveats (matters I could not confirm)

• Three references only partially characterized: US 2018/0029637 A1 (Ford), US 2018/0304918 A1 (Toyota), and KR 101878939 B1 (Mando) — I could not retrieve sufficient full text within this session's search budget. Their descriptions above are provisional and should be verified from the full documents before any § 102/§ 103 assertion. I did not invent content for them.
• No new/uncited prior art searched: You asked specifically about the citations of record for 11577776; I constrained the analysis to those. Prior art may exist outside this list (including the "Similar Documents" surfaced, e.g., US 12,179,805 B2 "Transitioning steering control from an autonomous vehicle to a driver," US 11,603,128 B2, and the EP search-report art), and third-party/IDS references may differ from examiner citations.
• Anticipation vs. obviousness: The distinction matters legally. I have stated anticipation only where a single reference discloses every claim element; where I list "relevant claims," I am identifying § 103 relevance, not § 102. No reference here meets § 102 for claims 1, 9, or 16.
• Prior-art status of items #9, #10, and US 2025/0333104: determined strictly from their earliest dates versus the '776 priority date (2020-03-05); none qualifies as prior art to the '776 patent. If you want them evaluated as art against a different family member with a later effective date, that requires a separate effective-date analysis.
• Identifiers interpreted literally: I did not auto-correct any number. Note the '776 record shows a certificate of correction dated 2023-04-11, so the printed claim text may differ in minor respects from the as-filed/as-granted text; the claim numbering used above follows the record you supplied.

If you'd like, I can (a) pull the full text of the three partially-retrieved references to firm up their § 103 mapping, or (b) run the cited-art set against US 12,049,270 B2 (the continuation), for which several of these references would be timely prior art.

I'll ground the obviousness analysis in the actual disclosures of the references cited on the patent's face. Let me pull each one.

Now the remaining references on the face of the patent.

Obviousness Analysis — U.S. Patent No. 11,577,776 B2

"Managing redundant steering system for autonomous vehicles" — 35 U.S.C. § 103

0. Scope, evidence base, and two threshold flags

0.1 Which prior art I used. You asked me to work from the Prior Art section of the patent page. That section supplies two distinct sets:

0.2 FLAG — one cited reference is NOT available as prior art. US20220089211A1 (Ford, "Methods and apparatus to improve electronic power assisted steering in a vehicle") carries a priority date of 2020-09-18 and a publication date of 2022-03-24. The '776 patent's effective filing/priority date is 2020-03-05. US20220089211A1 therefore post-dates the '776 priority date and cannot be § 102(a)(1) art (published after) or § 102(a)(2)/§ 102(e) art (effectively filed after). It must be excluded from any § 103 combination. Similarly, US20210276615A1 is the '776's own pre-grant publication (same application, US 16/810,781) and is not art against itself. That leaves eight genuinely citable references.

0.3 FLAG — date inconsistency in the task header. This task states "Current Date: April 26, 2026," while the operating instructions for this session state today is 2026-09-29, and the previously generated litigation section reports a Texas Fifteenth Court of Appeals opinion filed 2026-09-15. I treat 2026-09-29 as the operative date and note the header's April date as inconsistent. Nothing in the analysis below turns on the difference.

0.4 Evidentiary honesty. I retrieved substantive disclosure text for US20170088167A1, US20180304918A1, WO2017028788A1 (including its US national-phase counterpart US2018/0237059A1), US20120055730A1, and US20120303218A1. I did not retrieve full text for DE102015116937A1, KR101878939B1, US20180029637A1, US20220089211A1, or US10996673B1 (my retrieval budget was exhausted). Where I rely on those, I say so and label the inference as title/abstract-level only. Any conclusion resting on them should be re-verified against the full documents before being asserted in a validity opinion or an IPR petition.

1. Legal framework applied

• 35 U.S.C. § 103: a claim is obvious if the differences between the claimed subject matter and the prior art are such that the subject matter as a whole would have been obvious to a person having ordinary skill in the art (POSITA) at the time of the invention.
• Graham v. John Deere Co., 383 U.S. 1 (1966): (1) scope and content of the prior art; (2) differences between the prior art and the claims; (3) level of ordinary skill; (4) secondary considerations.
• KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007): a combination need not be taught expressly; any articulated reasoning with rational underpinning suffices. Relevant rationales here: (A) known technique used to improve a similar device in the same way; (B) combination of familiar elements according to known methods yielding predictable results; (C) a design incentive or market pressure present in the field; (D) "obvious to try" where there is a finite number of identified, predictable solutions.
• MPEP § 2143 lists the permissible rationales; MPEP § 2144.03 addresses applicant's own disclosure being usable as art. No teaching away, unexpected results, or criticality argument appears to be available here from the intrinsic record.

2. Level of ordinary skill in the art (POSITA)

A reasonable definition, calibrated to the '776 specification (FIG. 1–2; an in-vehicle control computer 150 running a steering control module 165, plus steering column steer drive 204 / steering gear steer drive 208):

A bachelor's degree in mechanical, electrical, or control-systems engineering (or equivalent) plus 2–4 years of experience in automotive steering systems and vehicle dynamics control, including electric power steering (EPS), steer-by-wire / redundant steering architectures, and vehicle-state estimation (speed, yaw rate, steering angle). A master's degree plus 1–2 years, or equivalent industry experience, would also qualify.

Significance: a POSITA at the 2020 priority date would be thoroughly familiar with (a) bicycle-model kinematics relating steering angle, vehicle speed, and yaw rate, (b) ASIL-D / fail-operational steering redundancy for SAE L3+ automation, and (c) plausibility / residual-based fault detection (comparing an actuator's commanded state to a model-predicted expected state).

3. Claim 1 — element-by-element mapping against the available art

Claim 1 has four limitations. I decompose and map them:

3.1 The single genuine novelty point

Limitation 1.3 — using speed and yaw rate to construct an expected range of steering angles as a diagnostic envelope (rather than as a control target) — is the only element not squarely disclosed in the verified art. Everything else is squarely in the cited references. That is also exactly the term the parties are fighting over in CreateAI, Inc. v. Bot Auto, Inc. et al., No. 4:25-cv-03371 (S.D. Tex., filed 2025-07-21), per the previously generated litigation section.

4. Primary § 103 combinations

Combination 1 (strongest): Toyota US20180304918A1 + Nippon Soken US20120055730A1

What each reference supplies

• US20180304918A1 (Toyota) — a control apparatus 20 (ECU with processor + memory) for an autonomous vehicle that (i) sends steering commands to a motor 8 of an electric power steering apparatus 2 to make the vehicle "travel along the target traveling route," (ii) receives speed, steering angle, and yaw rate (plus torque) after/based on those commands, and (iii) has an automatic steering control part 26 computing lateral and yaw-angle deviations from the target route. URL: https://patents.google.com/patent/US20180304918A1
• US20120055730A1 (Nippon Soken) — a steering control apparatus with two motors in one steering system: a first motor driving a variable gear ratio device (VGRS) on the column and a second motor performing power steering (EPS). Critically, it discloses fault detection in the first device ("VGRS apparatus abnormality check processing," FIG. 20) and, "when abnormality is detected in the variable gear ratio device," a second drive control part that takes over and "controls drive of the second motor based on the steering wheel angle and the speed increase ratio." URL: https://patents.google.com/patent/US20120055730A1

Why a POSITA would combine them (articulated reasoning)

1. Rationale (A) — known technique, same way. Nippon Soken's two-motor, fault-detecting steering architecture is a known steering-system redundancy technique. Applying it to Toyota's autonomous steering ECU is using the known technique for its known purpose (keeping steering authority when a steering actuator faults).
2. Rationale (C) — design incentive / market pressure. By the 2020 priority date, SAE L3+ automated driving programs required fail-operational steering (ASIL-D). Toyota's own reference is directed to "autonomous driving," so a POSITA tasked with hardening it would look precisely to redundant-actuator steering systems such as Nippon Soken's or BYD's.
3. Rationale (B) — predictable results. Both references are in the same field (B62D 5/04, B62D 6/00), and Toyota already routes steering angle, vehicle speed, and yaw rate through the same ECU that issues the steering command; the combination requires no change in the principal operating principle of either reference.
4. No teaching away. Nippon Soken's detection-and-takeover is presented as a benefit, not a disfavored option.

The remaining gap and how it is closed. Limitation 1.3's "expected range ... as a function of at least the speed of the vehicle and the yaw rate." Under KSR, the gap is bridged by the following reasoning (which needs to be supported by an additional reference or by the general knowledge of a POSITA):

• Toyota's vehicle information acquiring part 22 already delivers steering angle, vehicle speed, and yaw rate into one processor (US20180304918A1, [0046]-[0052] region). Computing, from those same three quantities, the band of steering angles consistent with the vehicle's observed speed and yaw rate is the application of the bicycle model / steady-state yaw-rate relationship — textbook vehicle-dynamics mathematics (γ ≈ V·δ/(L + K·V²)).
• Constructing a plausibility window around a model-predicted actuator state and flagging when a measured value falls outside it is a routine and well-known diagnostic technique in automotive electronics (model-based residual / plausibility checking, an established practice in EPS and ISO 26262 safety monitors). Nippon Soken's own FIG. 20 "abnormality check processing" shows that the field routinely builds fault criteria into the steering controller.
• Therefore: combining the Toyota ECU (which measures the inputs) with a conventional plausibility-window criterion (which a POSITA would naturally apply), and then wiring the window-exceedance output to the Nippon Soken takeover path, is the predictable use of familiar elements according to known methods. This is the weakest link in the chain, and it is where a validity challenge would be won or lost.

Combination 2 (alternative): Toyota US20170088167A1 + BYD WO2017028788A1

• US20170088167A1 (Toyota) supplies the automated driving apparatus with a steering actuator controlling steering angle to a first target steering angle while a steering wheel actuator controls wheel torque, plus steering-operation detection and the recognition of "speed, acceleration, yaw rate, and position" as the vehicle's traveling state.
  URL: https://patents.google.com/patent/US20170088167A1
  (The granted counterpart of this family is US 9,796,416 B2 — also a Toyota automated-driving reference, though not on the '776 citation list.)
• WO2017028788A1 (BYD) supplies fault-triggered redundancy with state feedback: a detection controller (6) "configured to determine whether a steering assist device (2) is normal, to continue to determine whether the steering assist device is normal if yes, and to control a drive rack (5A) of the first electric motor (4) to drive wheels ... to return and to output a steering failure signal, a steering wheel torque signal and a direction signal if no"; with a second electric motor (14) engaged thereafter. It also lists a yaw rate sensor (11), lateral acceleration sensor (10), and vehicle speed sensor (11A).
  URL: https://patents.google.com/patent/WO2017028788A1 (US counterpart: US 2018/0237059 A1)
• Why combine: same field (B62D 5/04, B62D 6/00); BYD expressly frames its problem as "how to prevent a steering control malfunction," which is precisely the safety problem an automated-driving system presents; and BYD already fuses yaw rate + speed with steering-failure logic, giving a POSITA a ready-made information set on which to base a plausibility criterion.

Weakness of Combination 2: BYD's second actuator is a traction motor driving a wheel, not a motor "coupled to the steering wheel," so it maps less cleanly onto claim 1 step 1.4 than Nippon Soken's second (EPS) motor. Combination 1 is the better vehicle for that limitation.

Combination 3 (secondary failover architecture): Nippon Soken US20120055730A1 as primary

Nippon Soken alone discloses nearly the entire architecture: two motors in the steering system, a first drive control part for the first motor, an abnormality check, and a second drive control part for the second motor triggered by that abnormality. Adding Toyota's yaw-rate/speed acquisition (Combination 1) or US20180304918A1's autonomous-route steering supplies the automated-driving context and the yaw-rate input. This is a useful alternative theory if a fact-finder rejects the Toyota-as-primary framing.

5. Dependent-claim analysis

6. The counter-case (what a patent owner will argue)

Any honest § 103 opinion must state the strongest non-obviousness positions, because the '776 patent survived prosecution and is now being asserted:

1. No reference teaches the diagnostic framing. In the cited art, yaw rate is a control target, not a fault criterion: US20170088167A1 uses target steering angles and reaction-force shaping; US20180304918A1 computes lateral/yaw deviations for path following; US20180304918A1's companion Toyota art (US 8,996,254, which appears in my search results but not on the '776 citation list) uses target yaw rate for behavior control. The '776 instead asks: given the command I just sent, what band of steering angles is physically consistent with the speed and yaw rate I now observe, and is the measured steering position inside it? That inversion — from control to diagnosis of the actuator itself — is a legitimate point of novelty.
2. The specific input triple is not arbitrary. Claim 1 requires speed AND yaw rate AND steering position, received after the command. None of Nippon Soken's, BYD's, or Toyota's fault tests is shown to use that triple in that way.
3. Secondary considerations are unquantified. There is no evidence of record of unexpected results, long-felt need, industry praise, or licensing tied to the '776 patent specifically. In litigation, the patent owner would likely press Nexus to the accused products' "full redundancy" marketing (referenced in the complaint at ¶76), but the previously generated litigation summary notes the state-court trade-secret case found the redundancy concept itself was not secret — which cuts against a strong secondary-considerations story.
4. The examiner allowed claim 1, and the cited references were before the examiner. A challenger therefore needs art beyond the eight citable references — which means the § 103 case above, standing alone on the patent's own citation list, is suggestive but not conclusive at the "expected range" limitation.

Conclusion on claim 1: Claims 1, 9, and 16 are vulnerable to a § 103 challenge built on Toyota (US20180304918A1 and/or US20170088167A1) in view of Nippon Soken (US20120055730A1), with BYD (WO2017028788A1) as an alternative redundancy reference and Ford (US20180029637A1) / Apple (US10996673B1) for claim 4. The combination is well-motivated under KSR rationales (A), (B), and (C). But the "expected range of steering angles as a function of speed and yaw rate" limitation is not expressly disclosed in any cited reference, and a validity attack that does not add a dedicated model-based plausibility-checking reference will likely fail. Whether the gap is bridged by "general knowledge of a POSITA" (bicycle-model kinematics + residual-based plausibility monitoring) is the decisive question, and it is a factual one for a jury or the PTAB.

7. Recommendations for a fuller analysis

1. Verify the four references I could not open. Obtain full text for DE102015116937A1 (Bosch), KR101878939B1 (Mando), US20180029637A1 (Ford), and US10996673B1 (Apple) before relying on claims 4, 5, 8, and 11 arguments.
2. Search for the missing limitation. Target art on model-based steering-actuator plausibility windows, i.e. comparisons of commanded steering versus an expected steering band derived from vehicle speed and yaw rate, and steering-angle/estimated-angle plausibility monitoring for EPS. This is the limitation that decides the case.
3. Consider § 112 and § 102 in parallel. The claim term "expected range of steering angles" is the disputed construction in CreateAI v. Bot Auto (4:25-cv-03371); an indefiniteness or written-description attack on that term may be more productive than § 103.
4. Check the family. The continuation US 12,049,270 B2 and pending US 2024/0375708 A1 (from US 18/784,321) contain the same disclosure; any art found against the '776 should be run against those claims too. The foreign family members are EP 3875349 A1 (withdrawn), CN 113353146 A (pending), and AU 2021200976 A1 (abandoned) — the EPO's Extended European Search Report dated 2021-07-02 (the sole non-patent citation of record) may itself reveal additional art.
5. Do not use US 2022/0089211 A1. As shown in § 0.2, it post-dates the '776 priority date despite appearing on the patent's own citation list (its citation was as later art, not as § 102/§ 103 art).

Sources cited

• Full patent text (authoritative): https://patents.google.com/patent/[US11577776](/patent/US11577776)/en
• US20180304918A1 (Toyota): https://patents.google.com/patent/US20180304918A1
• US20170088167A1 (Toyota): https://patents.google.com/patent/US20170088167A1 · https://www.freepatentsonline.com/y2017/0088167.html
• US20120055730A1 (Nippon Soken): https://patents.google.com/patent/US20120055730A1 · https://www.patents-review.com/a/20120055730-steering-control-apparatus.html
• WO2017028788A1 (BYD) and US counterpart US2018/0237059A1: https://patents.google.com/patent/WO2017028788A1 · https://patents.justia.com/patent/20180237059
• US20120303218A1 (Honda): https://patents.google.com/patent/US20120303218
• Litigation reference for the disputed claim term: CreateAI, Inc. v. Bot Auto, Inc. et al., No. 4:25-cv-03371 (S.D. Tex.) — https://portal.unifiedpatents.com/litigation/Texas%20Southern%20District%20Court/case/4%3A25-cv-03371