DETAILED ACTION
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on XXXXXXXXXXXXXX has been entered.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Status of Claims
Claims X are canceled.
Claims X are new.
Claims 1-20 are pending and have been examined.
This action is in reply to the papers filed on 09/02/2024 (effective filing date 09/02/2024).
Information Disclosure Statement
No Information Disclosure Statement has been filed.
The information disclosure statement(s) submitted: xxxxxxxx, has/have been considered by the Examiner and made of record in the application file.
Amendment
The present Office Action is based upon the original patent application filed on xxx as modified by the amendment filed on xxx.
Terminal Disclaimer
The terminal disclaimer filed on xxx disclaiming the terminal portion of any patent granted on this application which would extend beyond the expiration date of US Pat. No. xxxx has been reviewed and has been placed in the file.
Examiner acknowledges Applicant’s filed Terminal Disclaimer to prior art patent McCauley et al. US Pat. No. 5,930,775. A terminal disclaimer may be filed to overcome or obviate a nonstatutory double patenting rejection (37 CFR 1.321; MPEP 706.02; 1490).
Double Patenting - Withdrawn
The double patenting rejection is withdrawn per the filed terminal disclaimer noted above.
Reasons For Allowance
Prior-Art Rejection withdrawn
Claims xxx are potentially allowable over the prior-art, but, are still subject to the 35 USC §101 and/or 35 USC §112 rejections herein. The closest prior art (See PTO-892, Notice of References Cited) does not teach the claimed:
Claims xxx are allowed. Independent claims X, Y, and Z all contain the same inventive scope. The closest prior art (See PTO-892, Notice of References Cited) does not teach the claimed:
The invention teaches… and the prior-art teaches…, however, the prior-art does not teach…
The closest prior-art (xxx) teach the features as disclosed in Non-final Rejection (xxxx), however, these cited references do not teach and the prior-art does not teach at least the following combination of features and/or elements:
determining, at a second time after associating the information corresponding to the first loyalty card with the logged location, that a second user computing device is located within a specified distance of the logged location using a second positioning system of the second user computing device; in response to determining that the second user computing device is located within the specified distance of the logged location of the first user computing device at the first time of detecting: retrieving information corresponding to a second loyalty card, the second loyalty card being associated with the merchant and the second user computing device; and displaying, by the second user computing device, data describing the second loyalty card.
Claim Rejections - 35 USC §101 - Withdrawn
Per Applicant’s amendments and arguments and considering new guidance in the MPEP, the rejections are withdrawn. Specifically, in Applicant’s Remarks (dated 03/14/2017, pgs. 8-11), Applicant traverses the 35 USC §101 rejections arguing that the amended claims recite new limitations that are not abstract, amount to significantly more, are directed to a practical application, etc… For example, Applicant argues….
In support of their arguments, Applicant cites to the following recent Fed. Cir. court cases (i.e., Alice Corp. v. CLS Bank Int’l, SRI Int’l, Inc. v. Cisco Systems, Inc., Ultramercial, Inc. v. Hulu, LLC, Berkheimer, Core Wireless, McRO, Enfish, Bascom, DDR, etc…).
Claim Rejections - 35 USC § 101
35 U.S.C. § 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1–20 are rejected under 35 U.S.C. § 101 because the claimed invention is directed to a judicial exception (an abstract idea) without significantly more.
Independent Claims 1 and 11 — Step 1 (Statutory Category)
Claim 1 recites "[a] method of a pixel based with object based decision making for driving" and is therefore directed to a process, one of the four statutory categories under 35 U.S.C. § 101. Claim 11 recites "[a] non-transitory computer readable medium ... stores instructions executable by a processing circuit," and is therefore directed to a manufacture, likewise a statutory category. Because claim 11 expressly recites "non-transitory," it does not encompass a transitory propagating signal and no signal-per-se rejection is warranted. No software-per-se issue arises because claim 11 is directed to the medium storing the instructions, not to the software "per se" in the abstract.
Step 2A, Prong One — Recitation of a Judicial Exception
Claim 1 recites:
"receiving, at a first machine learning process of an artificial intelligence agent, a sensed information unit; receiving, at a second machine learning process of the artificial intelligence agent, object descriptive information regarding an object captured in the sensed information unit; ... generating, by the first machine learning process, a pixel-based path planning output ...; generating, by the second machine learning process, an object-based path planning output ...; and generating, by at least in part processing the pixel-based path planning output in correspondence with the object-based path planning output, a driving related output ..., such that the driving related output conforms to at least one of the pixel-based path planning output and the object-based path planning output."
Stripped of the generic computer/machine-learning labels ("machine learning process," "artificial intelligence agent") under which the steps are performed, claim 1 recites: (i) receiving two forms of information describing a scene; (ii) evaluating each independently to produce a candidate suggestion; and (iii) reconciling the two candidate suggestions — by comparing, weighting, or selecting between them — to reach a single output. The specification confirms that this reconciling step is, at bottom, a set of mental and mathematical operations: paragraph [0064] of the as-filed specification lists the available implementations of the "generating" step as including "[p]roviding weights," "[g]enerating ... based on the weights," "[d]etermining a difference," "[s]electing one of" the two outputs, "[f]using" the outputs, "[g]enerating ... by averaging," and "[s]electing ... the path segment that is safer" or has "a lower acceleration value," "a lower speed value," etc. Each of these is a mental process a person can (and routinely does) perform — weighing one impression against another and picking or blending between them — or a mathematical concept (a weighted sum, an average, a threshold comparison). This is squarely within the "[m]ental processes" and "[m]athematical concepts" groupings identified in MPEP § 2106.04(a)(2), and is analogous to the abstract idea of collecting data, analyzing it, and using the result to make a determination, held ineligible in Electric Power Group, LLC v. Alstom S.A., 830 F.3d 1350, 1353–54 (Fed. Cir. 2016) ("collecting information, analyzing it, and displaying certain results of the collection and analysis" is abstract even when limited to a particular technological environment), and to the abstract idea of a mathematical algorithm for analyzing data, held ineligible in SAP America, Inc. v. InvestPic, LLC, 898 F.3d 1161, 1167–68 (Fed. Cir. 2018) ("nothing in the claims, understood in light of the specification, requires anything other than off-the-shelf, conventional computer, network, and display technology").
Claim 11 recites the same steps, verbatim, as instructions executable by a processing circuit, and is directed to the same abstract idea for the same reasons.
Step 2A, Prong Two — Integration into a Practical Application
The only additional elements recited beyond the abstract idea are "a vehicle," "an artificial intelligence agent," "a first machine learning process," "a second machine learning process," and, in claim 11, "a processing circuit." None of these integrates the abstract idea into a practical application.
"[A] vehicle" merely identifies the technological field in which the abstract comparison-and-selection idea is applied; limiting an abstract idea to a particular technological environment or field of use does not integrate the idea into a practical application. MPEP § 2106.05(h); Electric Power Group, 830 F.3d at 1354. "[A]n artificial intelligence agent" and "a first/second machine learning process" are recited at a high level of generality and invoked only as black-box tools that "receiv[e]" data and "generat[e]" an output — the claim recites no particular network architecture, no particular training regime, and no particular technical mechanism by which the pixel-based and object-based outputs are actually fused beyond the functional, result-based phrase "processing ... in correspondence with." The claim does not recite any improvement to how a computer, a machine-learning model, a sensor, or a vehicle-control system operates; it recites only the result to be achieved (a "driving related output" that "conforms to at least one of" the two intermediate outputs), which is precisely the kind of result-oriented functional claiming that fails to integrate a judicial exception into a practical application. MPEP § 2106.05(f).
Step 2B — Inventive Concept ("Significantly More")
Considered individually and as an ordered combination, the additional elements are used in a purely conventional manner. "[R]eceiving" data and "generating" an output are the generic functions expected of any computer or machine-learning implementation. The specification confirms the conventionality of the recited hardware and software: the memory/storage units, processor, and communication system are described in wholly generic, off-the-shelf terms (¶¶ [0017]–[0034]), and the first machine learning process is expressly described as optionally "implemented by one or more transformers, by a transfuser" citing, as a known pre-existing example, “According to an embodiment the first machine learning process is implemented by one or more transformers, by a transfuser (when fed with two types of sensed information units)—see TransFuser: Limitation with Transformer-Based Sensor Fusion for Autonomous driving, K. Chitta, el at., arXiv:2205.15997” (¶ [0158]) — an admission that no new or improved machine-learning architecture is being claimed, only the abstract idea of running two already-known types of machine-learning processes and reconciling their outputs by conventional means (weighting, averaging, thresholding). Invoking already-known, generic technology to implement an abstract idea does not supply the inventive concept required by Step 2B. Alice, 573 U.S. at 225–26; buySAFE, Inc. v. Google, Inc., 765 F.3d 1350, 1355 (Fed. Cir. 2014) (adding a generic computer to perform generic functions does not confer eligibility).
[0158] A. A first machine learning process 521 is configured to receive a sensed information unit 529 and generate the pixel-based path planning 525. According to an embodiment the first machine learning process is implemented by one or more transformers, by a transfuser (when fed with two types of sensed information units)—see TransFuser: Limitation with Transformer-Based Sensor Fusion for Autonomous driving, K. Chitta, el at., arXiv:2205.15997. Using multiple transformers though is that the output is embeddings representing a birds eye view of the ego surroundings and it absorbs features at different depth in the backbone (high and low level abstraction).
Dependent Claims
No dependent claim adds an element or combination of elements sufficient to integrate the abstract idea into a practical application or to supply an inventive concept.
Claims 2 and 12 ("selecting the artificial intelligence agent out of a group of artificial intelligence agents") add only a further selection step — choosing which generic evaluator to consult — which is itself an unclaimed-technology mental/selection process. No technical mechanism for the selection is recited.
Claims 3 and 13 ("determining a scenario being faced by the vehicle, based on the sensed information unit; wherein the selecting ... is based [on] the scenario") add a data-classification step (categorizing the scenario) that is itself abstract data recognition, analogous to "collecting data, recognizing certain data within the collected data set" held abstract in *Content Extraction & Transmission LLC v. Wells Fargo Bank, Nat'l Ass'n*, 776 F.3d 1343, 1347 (Fed. Cir. 2014).
Claims 4 and 14 (confidence levels for each output, with the final output "responsive to" both) add a further mathematical weighting concept (a confidence-weighted composite), no different in character from the weighting/averaging already found abstract with respect to claim 1.
Claims 5 and 15 (object descriptive information limited to "object location information" and "kinematic information") merely narrow the content of the data gathered in claim 1's data-receiving step. Mere data-gathering limitations, including limitations on what data is gathered, do not add significantly more. *CyberSource Corp. v. Retail Decisions, Inc.*, 654 F.3d 1366, 1370 (Fed. Cir. 2011); *Electric Power Group*, 830 F.3d at 1355.
Claims 6 and 16 ("based in part on a safety parameter") and claims 7 and 17 ("based in part on a comfort ... of a passenger") each add a further factor or weighting criterion to the abstract weighting/selecting step, without reciting any technical mechanism for incorporating that factor.
Claims 8 and 18 ("identifying that the pixel-based path planning output contradict[s] the object-based path planning output and responding to the contradiction") recite a further mental comparison (noticing that two observations disagree) coupled with purely functional, result-based "responding" language that does not specify any particular technical response.
Claims 9 and 19 (a second, concurrently operating artificial intelligence agent performing the same generic receiving/generating steps) merely duplicate the same abstract analysis using a second generic component. Adding another instance of the same generic technology performing the same generic function does not confer eligibility. *buySAFE*, 765 F.3d at 1355.
Claims 10 and 20 ("generating ... a further driving related output" from the outputs of claim 9/19) recite the identical mathematical/mental combination step as claim 1's final "generating" limitation, applied one additional layer up, and add nothing beyond what was already found abstract.
Suggestion for Overcoming the Rejection
A claim amendment that recites a specific technical improvement in how the pixel-derived and object-derived information is actually fused within the machine-learning architecture itself — for example, a specific cross-attention or joint-embedding mechanism operating on pixel-level feature maps and object-level tokens within a single network (rather than reciting the result of "processing ... in correspondence"), together with a specific, non-generic recitation of the sensing and vehicle-control hardware that carries out the claimed steps and a specific technical benefit realized by that mechanism (e.g., a measurable reduction in path-planning latency or a measurable increase in perception robustness attributable to the specific fusion architecture) — would likely address this rejection by directing the claim to a specific improvement in autonomous-vehicle perception/planning technology rather than to the result of comparing two assessments.
Claim Rejections - 35 USC §112
The following is a quotation of 35 U.S.C. § 112(a) and (b):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art ... to make and use the same ....
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
§ 112(a) — Written Description
Claims 11–20 are rejected under 35 U.S.C. § 112(a) as failing to comply with the written description requirement. Independent claim 11's preamble recites "[a] non-transitory computer readable medium for interactive neural network training for pixel based with object based decision making for driving," yet the body of the claim recites only real-time, inference-time operations (receiving a sensed information unit and object descriptive information; generating outputs; generating a driving related output). No step of the claim performs or describes "training" — there is no recitation of, and the Detailed Description's corresponding disclosure of methods 200/300 (¶¶ [0048]–[0106]) does not describe, any ground-truth comparison, loss computation, weight/parameter update, or an "interactive" loop in which a human or system iteratively refines the network. The specification does not demonstrate possession of a method of "interactive neural network training" corresponding to the claimed steps; the claimed steps instead describe using an already-trained network at inference time. Claims 12–20 depend from claim 11 (directly or indirectly) and inherit the same defect without curing it. Applicant should either amend the preamble of claim 11 to remove "interactive neural network training" (if the body is intended to claim inference-time operation, as it appears to) or amend the body to actually recite training-specific steps supported by the specification.
§ 112(b) — Indefiniteness (including antecedent basis)
Claims 1 and 11 are rejected under 35 U.S.C. § 112(b) as indefinite. Each recites "wherein the first machine learning process and the second machine learning running concurrently" — omitting "process" after "the second machine learning." As literally written, "the second machine learning" (without "process") has no clear antecedent basis distinct from "the second machine learning process" introduced earlier in the claim, and it is unclear whether a person of ordinary skill would read this as referring back to "the second machine learning process" or as introducing an undefined new term "the second machine learning." The metes and bounds of the claim are consequently unclear. Correction to "the second machine learning process" is suggested.
Claims 3 and 13 are rejected as indefinite (minor issue). Each recites that "the selecting of the artificial intelligence agent is based in the scenario" — "based in" is nonstandard usage, and it is unclear whether "in" was intended as "on." Examiner suggests "based on the scenario."
Claims 5 and 15 are rejected as indefinite (minor issue). Each recites "kinematic information indicative a relative velocity," omitting "of" ("indicative of a relative velocity"), creating a grammatically incomplete limitation whose scope is not entirely clear on its face.
Claim 9 is rejected as indefinite. The claim recites "selecting, concurrently with the selecting of the artificial intelligence agent, another artificial intelligence agent." Claim 9 depends from claim 1, not claim 2. Claim 1 never recites a step of "selecting" the (first) artificial intelligence agent — only claim 2 does. Accordingly, "the selecting of the artificial intelligence agent" lacks antecedent basis when claim 9 is read together with claim 1, and the metes and bounds of claim 9 are unclear (it is not clear what act "concurrently" is measured against). This issue would be resolved if claim 9 depended from claim 2 instead of claim 1, or if claim 9 itself recited the missing "selecting" step. Claim 10 inherits this defect by dependency without adding a further issue.
Claim 19 is rejected as indefinite for the same reason as claim 9: claim 19 depends from claim 11, which, like claim 1, never recites a "selecting" step, so "the selecting of the artificial intelligence agent" again lacks antecedent basis. Claim 20 inherits this defect by dependency without adding a further issue.
Claims 11 and, by dependency, 12–20 are further rejected as indefinite for the same "interactive neural network training" preamble/body mismatch identified under § 112(a) above: it is not clear whether the "for interactive neural network training" language is (i) mere intended-use language that does not limit claim scope, or (ii) a substantive limitation requiring the instructions to actually perform training — and if the latter, the body of the claim does not appear to recite any training step, creating an internal inconsistency that renders the scope of claims 11–20 unclear.
Claims 13, 14, 18, and 19 are rejected as indefinite for a shared drafting defect in the boilerplate "for selecting" lead-in used to introduce each dependent CRM claim's added instructions. Claim 13 recites "further storing instructions executable by the processing circuit for selecting determining a scenario being faced by the vehicle" — "for selecting determining" is not grammatical English, and it is unclear whether "selecting" was intended to be deleted (leaving "for determining") or whether a separate, unrecited "selecting" step and object were intended; to the extent claim 13 (via claim 12, via claim 11) also intends to mirror claim 3's "based [on/in] the scenario" language, the same minor issue noted for claim 3 applies as well. Claim 14 recites "further storing instructions executable by the processing circuit for selecting: determining, by the first machine learning process, a suggested pixel-based path segment confidence level; and determining ..." — again, the lead-in "for selecting:" is followed by two "determining" sub-steps that do not select anything, making it unclear what "selecting" refers to. Claim 18 recites "for selecting identifying that the pixel-based path planning output contradict the object-based path planning output" — the same "for selecting [X]" indefiniteness defect, compounded by a subject-verb agreement error ("output contradict" rather than "output contradicts") that, while primarily grammatical, contributes to the lack of clarity as to what is actually being claimed. Claim 19 recites the same "for selecting: selecting, ..." construction, compounding its independent antecedent-basis defect (discussed above) with this indefiniteness issue.
Claims 2, 4, 6, 7, 8, 10 (beyond the claim 9 issue it inherits), 12, 16, 17, and 20 (beyond the claim 19 issue it inherits) raise no further issue under § 112(b) and are not otherwise rejected on this ground. Claim 15 inherits only the minor issue noted for claim 5 above.
§ 112(a)/(b) — Suggested Claim Fixes
For claims 1 and 11: insert "process" after "the second machine learning" in the concurrency clause.
For claim 3 (and its CRM counterpart claim 13): change "based in the scenario" to "based on the scenario."
For claim 5 (and claim 15): change "indicative a relative velocity" to "indicative of a relative velocity."
For claim 9: either amend to depend from claim 2, or add a "selecting the artificial intelligence agent" step directly in claim 9.
For claim 19: either amend to depend from claim 12, or add the corresponding "selecting" instruction directly.
For claims 13, 14, 18, and 19: delete the stray word "selecting" from the "for selecting [X]" lead-in phrase (i.e., "for determining ...," "for identifying ...," "for selecting, ...," matching the corresponding method claim's actual verb) throughout claims 12–20's boilerplate lead-in.
For claim 11 (and claims 12–20): either delete "for interactive neural network training" from the preamble or amend the body to actually recite a training step.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-3, 5, 11-13, 15 are rejected under 35 U.S.C. 103 as being unpatentable over: Raichelgauz et al., US 2023/0064387 A1 (Raichelgauz); in view of Chen et al. US 2021/0264167 A1 (Chen).
18/822,285 – Claim 1. Raichelgauz et al., 2023/0064387 (Raichelgauz) teaches A method of a pixel based with object based decision making for driving, the method comprises (Raichelgauz [0022 - pixel][0085 - Step 450 may be executed without human driver intervention and may include changing the speed and/or acceleration and/or the direction of progress of the vehicle. This may include performing autonomous driving or performing advanced driver assistance system (ADAS) driving operations that may include momentarily taking control over the vehicle and/or over one or more driving related unit of the vehicle.]; Method 100 (¶¶ [0030]–[0039]) and Method 400 (¶¶ [0056]–[0086])): receiving, at a first machine learning process of an artificial intelligence agent, a sensed information unit (Raichelgauz ¶ [0098] (Method 700, Step 710): "receiving a sequence of panoptically segmented images over short time window from ego perspective (images obtained by the ego vehicle), relative distance to individual objects X_rel,i"; ¶ [0099] (Step 720): "applying spatio-temporal CNN to individual instances (objects) to capture high-level spatio-temporal features X_i" (first ML process = the spatio-temporal CNN of Steps 710–720; "artificial intelligence agent" = the neural network selected from Raichelgauz's own "group of neural networks," Method 400, Steps 431–432 (¶¶ [0078]–[0079]).); receiving, at a second machine learning process of the artificial intelligence agent, object descriptive information regarding an object captured in the sensed information unit (Raichelgauz ¶ [0105] (Method 800, Step 810): "receiving a list of detected object relative kinematics (X_rel,i, V_rel,i) wherein X_rel,i is a relative location of detected object i-in relation to the ego vehicle and V_rel,i is a relative velocity of detected object i— in relation to the ego vehicle. Also receiving the ego vehicle velocity V_ego"; ¶ [0106] (Step 820): calculating, for each object, the perception field f_θ(X_rel,i, V_rel,i, V_ego,i)); the object descriptive information is less detailed than the sensed information unit (Inherent from the contrast between Raichelgauz ¶ [0098] (a full sequence of panoptically segmented images) and ¶ [0105] (a compact per-object list of two scalar kinematic quantities, X_rel,i and V_rel,i, plus V_ego) This is a matter of common sense/inherency to a person of ordinary skill (MPEP § 2144); Applicant's own specification confirms this is simply an expected, non-technical property of any object list versus raw sensed data (¶ [0159]: object descriptive information may be "less than 0.1 ... 10 percent" the size of the sensed information unit).); generating, by the first machine learning process, a pixel-based path planning output related to a suggested pixel-based path segment of a vehicle (Raichelgauz ¶ [0100] (Step 730): "computing individual perception fields f_θ(X_i,i)" and summing "Σf_θ(X_rel,i, X_i,i)"; ¶ [0101] (Step 740): "constructing a differential equation that describes ego acceleration applied on the ego vehicle: a = Σf_θ(X_rel,i, X_i,i)" (output of the inference process, ¶ [0102]); Raichelgauz ¶¶ [0037]–[0038] (Method 100, Steps 150–160): converting the resulting virtual force to "a desired (or target) virtual acceleration" and then "to one or more vehicle driving operations that will cause the vehicle to propagate according to the desired virtual acceleration" — i.e., a resulting path segment. Claim interpretation: under BRI, a "path planning output related to a suggested path segment" reads on a virtual force/desired acceleration output that dictates how the vehicle will propagate, because converting that force to a driving operation necessarily determines a resulting path segment.); generating, by the second machine learning process, an object-based path planning output related to a suggested object-based path segment of the vehicle (Raichelgauz ¶ [0106] (Step 820): "calculating for each object the perception field f_θ(X_rel,i, V_rel,i, V_ego,i)"; ¶ [0107] (Step 830): "summing the contributions from individual perception fields," with normalization; ¶ [0108] (Step 840): "constructing a differential equation that describes ego acceleration applied on the ego vehicle: a = N*Σf_θ(X_rel,i, V_rel,i, V_ego,i)" (output of the inference process, ¶ [0109]). Same claim interpretation as above, applied to the kinematics-derived pathway.); and generating, by at least in part processing the pixel-based path planning output in correspondence with the object-based path planning output, a driving related output with respect to the vehicle, such that the driving related output conforms to at least one of the pixel-based path planning output and the object-based path planning output (Raichelgauz ¶ [0036] (Step 140): "performing a vector weighted sum (or other function) on the one or more virtual forces associated with the one or more objects" to obtain the total force from which the desired acceleration/driving operation is derived (¶¶ [0037]–[0038], Steps 150–160); Method 400, Steps 431–432 (¶¶ [0078]–[0079]) (selecting the neural network(s) used, based on the determined situation)), wherein the first machine learning process and the second machine learning running concurrently for decision making driving of the vehicle (Raichelgauz, overall real-time driving-operation pipeline (¶¶ [0036]–[0039], Steps 140–160), necessarily operating on whichever perception-field-generating processes (Methods 700/800) are active within a given real-time control cycle).
Raichelgauz may not expressly disclose the “artificial intelligence…” features, however, Chen teaches these features (Chen [0044 - The processing unit 116 may be configured to provide processing capabilities to be configured to utilize machine learning/deep learning to be utilized to analyze inputted data in the form of image data and may utilize a mask R-CNN 120 and a long short-term memory module (LSTM) 118 to provide artificial intelligence capabilities.][0052 - The object masking module 124 may be configured to utilize the neural network 108 to execute machine learning/deep learning processing to provide a one-channel binary mask on subsets of pixels of the image frame that are encapsulated within each of the bounding boxes that include each of the dynamic objects located within the driving scene.][0053 - he neural network 108 may complete image inpainting to electronically remove and replace each of pixels associated with each of the dynamic objects independently, such that each dynamic object is singularly removed and its removal is analyzed to output a level of driving behavior change with respect to the particular removed dynamic object] Before the effective filing date of the claimed invention, it would have been obvious for one of ordinary skill in the art to have modified Raichelgauz to include the features as taught by Chen. One of ordinary skill in the art would have been motivated to do so to utilize well known tools and features useful for pixel based with object based decision making for driving which should prove to improve user experience.).
18/822,285 – Claim 11. Raichelgauz further teaches A non-transitory computer readable medium for interactive neural network training for pixel based with object based decision making for driving, the non-transitory computer readable medium stores instructions executable by a processing circuit for (Raichelgauz [0016-0020; 0023; 0087; Fig. 5; claim 22]): … receiving, at a first machine learning process of an artificial intelligence agent, a sensed information unit; receiving, at a second machine learning process of the artificial intelligence agent, object descriptive information regarding an object captured in the sensed information unit; the object descriptive information is less detailed than the sensed information unit; generating, by the first machine learning process, a pixel-based path planning output related to a suggested pixel-based path segment of a vehicle; generating, by the second machine learning process, an object-based path planning output related to a suggested object-based path segment of the vehicle; and generating, by at least in part processing the pixel-based path planning output in correspondence with the object-based path planning output, a driving related output with respect to the vehicle, such that the driving related output conforms to at least one of the pixel-based path planning output and the object-based path planning output, wherein the first machine learning process and the second machine learning running concurrently for decision making driving of the vehicle.
Claim 11 recites the identical steps of claim 1 as instructions stored on a non-transitory computer readable medium and executable by a processing circuit, rather than as a method. Claim 11, has similar limitations as of Claim 1, therefore it is REJECTED under the same rationale as Claim 1.
18/822,285 – Claim 2. Raichelgauz further teaches The method according to claim 1, further comprising selecting the artificial intelligence agent out of a group of artificial intelligence agents (Raichelgauz, Abstract: determining, using one or more neural network (NNs); ¶ [0079] (Method 400, Step 432): "selecting the one or more NNs based on the situation"; corresponding as-published claim 10: "The method according to claim 9 comprising selecting the one or more NNs based on the situation". Raichelgauz teaches this limitation, motivated by the express benefit (stated in Raichelgauz ¶ [0042], describing the use of narrow, situation-specific driving functions (e.g., ACC, AEB, LCA) and training on atomic scenarios so the model can still properly handle more complicated ones, and corroborated by Applicant's own ¶¶ [0013]–[0014]) of improving accuracy by tailoring the active network to the situation presented.).
18/822,285 – Claim 12. The non-transitory computer readable medium according to claim 11, further storing instructions executable by the processing circuit for selecting the artificial intelligence agent out of a group of artificial intelligence agents.
Claim 12, has similar limitations as of Claim 2, therefore it is REJECTED under the same rationale as Claim 2.
18/822,285 – Claim 3. Raichelgauz further teaches The method according to claim 2, further comprising determining a scenario being faced by the vehicle, based on the sensed information unit; wherein the selecting of the artificial intelligence agent is based in the scenario (Raichelgauz ¶ [0022 – sensors and sensed information], ¶ [0028; 0042 - scenarios], ¶ [0078] (Method 400, Step 431): "determining a situation of the vehicle, based on the object information," followed by ¶ [0079]'s situation-based selection; corresponding as-published claim 9: "The method according to claim 1 comprising determining a situation of the vehicle, based on the object information".).
18/822,285 – Claim 13. The non-transitory computer readable medium according to claim 12, further storing instructions executable by the processing circuit for selecting determining a scenario being faced by the vehicle, based on the sensed information unit; wherein the selecting of the artificial intelligence agent is based in the scenario.
Claim 13, has similar limitations as of Claim 3, therefore it is REJECTED under the same rationale as Claim 3.
18/822,285 – Claim 5. Raichelgauz further teaches The method according to claim 1, wherein the object descriptive information comprises object location information relating to the captured object in an environment of the vehicle, and kinematic information indicative a relative velocity between the captured object and the vehicle (Raichelgauz ¶ [0105 - detected object relative kinematics] (Method 800, Step 810, quoted above): "X_rel,i is a relative location of detected object i-in relation to the ego vehicle and V_rel,i is a relative velocity of detected object i— in relation to the ego vehicle").
18/822,285 – Claim 15. The non-transitory computer readable medium according to claim 11, wherein the object descriptive information comprises object location information relating to the captured object in an environment of the vehicle, and kinematic information indicative a relative velocity between the captured object and the vehicle.
Claim 15, has similar limitations as of Claim 5, therefore it is REJECTED under the same rationale as Claim 5.
Claims 4, 6, 12, 16 are rejected under 35 U.S.C. 103 as being unpatentable over: Raichelgauz et al., US 2023/0064387 A1 (Raichelgauz); in view of Chen et al. US 2021/0264167 A1 (Chen); in further view of Senmyo US 2021/0261139 A1 (Senmyo).
18/822,285 – Claim 4. Raichelgauz further teaches The method according to claim 1, further comprising: determining, by the first machine learning process, a suggested pixel-based path segment confidence level; and determining, by the second machine learning process, a suggested object-based path segment confidence level; wherein the generating of the driving related output is responsive to the suggested pixel-based path segment confidence level and to the suggested object-based path segment confidence level (Raichelgauz, [0022 – pixel][0032 – path segment])(Senmyo ¶ [0054]: risk level evaluator 91 converts "the degrees of certainty of the risk states of the risk state prediction result obtained by the collision and contact determination unit 57 and the risk state prediction result obtained by the risk prediction unit 55 ... into numbers, and regard[s] the larger one of the numbers as the risk level," or alternatively "the sum of the numbers"; ¶ [0055]: the risk level evaluator "may weight each of" the two prediction results. Before the effective filing date of the claimed invention, it would have been obvious for one of ordinary skill in the art to combine Senmyo's confidence/reliability-responsive weighting technique with Raichelgauz's weighted-sum aggregation (¶ [0036], Step 140, which already sums pathway contributions) applies a known technique (confidence-based weighting, already used by Senmyo to combine two parallel risk assessments) to improve a similar method (Raichelgauz's weighted-sum combination of two parallel perception-field assessments) in the same way Senmyo already uses it, yielding the predictable result of a more reliable combined output.).
18/822,285 – Claim 14. The non-transitory computer readable medium according to claim 11, further storing instructions executable by the processing circuit for selecting: determining, by the first machine learning process, a suggested pixel-based path segment confidence level; and determining, by the second machine learning process, a suggested object-based path segment confidence level; wherein the generating of the driving related output is responsive to the suggested pixel-based path segment confidence level and to the suggested object-based path segment confidence level.
Claim 14, has similar limitations as of Claim 4, therefore it is REJECTED under the same rationale as Claim 4.
18/822,285 – Claim 6. Raichelgauz further teaches The method according to claim 1, wherein the generating of the driving related output is based in part on a safety parameter (Raichelgauz, Abstract) (Senmyo ¶¶ [0047]–[0050] (risk prediction unit 55 predicts a risk state of the vehicle) and ¶¶ [0054]–[0055] (combination of risk-state predictions is expressly risk/safety-driven). Before the effective filing date of the claimed invention, it would have been obvious for one of ordinary skill in the art to Incorporate a safety/risk parameter into Raichelgauz's combination step, as demonstrated by Senmyo's own use of a risk parameter for the same purpose (combining two parallel machine assessments), applies a known technique to a similar method in the same way, addressing the well-known, predictable design goal of prioritizing collision avoidance in driver-assistance/autonomous systems.
18/822,285 – Claim 16. The non-transitory computer readable medium according to claim 11, wherein the generating of the driving related output is based in part on a safety parameter.
Claim 16, has similar limitations as of Claim 6, therefore it is REJECTED under the same rationale as Claim 6.
Claims 7, 17 are rejected under 35 U.S.C. 103 as being unpatentable over: Raichelgauz et al., US 2023/0064387 A1 (Raichelgauz); in view of Chen et al. US 2021/0264167 A1 (Chen); in further view of Elsner, Optimizing Passenger Comfort in Cost Functions for Trajectory Planning, arXiv:1811.06895 (submitted Nov. 2018), Technische Universität München (Elsner).
18/822,285 – Claim 7. Raichelgauz further teaches The method according to claim 1, wherein the generating of the driving related output is based in part on a comfort of a passenger of the vehicle (Raichelgauz, [0041 - Representing ego movement as the composition of individual perception fields implies decomposing actions into more fundamental components and is in itself a significant step towards explainability. The possibility to visualize these fields and to apply intuition from physics in order to predict ego motion represent further explainability as compared to common end-to-end, black-box deep learning approaches. This increased transparency also leads to passengers and drivers being able to trust AV or ADAS technology more.])(Elsner, entire reference: incorporating a passenger-comfort term into the trajectory-planning cost function. Before the effective filing date of the claimed invention, it would have been obvious for one of ordinary skill in the art to incorporate Elsner's comfort-weighting term into Raichelgauz's combination step is a predictable application of a known trajectory-optimization technique to a known driving-decision framework, addressing the long-recognized, independent design objective of ride quality identified by Elsner.).
18/822,285 – Claim 17. The non-transitory computer readable medium according to claim 11, wherein the generating of the driving related output is based in part on a comfort of a passenger of the vehicle.
Claim 17, has similar limitations as of Claim 7, therefore it is REJECTED under the same rationale as Claim 7.
Claims 8, 18 are rejected under 35 U.S.C. 103 as being unpatentable over: Raichelgauz et al., US 2023/0064387 A1 (Raichelgauz); in view of Chen et al. US 2021/0264167 A1 (Chen); in further view of Molloy et al., Safety Assessment for Autonomous Systems' Perception Capabilities, arXiv:2208.08237 (posted Aug. 17, 2022)( Molloy).
18/822,285 – Claim 8. Raichelgauz further teaches The method according to claim 1, further comprising identifying that the pixel-based path planning output contradict the object-based path planning output and responding to the contradiction (Raichelgauz, Abstract)(Molloy, discussion of "[c]onflict between sensors' perception," "[c]onflict between Decision and Understanding," and the safety value of detecting such conflicts. Before the effective filing date of the claimed invention, it would have been obvious for one of ordinary skill in the art to combine Raichelgauz's dual-pathway architecture with Molloy's conflict-detection teaching would be motivated by the well-recognized safety benefit — articulated by Molloy itself, independently of Applicant's disclosure — of catching disagreements between complementary/redundant perception pipelines before they propagate into an unsafe output.).
18/822,285 – Claim 18. The non-transitory computer readable medium according to claim 11, further storing instructions executable by the processing circuit for selecting identifying that the pixel-based path planning output contradict the object-based path planning output and responding to the contradiction.
Claim 18, has similar limitations as of Claim 8, therefore it is REJECTED under the same rationale as Claim 8.
No Prior-art Rejection / Potentially Allowable
Claims 9, 10, 19, 20 cannot be rejected with prior-art. Individual claimed features are taught in the prior-art, however, the unique combination of features and elements are not taught by the prior-art without hindsight reasoning. These claims are further rejected as being dependent upon a rejected base claim but might possibly be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
18/822,285 – Claim 9. Raichelgauz further teaches The method according to claim 1, further comprising: selecting, concurrently with the selecting of the artificial intelligence agent, another artificial intelligence agent; receiving, at another first machine learning process of the other artificial intelligence agent, the sensed information unit; receiving, at another second machine learning process of the other artificial intelligence agent, the object descriptive information; generating, by the other first machine learning process, another pixel-based path planning output related to another suggested pixel-based path segment of the vehicle; generating, by the other second machine learning process, another object-based path planning output related to another suggested object-based path segment of the vehicle; and generating, by at least in part processing the other pixel-based path planning output in correspondence with the other object-based path planning output, another driving related output with respect to the vehicle (Raichelgauz's own group-of-neural-networks framework (¶¶ [0051]–[0052], Method 400, Step 410: "receiving a group of NNs" / "training a group of NNs," multiple networks available for selection); Senmyo's own dual-unit architecture (¶ [0042] collision and contact determination unit 57; ¶ [0047] risk prediction unit 55) operating concurrently on the same traffic-situation input; general knowledge of ensemble/redundant-model processing, of which Molloy's "diversifying sensor suite" discussion is illustrative. Running a second selected agent on the same inputs concurrently with the first, and combining their outputs, is a predictable extension of Raichelgauz's own framework (reinforced by Senmyo's concurrently-operating dual-unit architecture) and a well-understood ensembling technique in machine learning generally (Official Notice is taken that concurrently operating multiple models on the same input and combining their outputs is a well-known machine-learning technique, MPEP § 2144.03).).
18/822,285 – Claim 19. The non-transitory computer readable medium according to claim 11, further storing instructions executable by the processing circuit for selecting: selecting, concurrently with the selecting of the artificial intelligence agent, another artificial intelligence agent; receiving, at another first machine learning process of the other artificial intelligence agent, the sensed information unit; receiving, at another second machine learning process of the other artificial intelligence agent, the object descriptive information; generating, by the other first machine learning process, another pixel-based path planning output related to another suggested pixel-based path segment of the vehicle; generating, by the other second machine learning process, another object-based path planning output related to another suggested object-based path segment of the vehicle; and generating, by at least in part processing the other pixel-based path planning output in correspondence with the other object-based path planning output, another driving related output with respect to the vehicle.
18/822,285 – Claim 10. Raichelgauz further teaches The method according to claim 9, further comprising generating, based on the driving related output and the other driving related output, a further driving related output ().
18/822,285 – Claim 20. The non-transitory computer readable medium according to claim 19, further storing instructions executable by the processing circuit for selecting generating, based on the driving related output and the other driving related output, a further driving related output.
Any comments considered necessary by applicant must be submitted no later than the payment of the issue fee and, to avoid processing delays, should preferably accompany the issue fee. Such submissions should be clearly labeled “Comments on Statement of Reasons for Allowance.”
Conclusion
PERTINENT PRIOR ART – Patent Literature
The prior-art made of record and considered pertinent to applicant's disclosure.
Bohnmann et al. 2023/0415755 [0002] The present invention relates to a computer-implemented method for generating a virtual vehicle environment for testing highly automated driving functions of a motor vehicle. The invention further relates to a system for generating a virtual vehicle environment for testing highly automated driving functions of a motor vehicle.
U.S. Patent No. 9,740,202 B2, Barton-Sweeney et al., "Fall Back Trajectory Systems for Autonomous Vehicles," assignee Google Inc. (now Google LLC, successor to Waymo LLC), Appl. No. 15/371,595, filed Dec. 7, 2016 (a continuation of Appl. No. 14/991,150, filed Jan. 8, 2016, now U.S. Patent No. 9,551,992), issued Aug. 22, 2017, published as Prior Publication US 2017/0199523 A1 (Jul. 13, 2017). Claim 1 recites: "A method of controlling a vehicle, the method comprising: generating, by a primary computing system, a nominal trajectory from a location for a vehicle in order to achieve a mission goal, the nominal trajectory being generated based on information received by the primary computing system from a perception system of the vehicle configured to detect objects in the vehicle's external environment; generating, by the primary computing system, a fall back trajectory from the location for the vehicle in order to safely stop the vehicle, the fall back trajectory being generated based on the information received by the primary computing system from the perception system of the vehicle, wherein the nominal trajectory and the fall back trajectory are identical between the location and a divergent point and where the nominal trajectory and the fall back trajectory diverge after the divergent point; sending, by the primary computing system, the fall back trajectory to a secondary computing system; receiving, by the secondary computing system, the fall back trajectory; controlling, by the secondary computing system, the vehicle according to the fall back trajectory; waiting, by the secondary computing system, for an updated trajectory from the primary computing system while controlling the vehicle; and when a predetermined threshold period of time from a time when the fall back trajectory was received by the secondary computing system has passed and an updated trajectory has not yet been received by the secondary computing system, continuing to control the vehicle, by the secondary computing system, according to the fall back trajectory in order to safely stop the vehicle." (Claim 11 recites the parallel system claim.) The specification confirms, at column 14 (approx. ll. 17–33): "the computing system 110 may generat[e] two different trajectories, only one of which is actually sent to the computing system 210 to be acted upon. The first trajectory may be a nominal trajectory that enables the vehicle to continue towards achieving the mission goal, while the second trajectory may be a fall back trajectory. For safety, only the second, fallback trajectory and corresponding instructions may be sent to the computing system 210." This reference is pertinent because it confirms that generating more than one candidate driving trajectory from vehicle sensor/perception data, and having a downstream control system select between them, was a known architectural pattern in autonomous-vehicle control well before the effective filing date, lending general plausibility to the motivation-to-combine analysis above. It does not, however, teach or suggest the specific architecture recited in the instant claims: both of its trajectories are generated by the same primary computing system from the output of a single perception system, rather than by two distinct machine learning processes respectively operating on two different categories of input information (a "sensed information unit" versus "object descriptive information," as recited in claim 1); its choice between the two trajectories is triggered by a system-availability/timeout condition (whether an updated trajectory is received by a threshold point) in order to fail safely, not by a comparison, weighting, or confidence-based reconciliation of the two outputs' respective content as recited in claim 1's "conforms to at least one of" limitation or claim 4's confidence-level limitation; and it does not disclose or suggest a "pixel-based" processing pathway as distinct from an "object-based" processing pathway.
PERTINENT PRIOR ART – Non-Patent Literature (NPL)
The NPL prior-art made of record and considered pertinent to applicant's disclosure.
Huang et al., *Multi-modal Sensor Fusion for Auto Driving Perception: A Survey*, arXiv:2202.02703 (originally posted Feb. 6, 2022; revision v3 dated Dec. 16, 2024) (authors Keli Huang (UCLA), Botian Shi, Xiang Li, Siyuan Huang, and Yikang Li (Shanghai AI Laboratory), and Xin Li (East China Normal University; Xiang Li also Beijing Institute of Technology)). As reflected in the paper's own Figure 2, the survey organizes fusion approaches into two major classes — "Strong-Fusion" and "Weak-Fusion" — with Strong-Fusion (Section 4.1) further divided into four minor classes: early-fusion, deep-fusion, late-fusion, and asymmetry-fusion, and Weak-Fusion (Section 4.2) described as a rule-based approach that uses one sensor modality as a supervision signal for a different modality rather than directly fusing their outputs. Late-fusion (object-level fusion) is described as "a kind of ensemble method that utilizes multi-modal information to optimize the final proposal." Per the paper's own scope diagram (Figure 1), the survey is confined to camera-and-LiDAR sensor fusion for perception-stage tasks — 2D and 3D object detection and semantic/instance segmentation. This reference is pertinent because it independently corroborates, from a different (non-Autobrains, non-Subaru) source, the general motivation relied upon above to combine complementary perception-derived pathways in an autonomous-driving system, and because it demonstrates that the general concept of selecting among different fusion stages/architectures (early versus late, among others) for combining two sources of driving-relevant sensor information was a well-developed area of study before the effective filing date. It does not, however, teach or suggest the specific claim limitations at issue: by its own express scope, it is directed only to fusing multiple raw *sensor modalities* (principally LiDAR and camera) at the perception/detection stage to produce a single object-detection or segmentation output, not to combining two already-computed, downstream *path-planning outputs* (a "pixel-based path planning output" and an "object-based path planning output," each itself the product of a separate machine learning process operating on already-processed information) into a "driving related output"; it does not disclose or suggest a "first machine learning process" and "second machine learning process" of an "artificial intelligence agent" in the sense claimed; and it does not disclose the "conforms to at least one of" reconciliation limitation, any confidence-level or safety/comfort-parameter weighting of the kind recited in claims 4, 6, and 7, or the contradiction-identification-and-response limitation of claim 8.
Chitta et al., TransFuser: Imitation with Transformer-Based Sensor Fusion for Autonomous Driving, IEEE TPAMI (2022) / arXiv:2205.15997 (posted May 31, 2022). This is Applicant-Admitted Prior Art insofar as it is cited in Applicant's own specification at ¶ [0158] as an example implementation of the claimed "first machine learning process." It is relevant because it confirms that transformer-based processing of raw sensor (pixel/point-cloud) data to produce driving-relevant outputs was well known before the effective filing date.
Liu, W., Xiang, Z., Fang, H., Huo, K., & Wang, Z. (2023). A Multi-Task Fusion Strategy-Based Decision-Making and Planning Method for Autonomous Driving Vehicles. Sensors, 23(16), 7021. https://doi.org/10.3390/s23167021.
THIS ACTION IS MADE FINAL
Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
THIS ACTION IS MADE FINAL
Applicant’s amendment necessitated new grounds of rejection and FINAL Rejection.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the date of this final action.
Contact Information
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MATTHEW T. SITTNER whose telephone number is (571) 270-7137 and email: matthew.sittner@uspto.gov. The examiner can normally be reached on Monday-Friday, 8:00am - 5:00pm (Mountain Time Zone). Please schedule interview requests via email: matthew.sittner@uspto.gov
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Sarah M. Monfeldt can be reached on (571) 270-1833.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/MATTHEW T SITTNER/
Primary Examiner, Art Unit 3629b
Prosecution Insights