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 .
DETAILED ACTION
Applicant filed an amendment on 6/25/26. Claims 1-20 were pending in the application. Claims 1, 10, and 19, and dependent claims 3, 7-9, 12, 16-18 have been amended, and claims 2, 11, and 20 have been canceled. Thus claims 1,3-10,12-19 are pending. After careful consideration of applicant arguments and amendments, the examiner finds them to be moot and/or non persuasive in view of new grounds of rejection. This action is a Final Rejection
35 USC § 101
Claims 1,3-10,12-19 overcome 35 USC 101 because they are based on a technical improvement in machine learning, in view of amendment.
Claim Rejections - 35 USC § 103
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, 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.
Claim(s) 1, 3-10, 12-19 is/are rejected under 35 U.S.C. 103 as being unpatentable over
US Patent Publication to Xu , 20220326023, in view of Wang 12293543
As per claim 1, Xu discloses;
receive a first output generated by a first machine learning (ML) model, Xu(0057, machine learning models) the first output of the first ML model comprises a depth map including a depth value for each pixel of the depth map; Wang(col. 4 lines 20-25)
receive one or more confidence levels of the first output generated by the first ML
model, Xu(0042) the one or more confidence levels of the first output generated by the first ML model comprises an error map indicative of a respective error for each pixel of the depth map; Wang(col. 4 lines 20-25) process, at a second ML model
including an object detection model, Xu(0023) the first output of the first ML model
and the one or more confidence levels of the first output of the first ML model as an input of the second ML model; Xu(0042) and generate a second output of the second ML model Xu(0042)
comprising a bounding box Xu(0079, bounding region)
based on the processing of the first output of the first ML model and the one or more confidence levels of the first output of the first ML model,
Xu(0034-35)
wherein the second output of the second ML model is based
on one or more pixels Xu(0038 each model, implies two, 0042, based on pixels)
of the depth map Wang(col. 4 lines 20-25)
that have the respective error being below an error threshold; Xu(0046)
provide the second output comprising the bounding box to a perception stack of a vehicle compute system, Wang(col. 20 lines 15-20, perception component, and bounding box col. 19, lines 1-10)
the vehicle compute system being operably coupled to a vehicle, Wang(col. 20 lines 10-20)
wherein the perception stack is configured to perceive an environment proximal to the vehicle based,
Wang(col. 20 lines 10-20)
at least in part, on the second output; and operate the vehicle, at least in part via the vehicle compute system, in the perceived environment. Wang(col. 20 lines 10-20, “in part” could be a little bit, that’s how an autonomous vehicle works)
It would therefore have been obvious to one of ordinary skill in the art before the effective filing date of the invention to combine the autonomous driving of Xu with the various technical elements of such driving of Wang for the motivation of better “controlling the autonomous vehicle” (col. 1 lines 6-11)
Claims 19 and 1 are similar to claim 10.
As per claim 3, Xu discloses;
The system of claim 1, wherein processing the first output of the first ML model and the one or more confidence levels of the first output of the first ML model comprises: processing at least a portion of the first output of the first ML model based on the one or more confidence levels of the first output of the first ML model, wherein a respective confidence level from the one or more confidence levels corresponding to at least the portion of the first output of the first ML model exceeds a confidence the respective error is below an error threshold.
Xu(0028 confidence values, 0033, threshold)
Claim 12 is similar to claim 3
As per claim 4, Xu discloses; The system of claim 1, wherein the one or more processors are configured to: train the second ML model with the second output of the second ML model, wherein training of the second ML model is independent of training of the first ML model. Xu(0097-98 multiple models …. )
Claim 13 is similar to claim 4.
As per claim 5 Xu discloses;
the system of claim 1, wherein the first ML model is trained with the first output of the first ML model and the one or more confidence levels of the first output, wherein training of the first ML model is independent of training of the second ML model.
Xu(0106)
Claim 14 is similar to claim 5.
As per claim 6, Xu discloses;
The system of claim 1, wherein the one or more confidence levels of the first output generated by the first ML model comprises a margin of error of the first output generated by the first ML model.
Xu(0108 and 0113)
Claim 15 is similar to claim 6.
As per claim 7 Xu discloses; The system of claim 1, wherein the first output of the first ML model includes a depth map comprising a depth value for each pixel of the depth map and the one or more confidence levels of the first output generated by the first ML model includes an error map indicative of a respective error for each pixel of the depth map, which comprises a probability of an error in the depth value of each pixel. Xu(0108, 0115, pixel confidence, 0019)
Claim 16 is similar to claim 7.
As per claim 8 Xu discloses;
The system of claim 7, wherein the second output of the second ML model is based on object detection model identifies the one or more pixels of the depth map that have the respective error that where a probability of an error is below an error threshold. (0046, error threshold)
Claim 17 is similar to claim 8.
As per claim 9 Xu discloses;
The system of claim 1, wherein the first ML model includes at least one of an object detection model and an object classification model.
Xu(0028, 0079 one of could be one, but definitely classification and detection)
Here in regards to depth map model Wang teaches what Xu fails to teach, (col. 4 lines 20-25) and the motivation would be similar to that provided for the independent claims.
Claim 18 is similar to claim 9.
Response to Arguments
Applicant filed an amendment on 6/25/26. Claims 1-20 were pending in the application. Claims 1, 10, and 19, and dependent claims 3, 7-9, 12, 16-18 have been amended, and claims 2, 11, and 20 have been canceled. Thus claims 1,3-10,12-19 are pending. After careful consideration of applicant arguments and amendments, the examiner finds them to be moot and/or non persuasive in view of new grounds of rejection. This action is a Final Rejection.
REJECTIONS UNDER 35 U.S.C. § 101- moot
REJECTIONS UNDER 35 U.S.C. 103
Claims 1-20 stand rejected under 35 U.S.C. § 103 as being unpatentable over Xu et al. (US20220326023; hereinafter "Xu").
In response to the rejections under 35 U.S.C. § 103, claims 1, 10, and 19 have
been amended to recite specific technical features that are neither taught nor suggested by Xu. The Office Action cited Xu paragraphs [0042] and [0056] for the broad limitations of
receiving a first ML-model output, receiving confidence levels, processing those items at another ML model, and generating a second output. However, the amended independent claims now require a perception architecture in which the first ML-model output comprises a depth map including a depth value for each pixel, the confidence information comprises an error map indicative of a respective error for each pixel of the depth map, and the second ML model includes an object detection model that generates a bounding box based on one or more pixels of the depth map having the respective error below an error threshold. The amended claims further require providing that bounding box to a perception stack of a vehicle control system and actuating vehicle mechanical systems to navigate the vehicle in the perceived environment.
Xu does not teach or suggest this claimed combination. Xu paragraph [0042]
describes generating estimated map data from top-down representations using a machine learned model having channels for road-policy features such as off-road, on-road, solid lane-line, and dashed lane-line channels. Xu paragraph [0056] generally describes various machine-learned models and training using policy-map ground truth. Those passages do not disclose a depth map having a depth value for each pixel, an error map indicating a respective error for each pixel of the depth map, or using such a depth-map error map at an object detection model to generate a bounding box from pixels whose error is below an error threshold.
The additional Xu passages cited for the dependent claims likewise do not cure this deficiency. Xu paragraph [0028] describes estimated map data and per-pixel confidence data associated with segmentation or policy-map information, and Xu paragraph [0033] describes comparing estimated map data with stored map data and a proposed vehicle trajectory to assess consistency. Xu paragraph [0079] separately describes object detection, segmentation, or classification from sensor data and associating a bounding region or confidence score with an identified object. Xu paragraph [0115] refers to per-pixel confidence value data associated with a location for an overlap score. These teachings, individually or collectively, do not amount to the
claimed ordered arrangement in which a depth map model generates both a depth map and a per-pixel error map for the depth map, and an object detection model generates a bounding box based on depth-map pixels whose respective errors are below an error threshold.
The distinction is not merely one of terminology. In the amended claims, the error map qualifies the reliability of depth values on a per-pixel basis, and the object detection model uses that depth-value error information to determine which depth-map pixels are used to generate the bounding box. Xu's cited teachings concern estimated map or policy-map confidence, consistency checking against stored map data, and general object detection or bounding-region association. The Office Action has not identified, and Xu does not disclose the claimed depth-map/error-map/object-detection chain or the claimed error-threshold-based generation of a bounding box for downstream vehicle perception and navigation.
Respectfully, the Examiner also fails to provide any "suggestion or incentive" that would motivate a skilled artisan to modify Xu disclose or suggest the missing elements in the way claimed in the present application.
Accordingly, Xu fails to teach or suggest at least the above-recited limitations of amended independent claims 1, 10, and 19. Applicant respectfully submits that the amendments overcome the rejection under
35 U.S.C. § 103, and respectfully requests withdrawal of the § 103 rejection of claims 1, 10, and 19.
The dependent claims are argued by virtue of dependency only.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Handling Occlusions in Automated Driving Using a Multiaccess Edge Computing Server-Based Environment Model From Infrastructure Sensors, (Year: 2021)
Uncertainty-Aware Driver Trajectory Prediction at Urban Intersections, IEEE (Year: 2019)
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 nonprovisional extension fee (37 CFR 1.17(a)) 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.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BRUCE I EBERSMAN whose telephone number is (571)270-3442. The examiner can normally be reached 8:00 am - 5:00 pm Monday-Friday.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Michael W Anderson can be reached at 571-270-0508. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/BRUCE I EBERSMAN/Primary Examiner, Art Unit 3693
Prosecution Insights