DETAILED ACTION
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 .
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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 May 6th, 2026 has been entered.
Claim Status
Claims 1-6 and 8-21 were pending for examination in the amendments filed for Application No. 18/520,098, filed February 12th, 2026. In the remarks and amendments received on May 6th, 2026, claims 1-6, 8, 11-17, and 19 are amended, no claims are cancelled, and no claims are added. Accordingly, claims 1-6 and 8-21 are pending for examination in the application.
Response to Arguments
Applicant’s arguments, filed May 6th, 2026, with respect to the rejection of claims 1 and 12, specifically the ones relating to prior art “Hu”, have been fully considered but are moot because the arguments do not apply to the new combination of references, facilitated by Applicant’s newly submitted amendments being used in the current rejection.
In regard to the other arguments, however, Applicant asserts that the Office “acknowledges that Sano does not teach or suggest a “grid map”” and “one or more outputs representing, from a top-down perspective environment” (pg. 14 of Applicant’s Remarks). The examiner disagrees, the Office Action stated Sano was “not relied upon” (see pg. 5 of the Final Rejection dated 3/30/26) to teach these limitations at the time of the Final Rejection, the Office Action did not state that Sano “does not teach or suggest”. The examiner asserts that Sano does teach a grid map, see Fig. 14 below, where the grid map is used by an autonomous vehicle to determine travelable regions. Additionally, this grid map is from the top-down perspective. Therefore, Sano does teach and suggest some limitations of “generating, by one or more machine learning models and based at least on the sensor data, one or more outputs representing a grid map of the environment, one or more first portions of the grid map including one or more first values that indicate a nearest boundary of an object and one or more second portions of the grid map including one or more second values that indicate one or more portions of the object other than the nearest boundary; determining, based at least on the grid map, one or more locations within the environment that are associated with the nearest boundary of the object”. See the rejection below for specific teachings and citations of Sano.
PNG
media_image1.png
338
360
media_image1.png
Greyscale
Additionally, Applicant asserts that the Office acknowledges that Sano does not teach or suggest “one or more machine learning models” (pg. 14 of Applicant’s Remarks). The examiner disagrees, the Office Action stated Sano was “not relied upon” (see pg. 5 of the Final Rejection dated 3/30/26) to teach these limitations at the time of the Final Rejection, the Office Action did not state that Sano “does not teach or suggest”.
Applicant also asserts that “it would not have been obvious to combine the teachings of Hu with the teachings of Sano” (pg. 17 of Applicant’s Remarks), because Sano uses algorithms in their object localization technique while Hu uses a trained machine learning model (paraphrased from Applicant’s Remarks on pg. 17). While Hu is not relied upon to reject claim 1 in this Office Action, Sano (using algorithms) is combined with new reference Philbin (as recited below in the Rejection), who uses machine learning models to create occupancy maps for an autonomous vehicle. Although Sano is not relied upon to teach machine learning in their method of determining the likelihood of an object’s nearest position (paras. [0099-0104] of Sano), it would have been obvious to do so in view of Philbin, as machine learning would enable Sano’s method to become more adaptable to an autonomous vehicle’s fast-changing environment.
Additionally, Applicant asserts that “the polar coordinate map of Sano does not teach or suggest the “boundary map” in amended claim 12” (pg. 18 of Applicant’s Remarks). While the examiner does not rely on Sano’s polar coordinate map to teach the “boundary map” of claim 12 in this Action, the examiner disagrees with Applicant’s assertion. Sano’s polar coordinate map, Fig. 12, indicates points “P” that are “closest detection point[s]” (para. [0171]). This map is converted into an orthogonal coordinate map (para. [0179], also Fig. 14) which provides a bird’s eye view of boundaries an autonomous vehicle must avoid. Therefore, even though Sano is not relied upon to teach some limitations (as highlighted in the current rejection below) of “generate, by one or more machine learning models and based at least on the sensor data, a boundary map of the environment surrounding the machine, the boundary map indicating at least one or more locations of one or more nearest boundaries of one or more objects from the machine; generate, based at least on the boundary map, an obstacle map indicating one or more occupied areas associated with the one or more objects within the environment”, Sano does teach and suggest some limitations, specifically, but not limiting, the “boundary map”, as shown above in Fig. 14. See the rejection for Claim 12 below for specific teachings and citations of Sano that are relied upon.
PNG
media_image2.png
336
454
media_image2.png
Greyscale
Applicant also asserts that Sano “does not teach or suggest using a first map to generate a second map” (pg. 18 of Applicant’s Remarks). The examiner disagrees with Applicant’s assertion. Examiner asserts that, for clarity of the record and not currently applied to the rejection, Sano’s polar coordinate map, Fig. 12, is converted into an orthogonal coordinate map (Fig. 14).
Election/Restrictions
Newly amended claims 19-21 are directed to an invention that is independent or distinct from the invention originally claimed for the following reasons: the claims are directed to processors for machine learning models that are trained to detect the one or more nearest boundaries of the one or more objects while refraining from detecting one or more other portions of the one or more objects. These claims were amended with the Request for Continued Examination, filed May 6th, 2026, and introduce a separate and distinct species that was not previously claimed.
Since applicant has received an action on the merits for the originally presented invention, this invention has been constructively elected by original presentation for prosecution on the merits. Accordingly, claims 19-21 are withdrawn from consideration as being directed to a non-elected invention. See 37 CFR 1.142(b) and MPEP § 821.03.
To preserve a right to petition, the reply to this action must distinctly and specifically point out supposed errors in the restriction requirement. Otherwise, the election shall be treated as a final election without traverse. Traversal must be timely. Failure to timely traverse the requirement will result in the loss of right to petition under 37 CFR 1.144. If claims are subsequently added, applicant must indicate which of the subsequently added claims are readable upon the elected invention.
Should applicant traverse on the ground that the inventions are not patentably distinct, applicant should submit evidence or identify such evidence now of record showing the inventions to be obvious variants or clearly admit on the record that this is the case. In either instance, if the examiner finds one of the inventions unpatentable over the prior art, the evidence or admission may be used in a rejection under 35 U.S.C. 103 or pre-AIA 35 U.S.C. 103(a) of the other invention.
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) 1-6, 8, 10, and 12-18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Philbin et al. (US-20210101624-A1) in view of Sano et al. (US-20180189599-A1).
Regarding claim 1, Philbin teaches:
A method comprising:
obtaining sensor data obtained in a perspective view using one or more sensors disposed on a machine navigating within an environment (“the vehicle 102 may receive sensor data from sensor(s) 104 of the vehicle 102… a depth position sensor (e.g., a lidar sensor, a radar sensor, a sonar sensor, a time of flight (ToF) camera, a depth camera, an ultrasonic and/or sonar sensor, and/or other depth-sensing sensor), an image sensor (e.g., a camera),” Para [0024]);
generating, by one or more machine learning models (Fig. 6A, “lidar and/or radar ML model architecture 600”, or Fig. 6B, “image ML model architecture 604”) and based at least on the sensor data (Fig. 6A, “Lidar data 306”, or Fig. 6B, “Image data 304”), one or more outputs representing a grid map of the environment (Fig. 6A, “set of occupancy maps 310(N)” is output, or Fig. 6B, “set of occupancy maps 310(1)”, where the occupancy maps can look like those in Fig. 5A),
PNG
media_image3.png
498
728
media_image3.png
Greyscale
PNG
media_image4.png
235
703
media_image4.png
Greyscale
and one or more second portions of the grid map including one or more second values that indicate one or more portions of the object other than the nearest boundary (Fig. 5A, occupancy grid maps 500, 502, and 504 have “a confidence score (e.g., a likelihood/posterior probability that the portion is occupied/unoccupied),” Para [0059]).
Philbin is not relied upon to teach the following limitations. Sano, however, further teaches:
one or more first portions of the grid map (Sano’s non-polar grid map, Fig. 14, that is derived from polar grid map Fig. 12) including one or more first values that indicate a nearest boundary of an object (“the calculation unit 20E derives, for each angular direction φ, α according to the first distance and the second distance of a detection point P which exists in the closest position to the mobile object 10 in the corresponding angular direction φ. Then, the calculation unit 20E calculates, for each angular direction φ, the existence probability p′ (r) for each distance r from the mobile object 10 along each of the lines extending along the respective angular directions φ,” Para [0126]);
PNG
media_image5.png
426
443
media_image5.png
Greyscale
determining, based at least on the grid map, one or more locations within the environment that are associated with the nearest boundary of the object (Fig. 14, the dark grey regions indicate a boundary of an object); and
causing, based at least on the one or more locations associated with the nearest boundary of the object, the machine to perform one or more operations (“To automatically drive the mobile object 10, the power control unit 10G controls the power unit 10H on the basis of information that can be acquired from the external sensor 10B and the internal sensor 10C, and existence probability information derived from processing,” Para [0031]).
Sano is considered to be analogous to the claimed invention because they are both in the field of determining drivable regions for an autonomous vehicle. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated the teachings of Sano into Philbin for the benefit of knowing precisely where an obstacle is located, and in particular, the part of the object that is most likely to collide with the vehicle.
Regarding claim 2, the rejection of claim 1 is incorporated herein. Philbin in view of Sano teaches the method of claim 1, and Sano further teaches:
wherein the determining that the one or more locations are associated with of the nearest boundary of the object comprises:
processing first data that represents the grid map (Fig. 12; and “the calculation unit 20E converts the polar coordinate map M1 into an orthogonal coordinate space,” Para [0174])
generating, closest region B′, among the regions B′ rectangularly divided in the polar coordinate space, as the existence probability of an obstacle in the corresponding partitioned region B,” Para [0176]).
Sano is not relied upon to teach “one or more machine learning models”. However, Philbin teaches a machine learning model that “appl[ies] a threshold to the data structure to determine binary indications of whether a portion [of the grid map] is occupied or unoccupied,” Para [0077]. See Para [0079] for more information on how the machine learning model processes the grid map data using thresholding.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated the teachings of Sano into Philbin for the benefit of knowing precisely where an obstacle is located, and in particular, the part of the object that is most likely to collide with the vehicle.
Regarding claim 3, the rejection of claim 1 is incorporated herein. Philbin in view of Sano teaches the method of claim 1, and Sano further teaches:
wherein the determining that the one or more locations are associated with the nearest boundary of the object (Fig. 9, “B’8”) comprises:
PNG
media_image6.png
310
506
media_image6.png
Greyscale
determining that the one or more first values include a higher value as compared to the one or more second values (Fig. 9, the existence probability at point P, or B’8, is higher than at B’9); and
determining, based at least on the one or more first values including the higher value, that the one or more first portions of the grid map correspond to the one or more locations associated with the nearest boundary of the object (“The detection point P is the one which exists in the closest position to the mobile object 10,” Para [0159]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated the teachings of Sano into Philbin for the benefit of knowing precisely where an obstacle is located, and in particular, the part of the object that is most likely to collide with the vehicle.
Regarding claim 4, the rejection of claim 1 is incorporated herein. Philbin in view of Sano teaches the method of claim 1, and Philbin further teaches:
the one or more machine learning models (Fig. 6A, “lidar and/or radar ML model architecture 600”, or Fig. 6B, “image ML model architecture 604”) are trained to generate
Philbin is not relied upon to teach the following limitations. Sano, however, further teaches:
the one or more first values indicating the nearest boundary (“the calculation unit 20E derives, for each angular direction φ, α according to the first distance and the second distance of a detection point P which exists in the closest position to the mobile object 10 in the corresponding angular direction φ. Then, the calculation unit 20E calculates, for each angular direction φ, the existence probability p′ (r) for each distance r from the mobile object 10 along each of the lines extending along the respective angular directions φ,” Para [0126]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated the teachings of Sano into Philbin for the benefit of knowing precisely where an obstacle is located, and in particular, the part of the object that is most likely to collide with the vehicle.
Regarding claim 5, the rejection of claim 1 is incorporated herein. Philbin in view of Sano teaches the method of claim 1, and Sano further teaches:
one or more third portions of the grid map (Fig. 14) include one or more third values that indicate one or more portions of the environment that are between the machine and the nearest boundary of the object (Fig. 14, the white/unshaded regions represent the region between the machine and the nearest boundary); and
the one or more first values (Fig. 9, B’8) are greater than the one or more second values (Fig. 9, B’9) and the one or more third values (Fig. 14, B’1 through B’6; the existence values of B’1 through B’6 and B’9 are less than the existence value of B’8).
PNG
media_image7.png
366
590
media_image7.png
Greyscale
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated the teachings of Sano into Philbin for the benefit of knowing precisely where an obstacle is not located, in addition to knowing where it is located, to minimize false positive readings from the vehicle sensors.
Regarding claim 6, the rejection of claim 1 is incorporated herein. Philbin in view of Sano teaches the method of claim 1, and Sano further teaches:
generating an obstacle map representing the environment (Fig. 14), the obstacle map indicating at least the one or more locations associated with the nearest boundary of the object (Fig. 14, the regions labeled with “occupied region” indicate the regions identified as the nearest boundaries of the object),
wherein the causing the machine to perform the one or more operations is based at least on the obstacle map (“To automatically drive the mobile object 10, the power control unit 10G controls the power unit 10H on the basis of information that can be acquired from the external sensor 10B and the internal sensor 10C, and existence probability information derived from processing,” Para [0031]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated the teachings of Sano into Philbin for the benefit of knowing precisely where an obstacle is located, and in particular, the part of the object that is most likely to collide with the vehicle.
Regarding claim 8, the rejection of claim 1 is incorporated herein. Philbin in view of Sano teaches the method of claim 1, and Philbin further teaches:
wherein the generating the one or more outputs comprises:
generating, using the one or more machine learning models (Fig. 4, step 404(1)) and based at least on a first portion of the sensor data corresponding to a first sensor modality (Fig. 4, step 402(1)), one or more second outputs (Fig. 4, step 406(1));
generating, using the one or more machine learning models (Fig. 4, step 404(N)) and based at least on a second portion of the sensor data corresponding to a second sensor modality (Fig. 4, step 402(N)), one or more third outputs (Fig. 4, step 406(N)); and
generating, by the one or more machine learning models (“setting a threshold may comprise training the ML models based at least in part on a training data set, which may include sensor data recorded from live operation of a vehicle and/or simulated sensor data, top-down representations generated by a previously-trained ML model, aerial footage, and/or labels generated by humans and/or another ML model;” Para [0079]) and based at least on the one or more second outputs and the one or more third outputs, the one or more outputs (“example occupancy map 506 that may result from setting a threshold,” Para [0080], where this map was created by aggregating the other maps, as shown in Fig. 4).
PNG
media_image8.png
703
478
media_image8.png
Greyscale
Regarding claim 10, the rejection of claim 1 is incorporated herein. Philbin in view of Sano teaches the method of claim 1, and Philbin further teaches:
wherein the sensor data comprises one or more of:
image data generated using the machine (“the vehicle 102 may receive sensor data from… an image sensor (e.g., a camera),” Para [0024);
LiDAR data generated using the machine (“a lidar sensor,” Para [0024]);
RADAR data generated using the machine (“a radar sensor,” Para [0024]);
or ultrasonic data generated using the machine (“an ultrasonic and/or sonar sensor,” Para [0024]).
Regarding claim 12, Philbin teaches:
one or more processors to:
obtain sensor data obtained using a machine navigating within an environment (“the vehicle 102 may receive sensor data from sensor(s) 104 of the vehicle 102… a depth position sensor (e.g., a lidar sensor, a radar sensor, a sonar sensor, a time of flight (ToF) camera, a depth camera, an ultrasonic and/or sonar sensor, and/or other depth-sensing sensor), an image sensor (e.g., a camera),” Para [0024]);
generate, by one or more machine learning models (Fig. 6A, “lidar and/or radar ML model architecture 600”, or Fig. 6B, “image ML model architecture 604”) and based at least on the sensor data (Fig. 6A, “Lidar data 306”, or Fig. 6B, “Image data 304”), a boundary map of the environment surrounding the machine (Fig. 5A, and where “The occupancy maps 500-510 each comprise multiple portions depicted as squares and each portion is associated with a respective portion of an environment surrounding the vehicle,” Para [0073]),
generate, based at least on the boundary map, an obstacle map (Figs. 5B-5D) indicating one or more occupied areas associated with the one or more objects within the environment (“the set of occupancy maps 314 that result from aggregation may comprise a field of confidence scores and applying the threshold to the confidence scores may result in binary indications of occupied and unoccupied portions of the environment,” Para [0066]; in other words, the occupancy maps are combined and then binarized); and
PNG
media_image9.png
275
797
media_image9.png
Greyscale
cause, based at least on the obstacle map, the machine to perform one or more operations (“example process 400 may comprise controlling an autonomous vehicle based at least in part on the data structure,” Para [0083]).
Philbin is not relied upon to teach the following limitations. Sano, however, further teaches:
the boundary map indicating at least one or more locations of one or more nearest boundaries of one or more objects from the machine (Fig. 14, the regions labeled with “occupied region” indicate the regions identified as the nearest boundaries of the object).
Sano is considered to be analogous to the claimed invention because they are both in the field of determining drivable regions for an autonomous vehicle. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated the teachings of Sano into Philbin for the benefit of knowing precisely where an obstacle is located, and in particular, the part of the object that is most likely to collide with the vehicle.
Regarding claim 13, the rejection of claim 12 is incorporated herein. Philbin in view of Sano teaches the system of claim 12, and Philbin further teaches: wherein the generation of the boundary map comprises:
determining, using the one or more machine learning models (Fig. 6A, “lidar and/or radar ML model architecture 600”, or Fig. 6B, “image ML model architecture 604”) and based at least on the sensor data (Fig. 6A, “Lidar data 306”, or Fig. 6B, “Image data 304”), one or more values associated with the one or more locations within the environment (Fig. 6A, “set of occupancy maps 310(N)” is output, or Fig. 6B, “set of occupancy maps 310(1)”, where the occupancy maps can look like those in Fig. 5A).
Philbin is not relied upon to teach all the following limitations. Sano, however, further teaches:
determining, using the one or more machine learning models (Philbin, Fig. 6A, “lidar and/or radar ML model architecture 600”, or Fig. 6B, “image ML model architecture 604”) and based at least on the one or more values (Sano, Fig. 9, “existence probability”), that the one or more locations are associated with the one or more nearest boundaries of the one or more objects from the machine (Sano, “the calculation unit 20E derives, for each angular direction φ, α according to the first distance and the second distance of a detection point P which exists in the closest position to the mobile object 10 in the corresponding angular direction φ. Then, the calculation unit 20E calculates, for each angular direction φ, the existence probability p′ (r) for each distance r from the mobile object 10 along each of the lines extending along the respective angular directions φ,” Para [0126]); and
generating, by the one or more machine learning models, the boundary map indicating at least the one or more locations of the one or more nearest boundaries of the one or more objects from the machine (Sano, Fig. 14, a boundary map indicating nearest boundaries).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated the teachings of Sano into Philbin for the benefit of knowing precisely where an obstacle is located, and in particular, the part of the object that is most likely to collide with the vehicle.
Regarding claim 14, the rejection of claim 13 is incorporated herein. Philbin in view of Sano teaches the system of claim 13, and Sano further teaches:
wherein the determining that the one or more locations are associated with the one or more nearest boundaries of the one or more objects from the machine comprises:
determining that the one or more locations are associated with one or more maximum values of the one or more values (Fig. 9, location “B’8” or “P” is designated as having the highest existence probability);
and determining, based at least on the one or more locations being associated with the one or more maximum values, that the one or more locations are associated with the one or more nearest boundaries of the one or more objects from the machine (“the calculation unit 20E subsequently calculates the first distance of the detection point P closest to the mobile object 10 (the detection point P at the distance r.sub.0) in the corresponding angular direction φ to be processed,” Para [0131]).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated the teachings of Sano into Philbin for the benefit of knowing precisely where an obstacle is located, and in particular, the part of the object that is most likely to collide with the vehicle.
Regarding claim 15, the rejection of claim 12 is incorporated herein. Philbin in view of Sano teaches the system of claim 12, and Philbin further teaches:
wherein the one or more occupied areas indicated by the obstacle map (Figs. 5B-5D, the shaded/grey areas) include one or more second locations that the boundary map indicates as not being occupied by the one or more objects (Figs. 5B-5D, the non-shaded/white areas).
Regarding claim 16, the rejection of claim 12 is incorporated herein. Philbin in view of Sano teaches the system of claim 12, and Sano further teaches:
wherein the boundary map (Fig. 14) indicates at least:
one or more first values associated with the one or more locations within the environment (Fig. 9, location “B’8” or “P” with the highest existence probability);
one or more second values associated with one or more second locations within the environment that are closer to the machine (Fig. 9, the machine location is when r = 0) than the one or more first locations (Fig. 9, location “B’7” with an associated existence probability);
and one or more third values associated with one or more third locations within the environment that are farther from the machine than the one or more first locations (Fig. 9, locations “B’9 – B’11”, each with an associated existence probability);
and the one or more first values are greater than the one or more second values and the one or more third values (Fig. 9, existence probability at “B’8” is higher than at “B’7” and “B’9”).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated the teachings of Sano into Philbin for the benefit of knowing precisely where an obstacle is located, and in particular, the part of the object that is most likely to collide with the vehicle.
Regarding claim 17, the rejection of claim 12 is incorporated herein. Philbin in view of Sano teaches the system of claim 12, and Philbin further teaches:
wherein the generation of the obstacle map (Figs. 5B-5D) comprises:
determining, based at least on the boundary map (Fig. 5A), the one or more locations ofcomprise aggregating occupancy maps generated by different pipelines and associated with a same time into a single occupancy map,” Para [0075]);
determining,
generating the obstacle map to indicate at least the one or more occupied areas associated with the one or more objects within the environment (Figs. 5B-5D, non-limiting examples of the resulting obstacle maps based on the averaging/voting techniques).
Philbin is not relied upon to teach the following limitations. Sano, however, further teaches:
determining, based at least on the boundary map, the one or more locations of the one or more nearest boundaries of the one or more objects from the machine (Sano, Fig. 14, a boundary map indicating nearest boundaries).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have incorporated the teachings of Sano into Philbin for the benefit of knowing precisely where an obstacle is located, and in particular, the part of the object that is most likely to collide with the vehicle.
Regarding claim 18, the rejection of claim 12 is incorporated herein. Philbin in view of Sano teaches the method of claim 12, and Philbin further teaches:
wherein the system is comprised in at least one of:
a control system for an autonomous or semi-autonomous machine (“the vehicle 102 may be an autonomous vehicle configured to operate according to a Level 5 classification issued by the U.S. National Highway Traffic Safety Administration,” Para [0023]);
a perception system for an autonomous or semi-autonomous machine;
a system for performing simulation operations;
a system for performing digital twin operations;
a system for performing light transport simulation;
a system for performing collaborative content creation for 3D assets;
a system for performing deep learning operations;
a system implemented using an edge device;
a system implemented using a robot;
a system for performing conversational AI operations;
a system implementing one or more large language models (LLMs);
a system for performing generative AI operations;
a system for generating synthetic data;
a system incorporating one or more virtual machines (VMs);
a system implemented at least partially in a data center;
or a system implemented at least partially using cloud computing resources.
Examiner’s Note: Claim 18 as recited is treated as a “field of use” or “intended use” limitation and therefore carries no patentable weight although it has been examined in view of Philbin in view of Sano above. The system as recited has been examined as evidenced in claim 12 above. With respect to the enumerated environments that said system is “comprised in”, the specification as disclosed merely mentions these environments as preferred intended use environments without specific details that warrant said system comprised in these environments resulted in a novel and non-obvious structural change to the system. Reference to MPEP 2112.01 is also made for applicant’s attention.
Claim(s) 9 is/are rejected under 35 U.S.C. 103 as being unpatentable over Philbin et al. (US-20210101624-A1) in view of Sano et al. (US-20180189599-A1) as applied to claim 1 above, and further in view of You et al. (US-20180164832-A1).
Regarding claim 9, the rejection of claim 1 is incorporated herein. Philbin in view of Sano teaches the method of claim 1, but are not completely relied upon to teach the following limitations. You, however, further teaches generating, using the one or more machine learning (Philbin, Fig. 6A, “lidar and/or radar ML model architecture 600”, or Fig. 6B, “image ML model architecture 604”), at least one of:
a height map indicating one or more heights associated with the one or more locations (You, “generating a height map of the front image by transforming the generated depth map,” Para [0007]); or one or more uncertainty values associated with the one or more locations.
You is considered to be analogous to the claimed invention because they are both in the same field of avoiding obstacles in a vehicle’s field of view. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to have incorporated the teachings of You into Philbin and Sano for the benefit of more accurately determining if an obstacle exists in a field of view.
Claim(s) 11 is/are rejected under 35 U.S.C. 103 as being unpatentable over Philbin et al. (US-20210101624-A1) in view of Sano et al. (US-20180189599-A1) as applied to claim 1 above, and further in view of Lee et al. (US-20200294310-A1) and Kentley-Klay (US-20200064842-A1).
Regarding claim 11, the rejection of claim 1 is incorporated herein. Philbin in view of Sano teach the method of claim 1, but are not relied upon to teach the following limitations. Lee, however, further teaches wherein the one or more machine learning models are trained using at least:
input data (object detector is trained using “training images,” Para [0057]) that includes at least
and ground truth data (“ground-truth data for a parking space”) representing at least one or more ground truth values indicating whether one or more
Lee is not relied upon to teach the following limitations. Kentley-Klay, however, further teaches “second” sensor data, machines, environments, values, locations, boundaries, and objects (“In some examples, the first vehicle 104(1) may migrate testing and/or training to second vehicle 104(2) by transmitting… an updated model 408 (which may include a trained target ML model) and/or an experimental model 410 to the second vehicle 104(2),” Para [0066]).
Lee is considered to be analogous to the claimed invention because they are both in the field of object detection in autonomous vehicles. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to have incorporated the teachings of Lee into Philbin and Sano for the benefit of up-to-date training information for more reliable object avoidance.
Kentley-Klay is considered to be analogous to the claimed invention because they are both in the field of vehicle object avoidance. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date to have incorporated the teachings of Kentley-Klay into Philbin, Sano, and Lee for the benefit of up-to-date training information for more reliable object avoidance.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Philbin et al. (US-20210103285-A1) teaches a method for controlling an autonomous vehicle and avoiding collisions based on existence probability values of other objects around the vehicle.
Unnikrishnan et al. (US-20220237402-A1) teaches a method for object occupancy tracking using boundary information of objects surrounding an autonomous vehicle.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to RACHEL A OMETZ whose telephone number is (571)272-2535. The examiner can normally be reached 6:45am-4:00pm ET Monday-Thursday, 6:45am-1:00pm ET every other 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, Vu Le can be reached at 571-272-7332. 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.
/Rachel Anne Ometz/Examiner, Art Unit 2668 6/15/26
Rachel.ometz@uspto.gov
/VU LE/Supervisory Patent Examiner, Art Unit 2668
Prosecution Insights