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 .
Notice to Applicants
This communication is in response to the amendment filed on 03/17/2026.
Claim 1-20 are pending.
Response to Arguments
Applicant’s arguments with respect to claim(s) 1-20 have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Drawings
The drawings are objected to because the arrow extending from block 630 in FIG. 6 should connect to block 610, rather than block 615 (Refer to para. [0078] in the original spec. filed on 1/19/2024 and the amended spec. filed on 3/17/2026). Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. The figure or figure number of an amended drawing should not be labeled as “amended.” If a drawing figure is to be canceled, the appropriate figure must be removed from the replacement sheet, and where necessary, the remaining figures must be renumbered and appropriate changes made to the brief description of the several views of the drawings for consistency. Additional replacement sheets may be necessary to show the renumbering of the remaining figures. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(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.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 1-20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 1 recites the limitation "the pose" in line 13. There is insufficient antecedent basis for this limitation in the claim. Clarification/explanation is required.
Claim 1 recites the limitation "an object" in line 13. It is unclear if “an object” is referring back to “object” in line 6, 10 and 12. Clarification/explanation is required.
Claim 1 recites the limitation "the first and second sensor scans" in line 15. It is unclear if “the first and second sensor scans” are referring back to “the first and second scans” in line 7-8 or something else. Clarification/explanation is required.
Claim 1 recites the limitation "receive a parameter from the memory of the computer, the parameter being determined from a training process” in line 9-10. The system must use the “modified amodal representation” to interpolate the parameter which represents the first validity time (Refer to para. [0031] & [0032] in the spec. and claim 10 & 11). However, the system requires the parameter to modify the amodal representation in the first place. Because calculating the parameter requires the modified representation, but generating the modified representation requires the parameter, the logic is circular.
Also, the timing relationship between “determine a first velocity-compensated position of an object represented by a third set of points at a first validity time” and “receive a parameter from the memory of the computer, the parameter being determined from a training process to modify an amodal representation of the object” is unclear. The specification indicates the parameter represents the first validity time at which the modified amodal representation of the object (Refer to para. [0031] & [0041] in the spec.), but in the claim 1, the parameter is received after the determination of the first velocity-compensated position has already been performed. It is unclear how the first validity time is available for use in determining the first velocity-compensated position before receiving the parameter representing a first validity time.
If the parameter is received or determined after the first execution, which corresponds to the second execution of iteratively refining compensated object positions (Refer to para. [0020] and [0072] in the spec. and FIG. 6), its timing should reflect the claim language. Note that the first execution is equivalent to determining a first velocity-compensated position of an object represented by a third set of points at a first validity time.
Claim 1 recites the limitation “modify an amodal representation of the object” in line 10, but defines the amodal representation as “an aggregation of one or more historical sets of points, obtained previous to the first and second sensor scans” in line 14-15. So, the limitation “modify an amodal representation of the object” appears inconsistent with the rest of claims and the specification. The specification supports that the system modifies the first amodalized representation of the moving object at a first interpolated time between first and second scans (which is related to “a first velocity-compensated position of an object”) which may include bent, bowed, or fattened lines (Refer to para. [0072] - [0079] in the spec. and FIG. 6). The specification does not support the limitation of “modify an amodal representation of the object” which “obtained utilizing an aggregation of one or more historical sets of points, obtained previous to the first and second sensor scans.” The examiner interprets the limitation of “modify an amodal representation of the object” as of “modify a first amodalized representation of the object.” (Refer to para. [0020] and [0072] in the spec. and FIG. 6)
Claim 1 recites the limitation “the parameter being determined from a training process to modify an amodal representation of the object, the modified amodal representation being determined from a difference between a second velocity-compensated position of the object and the amodal representation of the object” in line 9-13. It is unclear how the amodal representation is modified. The claim indicates the amodal representation is modified by “the parameter” and “a difference”, but it is unclear how these two concepts interact. The claim fails to clearly define the relationship between the parameter and the difference.
Claim 1 recites the limitation “a second velocity-compensated position of the object” in line 11-12. Even if the 112 (a) rejection is withdrawn by explaining how to enable to obtain “a second velocity-compensated position of the object” (Refer to the remarks filed on 3/17/2026), but the claim skips the essential step of calculating the second velocity-compensated position.
Claim 1 recites the limitation “a difference between a second velocity-compensated position of the object and the amodal representation of the object”. The claim does not define how the second velocity-compensated position and the amodal representation of the object are quantitatively compared.
Claim 1 recites the limitation “the modified amodal representation being determined from a difference between a second velocity-compensated position of the object and the amodal representation of the object”. However, the sequence of the steps is unclear and contradicts the specification. First, the claim recites that the modified amodal representation is determined from a difference utilizing the second velocity-compensated position of the object. It requires the second position to exist prior to the modification. However, it contradicts the specification and drawings, which explicitly states that the system performs the modifying step first (Refer to para. [0077] in the spec. and block 630 in FIG. 6), and after the modifying step, computing a second amodalized object position (Refer to para. [0078] in the spec. and returned block 610 in FIG. 6).
Claim 2 recites the limitation "the parameter is generated from an iterative adjustment of the amodal representation of the object, the iterative adjustment of the amodal representation being based on a difference between the second velocity-compensated position of the object and the modified amodal representation of the object being greater than a threshold value". Claim 2 contradicts regarding the determination of the parameter when read in light of claim 1. Claim 1 recites that the parameter is determined from the training process, but claim 2 recites that the same parameter is generated from the iterative adjustment. It is unclear if the training process and the iterative adjustment are intended to be the same process or not.
Also, the claim 1 uses the parameter to generate the modified amodal representation, while claim 2 calculates the same parameter based on the modified amodal representation. It is chronologically impossible.
Claim 3 recites the limitation "the parameter is determined from the iterative adjustment of the amodal representation of the object terminating in response to the second velocity-compensated position of the object and the modified amodal representation of the object being less than the threshold value". It is impossible to compare two distinct spatial representations directly to a threshold value without first defining a metric such as a difference between them.
With respect to claim 12, arguments analogous to those presented for claim 1, are applicable.
With respect to claim 13, arguments analogous to those presented for claim 2, are applicable.
With respect to claim 14, arguments analogous to those presented for claim 3, are applicable.
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 (i.e., changing from AIA to pre-AIA ) 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, 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 1, 5-9, 12 and 16-19 are rejected under 35 U.S.C. 103 as being unpatentable over Blaes et al. (U.S Patent No. 12,221,115) (hereafter, "Blaes") in view of Purdy (U.S Patent No. 12,333,823) and further in view of PARK et al. (KR 20230102363A) (hereafter, "PARK").
Regarding claim 1, Blaes teaches a system, comprising: a computer having a processor and a memory, the memory including instructions executable by the processor to ([column 15, line 26-29] The vehicle computing device 604 may include one or more processor(s) 612 and computer readable media 614 communicatively coupled with the one or more processor(s) 612): generate first and second sets of points from first and second scans obtained from a lidar sensor or from a radar sensor ([column 20, line 36-37] the radar sensor 702 may capture the radar data 708, e.g., via one or more radar scans; FIG. 1; [column 8, line 53-59] For each interval of radar data 102(A)-(O) captured by the one or more sensors ... register the radar-based point cloud stream using the platform's state in the global reference frame to generate a global reference representation 106 of the radar data 102(A)-(O); [column 21, line 5-14] The illustration of FIG. 7 comprises a visualization 714 of radar data 718. More specifically, the visualization 714 includes a plurality of representations of radar points illustrated as circular points, which may be representative of returns caused by objects, surfaces, or other reflective items, in an environment of the vehicle 704 … each of the points depicted as part of the visualization 714 may correspond to a position of a surface associated with one of the vehicle 710 that reflected a radio wave emitted by the radar device and was detected at the radar sensor 702); determine a first velocity-compensated position of an object represented by a third set of points at a first validity time that is between respective times of the first and second scans ([column 24, line 23-35] As illustrated in FIG. 8, a first probability map 802 illustrates probabilities of locations of objects at a first time. The first probability map 802 may be based on or include the radar data from the radar sensor. In the first probability map 802, clusters of points may be identified as related to one another to identify rigid bodies, as described herein. The second frame 804 illustrates a velocity output of the machine learning model based on the provided radar data. The velocity output from the machine learning model and the first predicted location from the first probability map 802 may be used to determine a predicted location 806. The predicted location is for a known time in the future of the first time. The predicted location 806 may then be compared against a second probability map 808, from a second time from a successive radar scan; [column 24, line 48-52] the model output at the second frame 804 may be determined using additional sensor data, such as additional range sensor data and/or image data. Such additional data may provide further positional and/or velocity information that may be used to refine the model) … determine a pose of the object represented by the third set of points based on the parameter ([column 2, line 30-34] the perception system may perform updates on object state data, such as position, orientation … using a discretized point cloud representation of captured radar and machine learned algorithms (such as deep neural networks); [column 5, line 17-20] a machine learning model (e.g., deep neural network or convolutional neural network) may be trained to output velocity data for objects in response to radar data input to the machine learning model; [column 24, line 48-52] the model output at the second frame 804 may be determined using additional sensor data, such as additional range sensor data and/or image data. Such additional data may provide further positional and/or velocity information that may be used to refine the model).
Blaes does not expressly teach receive a parameter from the memory of the computer, the parameter being determined from a training process to modify an amodal representation of the object, the modified amodal representation being determined from a difference between a second velocity-compensated position of the object and the amodal representation of the object, wherein the amodal representation is a representation of the pose of an object obtained utilizing an aggregation of one or more historical sets of points, obtained previous to the first and second sensor scans, that describe a distance to at least a portion of the object; and.
However, Purdy teaches receive a parameter from the memory of the computer ([column 25, line 11-14] the memory 1418 of the vehicle computing device(s) 1404 stores the localization component 1420, the perception component 120; [column 10, line 25-26 & 28-31] The perception component 120 may generate or otherwise determine the perception data 128 ... the perception data 128 may include the one or more attribute(s) 130 associated with the object that are determined by the perception component 120, such as the velocity 132 of the object), the parameter being determined from a training process to modify an amodal representation of the object ([column 16, line 1-4] the perception component may average the shapes of the first representation and the second representation to determine the adjusted representation; [column 7, line 49-53] the perception data 128 may include one or more attribute(s) 130 associated with the object 110(1) that are determined by the perception component 120, such as a velocity 132 of the object 110(1); [column 20, line 36-44] the machine-learned model 806 may be configured or otherwise trained to determine, as an output 808, the attribute(s) 130 of the object. For instance, the machine-learned model 806 may be configured to determine the velocity 132 ... the machine-learned model 806 may be associated with or apart of the perception component 120; [column 34, line 41-43] altering a parameter of the machine-learned model to minimize the difference and obtain a trained machine-learned model. For instance, the machine-learned model training component may alter the parameter of the machine-learned model 806 to minimize the difference and obtain the trained machine-learned model; [column 36, line 6-8] the predicted attribute of the object comprises at least one of a size of the object, a location of the object, an orientation of the object, or a velocity of the object), the modified amodal representation ([column 9, 5-7 & line 9-13] the prediction component 124 may determine predicted trajectories of the objects 110 in the environment 108 … the prediction component 124 may determine a predicted trajectory of the object 110(1) based at least in part on the attribute(s) 130 included in the perception data 128, such as the velocity 132, the bounding box 134, and/or other attribute(s) 130; [column 36, line 6-8] the predicted attribute of the object comprises at least one of a size of the object, a location of the object, an orientation of the object, or a velocity of the object) being determined from a difference between a second velocity-compensated position of the object and the amodal representation of the object ([column 7, line 20-23] the sensor data 106 includes a representation 112 of the object 110(1) that is distorted (e.g., horizontally compressed) relative to an actual shape 114 of the object 110(1); [column 7, line 60-63] the perception component 120 may determine the attribute(s) 130 based on a difference between the representation 112 of the object 110(1) and the actual shape 114 of the object 110(1); [column 20, line 36-41] the machine-learned model 806 may be configured or otherwise trained to determine, as an output 808, the attribute(s) 130 of the object ... the machine-learned model 806 may be configured to determine ... adjusted representation 136; [column 7, line 56-59] the bounding box 134 may be indicative of one or more attributes associated with the object 110(1), such as the location, the orientation, a size of the object 110(1)).
It would have been obvious before the effective filing date of the claimed invention to one having ordinary skill in the art to modify the method and device of Blaes to incorporate the step/system of obtaining a velocity from the memory, determining the velocity through the machine-learned model to refine an object's representation, determining refined object's representation (a size, a location, an orientation of the object) from difference between the adjusted representation of the object and the actual shape of the object taught by Purdy.
The suggestion/motivation for doing so would have been to improve the accuracy of detecting precise locations of objects in the environment for the safe operation of autonomous vehicles ([column 6, line 3-8 & 13-16] the techniques can also improve segmentation by detecting which portions of the sensor data are moving together and which portions are remaining stationary ... the techniques described herein improve the safe operation of autonomous vehicles ... the techniques allow for determining more precise locations of objects in the environment, helping the vehicle to avoid coming in too close of proximity to the objects and/or helping avoid collisions). Further, one skilled in the art could have combined the elements as described above by known method with no change in their respective functions, and the combination would have yielded nothing more than predicted results.
The combination of Blaes and Purdy does not expressly teach wherein the amodal representation is a representation of the pose of an object obtained utilizing an aggregation of one or more historical sets of points, obtained previous to the first and second sensor scans, that describe a distance to at least a portion of the object; and.
However, PARK teaches wherein the amodal representation is a representation of the pose of an object ([0045] the odometry program estimates the change in pose of the rotary lidar in the first to Nth pieces by aligning the point groups of the first to Nth pieces, newly measured through the pose estimation model, with the voxels converted from each piece of scan data measured at an earlier point in time; [0055] The pose estimation model (200) includes a prediction model that predicts an estimate of the pose change amount for N pieces of voxels measured at the current time, and an observation model that calculates a first measurement of the pose change amount based on the position difference between the i-th piece of voxel measured at the current time and the m-th piece of voxel measured at the corresponding preceding time) obtained utilizing an aggregation of one or more historical sets of points ([0045] the odometry program estimates the change in pose of the rotary lidar in the first to Nth pieces by aligning the point groups of the first to Nth pieces, newly measured through the pose estimation model, with the voxels converted from each piece of scan data measured at an earlier point in time), obtained previous to the first and second sensor scans ([0078] The distribution ... is modeled ... using the covariance value representing the shape of the l-th voxel of the current piece and the m-th voxel of the corresponding scan at the preceding time point), that describe a distance to at least a portion of the object; and ([0048] the corresponding relationship between the voxels of a scan from an earlier time point (previously measured point cloud) and the voxels of the current fragment (newly measured point cloud) can be obtained by finding the combination that minimizes the distance between the two voxels; [0108] The pose estimation model (200) includes a prediction model that predicts an estimate of the pose change amount for N pieces of voxels measured at the current time, and an observation model that calculates a first measurement of the pose change amount based on the position difference between the i-th piece of voxel measured at the current time and the m-th piece of voxel measured at the corresponding preceding time).
It would have been obvious before the effective filing date of the claimed invention to one having ordinary skill in the art to modify the method and device of Blaes and Purdy to incorporate the step/system of determining the LiDAR's changing pose by using voxels created from earlier scans taught by PARK.
The suggestion/motivation for doing so would have been to prevent the degradation of odometry performance due to rapid movement ([0016] The present invention provides a distortion compensation algorithm that processes a scan by splitting it into very short intervals with minimal distortion effects, thereby preventing the degradation of odometry performance caused by point cloud distortion due to rapid movement). Further, one skilled in the art could have combined the elements as described above by known method with no change in their respective functions, and the combination would have yielded nothing more than predicted results. Therefore, it would have been obvious to combine Blaes and Purdy with PARK to obtain the invention as specified in claim 1.
Regarding claim 5, the combination of Blaes and Purdy with PARK teaches all the limitations of claim 1 above. Blaes teaches wherein the amodal representation of the object is determined from an aggregated history of scans of the object ([column 6, line 56-64] The positions of the objects can be used to determine flow fields for the objects within the radar data … rather than explicit object determination, the flow field may be performed on a pixel-by-pixel basis, such that each pixel is identified in the successive radar data snapshots with a velocity flow field established for all of the pixels of radar data. The flow field may be decoded by aggregating velocity data for particular clusters of pixels corresponding to bodies of objects to determine object velocities).
Regarding claim 6, the combination of Blaes and Purdy with PARK teaches all the limitations of claim 1 above. Purdy teaches wherein the instructions further comprise instructions to ([column 28, line 54-57] The processor(s) 1416 of the vehicle 1402 and the processor(s) 1430 of the computing device(s) 1428 can be any suitable processor capable of executing instructions to process data and perform operations): generate a geometric container that includes the third set of points ([column 20, line 21-25] if the sensor data 106 is a lidar point cloud including multiple lidar points associated with three spatial dimensions, these lidar points may be represented in four dimensions by the time-dimensional component 802; [column 10, line 42-45] the perception component 120 may average the sizes of the first representation 112(1) and the second representation 112(2) to determine the approximate width, the bounding box 134, and/or the adjusted representation); and assign a class label to the geometric container ([column 25, line 42-47] the perception component 120 can associate a bounding region (e.g., bounding box or otherwise an instance segmentation) with an identified object and can associate a confidence score associated with a classification of the identified object with the identified object).
Regarding claim 7, the combination of Blaes and Purdy with PARK teaches all the limitations of claim 6 above. Purdy teaches wherein the class label assigned to the geometric container is ([column 25, line 42-47] the perception component 120 can associate a bounding region (e.g., bounding box or otherwise an instance segmentation) with an identified object and can associate a confidence score associated with a classification of the identified object with the identified object) a cuboid encompassing a vehicle ([column 10, line 28-34] the perception data 128 may include the one or more attribute(s) 130 associated with the object that are determined by the perception component 120, such as the velocity 132 of the object, the bounding box 134 associated with the object, the adjusted representation 136 of the object, a location of the object, an orientation of the object, and/or the like).
Regarding claim 8, the combination of Blaes and Purdy with PARK teaches all the limitations of claim 1 above. Blaes teaches wherein the instructions further comprise instructions to: actuate a vehicle component based on the determined pose of the object ([column 25, line 38-42 & 48-50] the perception component 910 may determine what is in the environment surrounding the vehicle 904 and the planning component 912 may determine how to operate the vehicle 904 according to information received from the perception component 910 ... The trajectory may comprise instructions for controller(s) of the vehicle 904 to actuate drive components of the vehicle 904; [column 2, line 30-31] the perception system may perform updates on object state data, such as position, orientation).
Regarding claim 9, the combination of Blaes and Purdy with PARK teaches all the limitations of claim 8 above. Blaes teaches wherein the vehicle component is a steering component or a propulsion component ([column 25, line 48-55] The trajectory may comprise instructions for controller(s) of the vehicle 904 to actuate drive components of the vehicle 904 to effectuate a steering angle and/or steering rate, which may result in a vehicle position, vehicle velocity, and/or vehicle acceleration. For example, the trajectory may comprise a target heading, target steering angle, target steering rate, target position, target velocity, and/or target acceleration for the controller(s) to track).
With respect to claim 12, arguments analogous to those presented for claim 1, are applicable.
With respect to claim 16, arguments analogous to those presented for claim 5, are applicable.
With respect to claim 17, arguments analogous to those presented for claim 6, are applicable. With respect to claim 18, arguments analogous to those presented for claim 8, are applicable.
With respect to claim 19, arguments analogous to those presented for claim 9, are applicable.
Conclusion
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 DANIEL C. CHANG whose telephone number is (571)270-1277. The examiner can normally be reached Monday-Thursday and Alternate Fridays 8:00-5:00.
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, Chan S. Park can be reached at (571) 272-7409. 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.
/DANIEL C CHANG/Examiner, Art Unit 2669 /CHAN S PARK/Supervisory Patent Examiner, Art Unit 2669
Prosecution Insights