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
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments
Applicant's arguments filed 7/9/2026 have been fully considered but they are not persuasive. Applicant asserts as follows with respect to the rejection(s) of Independent Claim 1:
As another example of features and functionality that the Ready-Campbell / Austen, Kikani, Maeda, Milton and Oasa references fail to disclose or otherwise render obvious, whether alone or in combination, independent claim 1 as amended is generally directed to determining, while a bulldozer vehicle is operating in a cutting or pushing or loading mode using its blade tool attachment, that automated operations include predicting the loading of the vehicle's blade tool using on-vehicle sensors of multiple types, and further training that machine learning model to generate such predictions by capturing images of vehicles' blades during operation and using a separate camera-based machine learning model to predict blade tool loading from the captured images for use as training data for the separate blade load prediction machine learning model.
Conversely, the Ready-Campbell / Austen, Kikani, Maeda, Milton and Oasa references appear to lack any disclosure of training a machine learning model to generate predictions of loading of a vehicle's blade tool attachment by capturing images of vehicles' blades during operation and using a separate camera-based machine learning model to predict blade tool loading from the captured images for use as training data for the separate blade load prediction machine learning model.
With respect to previously pending dependent claim 9 that generally recited such functionality (and is now canceled, with the elements of dependent claim 9 being added to independent claim 1), the Office alleges on pages 20-21 of the current Office Action that paragraphs 133 and 134 of Ready-Campbell / Austen disclose acquiring images of a vehicle's blade tool and directly analyzing the acquired images to determine the loading of the blade tool. However, even if it is assumed for the sake of argument that Ready-Campbell / Austen does disclose acquiring images of a vehicle's blade tool and directly analyzing the acquired images to determine the loading of the blade tool, there is nothing in the cited portions of Ready- Campbell / Austen, or in any other portion of Ready-Campbell / Austen that Applicant can identify, of using a first machine learning model to generate training data for a second machine learning model, in any manner or for any reason, let alone to do so by using the first machine learning model to analyze acquired images of a vehicle's blade tool during operations to generate output indicating the loading of the blade tool and to then use that output of the first machine learning model to generate training data for a second machine learning model used to control autonomous operations of a vehicle with a blade tool. Thus, Ready-Campbell / Austen appears to lack any disclosure of "obtaining multiple camera-based estimates of how full that blade tool attachment is at multiple time points based at least in part on images of that blade tool attachment captured at the multiple time points that are input to a camera-based blade load predictor machine learning model", as recited in independent claim 1 as amended.
Examiner respectfully disagrees. As set forth in the rejection to now-rejected Dependent Claim 9, now incorporated into the rejection of Independent Claim 1, Austen discloses in at least Paragraphs 0133 – 0134 & 0136 – 0138 wherein the fill estimate engine may acquire information from an imaging sensor, such as a point cloud or direct image, which may be used to determine a fill state of the tool by comparing the image or point cloud to an empty representation of the tool, including being based on the soil composition properties of the material. While Austen does not appear to specifically set forth wherein the model is a machine learning model, Examiner respectfully asserts that such limitations are anticipated by Kikani as further relied upon in the relevant rejection(s).
Kikani teaches in at least Paragraph 0081 wherein the fill level of a tool attachment to a vehicle may be measured through the use of a machine learning model, which uses imagery data from sensors as taught in at least Paragraph 0046 to make the relevant determination. The model of Kikani is trained through the use of labeled or unlabeled training data as taught in at least Paragraphs 0054 & 0083, and may be retrained based on imagery sensor data collected during the working process as taught in Paragraphs 0046, 0083, & 0087.
Thus, Examiner respectfully asserts that reading Austen in view of Kikani renders obvious the invention of Independent Claim 1. As Austen discloses estimating a fill level of the tool through image-based means, including through assessment of soil composition, and Kikani recites a similar means of doing so, incorporating machine learning to identify both fill levels an soil composition that would cause such a fill, using image-based means as well, Examiner respectfully asserts that the combination of references renders obvious the present claimed invention in identifying a fill level of a tool through image-based machine-learning. Further, with respect to training the machine learning model, Examiner respectfully asserts that Kikani teaches training both the fill prediction model, and the underlying soil composition model based on collected data, [soil composition being used in Austen as a determinative factor in assessing fill level based on images] the model(s) being actively updated based on new data acquired during the running of the vehicle, thus anticipating the training of the machine learning model as set forth in Claim 1 under the broadest reasonable interpretaiton. Thus Applicant arguments are not persuasive.
In fact, the Office acknowledges on page 21 of the current Office Action that Ready- Campbell / Austen lacks disclosure of using a machine learning model to analyze captured images in the indicated manner, but alleges that Kikani discloses such functionality in paragraph 81.
However, the cited portion of Kikani is shown below, with emphasis added, and appears to, in the same manner as Ready-Campbell / Austen, lack any disclosure of using a first machine learning model to generate training data for a second machine learning model, in any manner or for any reason, let alone to do so by using the first machine learning model to analyze acquired images of a vehicle's blade tool during operations to generate output indicating the loading of the blade tool and to then use that output of the first machine learning model to generate training data for a second machine learning model used to control autonomous operations of a vehicle with a blade tool.
There is nothing in the cited passage about generating training data for a machine learning model in any manner, let alone using a first machine learning model to generate training data for a second machine learning model by using the first machine learning model to analyze acquired images of a vehicle's blade tool during operations to generate output indicating the loading of the blade tool and to then use that output of the first machine learning model to generate training data for a second machine learning model used to control autonomous operations of a vehicle with a blade tool.
While Kikani generally discusses using training data, Applicant is unable to identify any disclosure in Kikani of using a first machine learning model to generate training data for a second machine learning model, in any manner or for any reason, let alone to do so by using the first machine learning model to analyze acquired images of a vehicle's blade tool during operations to generate output indicating the loading of the blade tool and to then use that output of the first machine learning model to generate training data for a second machine learning model used to control autonomous operations of a vehicle with a blade tool.
Accordingly, Kikani, like Ready-Campbell / Austen, appears to lack any disclosure of "obtaining multiple camera-based estimates of how full that blade tool attachment is at multiple time points based at least in part on images of that blade tool attachment captured at the multiple time points that are input to a camera-based blade load predictor machine learning model", as recited in independent claim 1 as amended.
Examiner respectfully disagrees. As set forth above, Examiner respectfully asserts that Austen in view of Kikani renders obvious the present claimed invention. Specifically, as set forth above with respect to Kikani, Kikani teaches in at least Paragraph 0081 wherein the fill level of a tool attachment to a vehicle may be measured through the use of a machine learning model [i.e. a first machine learning model], which uses imagery data from sensors as taught in at least Paragraph 0046 to make the relevant determination. The model of Kikani is trained through the use of labeled or unlabeled training data as taught in at least Paragraphs 0054 & 0083, and may be retrained based on imagery sensor data collected during the working process as taught in Paragraphs 0046, 0083, & 0087 [i.e. a second machine learning model]. And, as further noted above, Kikani uses a machine learning model to determine soil composition, which is also trained and updated based on received data, soil composition further being a factor used in determining the fill level of the tool from the image data in Austen.
With respect to the acquisition of multiple images, Austen discloses in at least Paragraphs 0133 & 0134 wherein image data, which may be used to determine a fill state of the tool, may be acquired for multiple height and angle measurements captured by the imaging system [i.e. multiple images] which may be stored in a lookup table as disclosed in at least Paragraph 0137 for future use. As the Examiner has mapped the “lookup table” updated in Austen to the training of the model set forth above in Kikani, Applicant Arguments that the relevant amended limitations are not anticipated by Austen in view of Kikani are not persuasive, as in each instance, the underlying model used for the estimate is updated based on newly acquired image data, Kikani specifically doing so via machine learning similarly to as recited by the present claimed invention.
Thus, for at least the reasons set forth above as well as those set forth below with respect to the specific claim limitations, Examiner respectfully maintains the rejection of Independent Claim 1 [and dependents as applicable] under 35 USC 103.
Applicant’s arguments, see Pages 12 - 14, filed 7/9/2026, with respect to Independent Claims 12, 14, & 18 have been fully considered and are persuasive. The rejections of Claims 12 - 21, as well as Dependent Claim 5 which recites similar subject matter, under 35 USC 103 have been withdrawn.
Allowable Subject Matter
Claims 12 - 21 are allowed.
The following is an examiner’s statement of reasons for allowance: With respect to Independent Claim 18, the claim recites at least the following limitations:
receiving, by the one or more hardware processors, data for a powered earth-moving vehicle with a blade tool attachment,
wherein the powered earth-moving vehicle is moving at least one of tracks or wheels and using the blade tool attachment to perform pushing of material in an environment of the powered earth-moving vehicle,
the data indicating at least one of engine revolutions per minute, fuel consumption, engine torque, engine load, speed of the at least one of the tracks or wheels, or a transmission gear ratio;
applying, by the one or more hardware processors, a machine learning model to the received data to generate a predicted estimated degree of loading of the blade tool attachment with the material during the pushing of the material;
determining, by the one or more hardware processors and based at least in part on the predicted estimated degree of loading of the blade tool attachment satisfying a defined threshold level, an output indicating that the powered earth-moving vehicle is experiencing slippage from the pushing of the material; and
initiating, by the one or more hardware processors and in response to the determining of the output, an automated remedial action.
No single prior art reference, nor combination of references, has been found to anticipate the above limitations, when read in view of the remaining limitations of the claim. Specifically, no single prior art reference, nor combination of references, has been found to anticipate a machine learning model that predicts an estimated degree of blade loading, with a subsequent output being provided based on said estimate that indicates that powered earth-moving vehicle is experiencing slippage from the pushing of the material [emphasis added]. While prior art has been found to anticipate the prediction of future slippage of a vehicle based on predicted future fill conditions, no single prior art reference, nor combination of references, has been found to anticipate wherein a current slippage condition of a vehicle is assessed based on an estimated degree of blade loading obtained from a machine learning model as recited by the present claimed invention.
The closest prior art found is Austen (US 2022/0120060 A1), Kikani (US 2022/0412051 A1), and Maeda (US 2025/0277351 A1), Austen reciting a bulldozer vehicle, in which a system is provided to monitor the fill level of a tool and initiate an end to operations based on the fill level exceeding a threshold, Kikani reciting wherein a machine learning model is used to determine a fill level of a tool based on sensor data, and Maeda reciting wherein a slippage of a work vehicle may be predicted by reinforcement learning.
More specifically, Austen recites a bulldozer vehicle, configured to determine a fill level of a tool, and end operations based on the fill level exceeding a threshold. A plurality of sensor systems may be provided on the vehicle, in order to assess the fill level, and the end of the operation may include raising or otherwise operating the blade tool to disengage from the operation. Austen however appears to be silent regarding determining an amount of slippage experienced by the vehicle and making control decisions based on such, much less wherein the amount of slippage is based on the amount of fill of the tool assessed by a machine learning model as recited by the present claimed invention.
Kikani recites wherein a fill level of an excavator bucket or other tool may be estimated mathematically using a trained machine learning model, based on the depth of the leading edge of the tool beneath the ground surface, imagery, sensor data, and/or through measured kinetic force on the tool. Kikani however appears to be silent regarding determining an amount of slippage experienced by the vehicle and making control decisions based on such, much less wherein the amount of slippage is based on the amount of fill of the tool assessed by a machine learning model as recited by the present claimed invention.
Maeda recites wherein a reinforcement learning based approach may be used to determine if a slippage of the work vehicle is predicted to occur along a movement path based on if the predicted slip of the vehicle exceeds the shoe slip limit of the vehicle. Based on slip being determined to occur, another candidate movement path may be assessed such that the slip of the work vehicle does not occur, which may include the adjustment of excavation depth. Maeda however appears to be silent regarding determining an amount of slippage currently being experienced by the vehicle and making control decisions based on such, much less wherein the amount of slippage is based on the amount of fill of the tool assessed by a machine learning model as recited by the present claimed invention.
Thus, Independent Claim 18 is found to be allowable. Independent Claims 12 & 14, while not identical, recite substantially the same limitations as those found to be allowable with respect to Independent Claim 18, and are therefore found to be allowable under similar rationale as set forth above. Dependent Claims 13, 15 – 17, & 19 – 21 are found to be allowable based on their dependence on an allowable base claim, as well as the limitations recited within.
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.”
Claim 5 is objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
The following is a statement of reasons for the indication of allowable subject matter: The aforementioned Dependent Claim(s) recite substantially similar limitations as those found in Independent Claims 12, 14, & 18, and are therefore objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims under similar rationale as set forth with respect to Claims 12 – 21 as set forth above.
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(s) 1 - 4 & 11 is/are rejected under 35 U.S.C. 103 as being unpatentable over Austen (US 2022/0120060 A1) in view of Kikani (US 2022/0412051 A1).
Regarding Claim 1:
Austen discloses: An autonomous vehicle steering system using a blade tool attachment, comprising: (Austen discloses in at least Paragraphs 0005 & 0030 a earth shaping vehicle operating and navigating autonomously in a dig site [i.e. an autonomous vehicle steering system], the earth shaping vehicle including a bulldozer blade as disclosed in at least Paragraph 0203 [i.e. using a blade tool attachment])
a bulldozer vehicle with a chassis, tracks, a blade tool attachment on a front of the chassis, hydraulic arms between the chassis and the blade tool attachment, one or more first controls for manipulating movement of the tracks, and one or more second controls for manipulating the blade tool attachment via the hydraulic arms; (Austen discloses in at least Paragraphs 0034 & 0203 wherein the earth shaping vehicle may include a bulldozer, the earth shaping vehicle including a chassis, drive system, and earth shaping tool as disclosed in at least Paragraph 0041 [i.e. a bulldozer vehicle with a chassis, tracks (see annotated Figure 2A, below], and a blade tool attachment on a front of the chassis]. At least Paragraphs 0035, 0042, & 0050 of Austen further disclose wherein arms, actuated by hydraulics, may be used to control the position of the tool [i.e. hydraulic arms between the chassis and the blade tool attachment]. At least Paragraphs 0057 & 0059 of Austen discloses wherein the earth shaping vehicle may include manual input devices, such as joysticks, for controlling the drive system and earth shaping tool [i.e. one or more first controls for manipulating movement of the tracks, and one or more second controls for manipulating the blade tool attachment via the hydraulic arms]. Figure 2A of Austen, annotated by the Examiner, is presented below which depicts many of the above features)
PNG
media_image1.png
544
680
media_image1.png
Greyscale
a microcontroller unit on the bulldozer vehicle that is capable of effecting movement of the first and second controls via piston displacement mechanisms; and a control system on the bulldozer vehicle that is configured to be in communication with the microcontroller unit and to perform automated operations including: (Austen discloses in at least Paragraphs 0104, 0128, & 0132 wherein a hydraulic distribution engine [i.e. a microcontroller unit on the bulldozer vehicle] may be provided to monitor and adjust hydraulic pressure provided to actuate the hydraulic tool actuators [i.e. effecting movement of the first and second controls via piston displacement mechanisms]. At least Paragraphs 0031, 0045, & 0123 of Austen further disclose wherein the vehicle may further include a controller to perform analysis of sensor data and issue control instructions [i.e. a control system on the bulldozer vehicle that is configured to be in communication with the microcontroller unit and to perform automated operations])
including, for each of one or more operational bulldozer vehicles each having a respective blade tool attachment, obtaining multiple camera-based estimates of how full that blade tool attachment is at multiple time points based at least in part on images of that blade tool attachment captured at the multiple time points that are input to a camera-based blade load predictor… model, and (Austen discloses in at least Paragraphs 0133 & 0134 wherein the fill estimate engine may acquire information from an imaging sensor, such as a point cloud or direct image, which may be used to determine a fill state of the tool by comparing the image or point cloud to an empty representation of the tool, which may be acquired for multiple height and angle measurements captured [i.e. obtaining multiple camera-based estimates of how full that blade tool attachment is at multiple time points based at least in part on images of that blade tool attachment captured at the multiple time points that are input to a camera-based blade load… model] and stored in a lookup table as disclosed in at least Paragraphs 0136 – 0138 of Austen)
supplying the multiple camera-based estimates of how full the blade tool attachment is as training data to the …model along with additional values that are captured at the multiple time points for that operational bulldozer vehicle from at least some of the types of sensors on that operational bulldozer vehicle; (Austen discloses in at least Paragraphs 0133 & 0134 wherein image data, which may be used to determine a fill state of the tool, may be acquired for multiple height and angle measurements captured by the imaging system, the amount of earth currently in the tool being determined based on features such as soil composition based in turn on the image data as disclosed in at least Paragraph 0136 [i.e. multiple camera-based estimates of how full the blade tool attachment is] which may be stored in a lookup table as disclosed in at least Paragraphs 0136 – 0138, with a particular emphasis on Paragraph 0137 for future reference in determining the fill level of the tool based on a comparison of the lookup table to the current state [i.e. supplying the multiple camera-based estimates of how full the blade tool attachment is as training data to the model along with additional values that are captured at the multiple time points for that operational bulldozer vehicle from at least some of the types of sensors on that operational bulldozer vehicle)
determining, while the bulldozer vehicle is in motion and is using the blade tool attachment to push ground material in at least one of a pushing mode or a cutting mode or a loading mode, and by a… model using multiple data readings from a plurality of sensors of the plurality of types on the bulldozer vehicle, at least one of a predicted degree or a predicted amount of how full the blade tool attachment is with the ground material; (Austen discloses in at least Paragraphs 0105 & 0117 wherein a fill estimate engine is configured to determine an estimate of the tool fill level as the tool is moved over a target path, such as when the tool is actuated below the earth surface and the vehicle is driven to fill the tool with earth as disclosed in at least Paragraph 0083 [i.e. determining, while the bulldozer vehicle is in motion and is using the blade tool attachment to push ground material in at least one of a pushing mode or a loading mode a predicted amount of how full the blade tool attachment is with the ground material]. At least Paragraphs 0118 & 0119 of Austen further disclose wherein the fill estimate engine may estimate the volume of earth in the tool using a previously trained prediction model and measurements from sensors, such as the measured force of the earth acting on the tool beneath the surface, visual/spatial sensors to detect the volume of earth, the distance travelled, and the like [i.e. by a model using multiple data readings from a plurality of sensors on the bulldozer vehicle])
determining, based at least in part on the determined at least one of the predicted degree or the predicted amount of how full the blade tool attachment of the bulldozer vehicle is with the ground material satisfying a defined threshold, to initiate an end to the at least one of the pushing mode or the cutting mode or the loading mode; and (Austen discloses in at least Paragraphs 0120 – 0122 wherein the fill estimate is compared to a threshold volume, which may be defined as the maximum available volume of the tool, or a volume manually set by a human operator [i.e. determining if the predicted amount of how full the blade tool attachment is with the ground material satisfies a defined threshold] to determine if the threshold volume has been met. If the estimated quantity is less than the threshold volume, the digging routine may resume, however if the estimated volume is greater than the threshold volume, the excavation vehicle is configured to raise the tool above the ground surface and perform a dump routine as disclosed in at least Paragraphs 0122 & 0123 of Austen [i.e. based at least in part on the defined threshold being satisfied, initiating an end to the at least one of the pushing mode or the loading mode])
initiating, in response to the determining to initiate the end to the at least one of the pushing mode or the cutting mode or the loading mode, autonomous operations of the bulldozer vehicle to at least one of raise the blade tool attachment to end contact with the ground material by using the second controls to manipulate the hydraulic arms via at least one of the piston displacement mechanisms, or stop the motion of the bulldozer vehicle by using the first controls to stop the movement of the tracks via at least one of the piston displacement mechanisms. (Austen discloses in at least Paragraphs 0122 & 0123 wherein when the measured fill volume exceeds the threshold quantity, the controller of the vehicle is configured to raise the tool above the ground surface and perform a dump routine, in which the vehicle moves the excavated earth to a dump pile location, deposits the earth, and returns to the target excavation location to continue excavating as disclosed in at least Paragraphs 0198 – 0200 of Austen [i.e. initiating, in response to the determining to initiate the end to the at least one of the pushing mode or the loading mode, autonomous operations of the bulldozer vehicle to at least one of raise the blade tool attachment to end contact with the ground material by using the second controls to manipulate the hydraulic arms via at least one of the piston displacement mechanisms])
Austen however appears to be silent regarding:
Wherein the predicted amount of how full the blade tool attachment is determined by a trained machine learning model
training a machine learning model to predict how full a bulldozer blade tool attachment is with material using data readings from a plurality of types of sensors,
However Kikani teaches wherein a machine learning model is utilized to determine the fill level of an excavator bucket or similar tool as the excavator traverses the field.
Wherein the predicted amount of how full the blade tool attachment is determined by a trained machine learning model (However Kikani teaches in at least Paragraph 0081 wherein a fill level of an excavator bucket or other tool may be estimated mathematically using a trained machine learning model, based on the depth of the leading edge of the tool beneath the ground surface, imagery, sensor data, and/or through measured kinetic force on the tool [i.e. a trained machine learning model is used to predict the amount of how full the blade tool attachment is])
training a machine learning model to predict how full a bulldozer blade tool attachment is with material using data readings from a plurality of types of sensors, However Kikani teaches in at least Paragraph 0081 wherein a fill level of an excavator bucket or other tool may be estimated mathematically using a trained machine learning model, based on the depth of the leading edge of the tool beneath the ground surface, imagery, sensor data, and/or through measured kinetic force on the tool [i.e. using data readings from a plurality of types of sensors]. The model of Kikani is trained through the use of labeled or unlabeled training data as taught in at least Paragraphs 0054 & 0083, and may be retrained based on imagery sensor data collected during the working process as taught in Paragraphs 0046, 0083, & 0087 [i.e. training a machine learning model to predict how full a bulldozer blade tool attachment is with material using data readings from a plurality of types of sensors])
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Austen by incorporating the determination of excavator tool fill level based on a trained machine learning model as taught by Kikani.
The motivation to do so is that, as acknowledged by Kikani in at least Paragraph 0081, the fill level of the tool may be better estimated through a plurality of factors, improving the determination of if the tool is filled or not during an earth-manipulation operation.
Regarding Claim 2:
The autonomous vehicle steering system of claim 1 wherein the determining to initiate the end to the at least one of the pushing mode or the cutting mode or the loading mode includes determining to switch to a carrying mode to move the ground material in the blade tool attachment to a different location, and wherein the initiated autonomous operations include using the second controls to raise the blade tool attachment to end contact with the ground material, and using the first controls to cause the movement of the tracks toward the different location.
Austen discloses in at least Paragraphs 0122 & 0123 wherein when the measured fill volume exceeds the threshold quantity, the controller of the vehicle is configured to raise the tool above the ground surface and perform a dump routine [i.e. determining to switch to a carrying mode to move the ground material in the blade tool attachment to a different location], in which the vehicle moves the excavated earth to a dump pile location, deposits the earth, and returns to the target excavation location to continue excavating as disclosed in at least Paragraphs 0198 – 0200 of Austen [i.e. wherein the initiated autonomous operations include using the second controls to raise the blade tool attachment to end contact with the ground material, and using the first controls to cause the movement of the tracks toward the different location].
Regarding Claim 3:
The autonomous vehicle steering system of claim 1 wherein the determining of the at least one of the predicted degree or the predicted amount of how full the blade tool attachment is with the ground material includes generating a binary output of the blade tool attachment being fully loaded and the defined threshold is that generated binary output, or wherein the determining of the at least one of the predicted degree or the predicted amount of how full the blade tool attachment is with the ground material includes generating a degree of loading of the blade tool attachment relative to being fully loaded and the defined threshold is an indicated degree of loading that is exceeded by the generated degree of loading, or wherein the determining of the at least one of the predicted degree or the predicted amount of how full the blade tool attachment is with the ground material includes generating a predicted estimate of the quantity of the ground material loaded in the blade tool attachment and the defined threshold is an indicated quantity that is exceeded by the generated estimate of the quantity.
Austen discloses in at least Paragraphs 0120 – 0122 wherein the fill estimate is compared to a threshold volume, which may be defined as the maximum available volume of the tool, or a volume manually set by a human operator [i.e. wherein the determining of the at least one of the predicted degree or the predicted amount of how full the blade tool attachment is with the ground material includes generating a predicted estimate of the quantity of the ground material loaded in the blade tool attachment and the defined threshold is an indicated quantity that is exceeded by the generated estimate of the quantity] to determine if the threshold volume has been met. If the estimated quantity is greater than the threshold volume, the excavation vehicle is configured to raise the tool above the ground surface and perform a dump routine as disclosed in at least Paragraphs 0122 & 0123 of Austen.
Regarding Claim 4:
The autonomous vehicle steering system of claim 1 wherein the determining of the at least one of the predicted degree or the predicted amount of how full the blade tool attachment is with the ground material is performed repeatedly during the motion of the bulldozer vehicle using the blade tool attachment to push the ground material in the at least one of the pushing mode or the cutting mode or the loading mode, and wherein the automated operations include determining, before the determining to initiate the end to the at least one of the pushing mode or the cutting mode or the loading mode, to continue the at least one of the pushing mode or the cutting mode or the loading mode in response to one or more prior determinations of the at least one of the predicted degree or the predicted amount of how full the blade tool attachment is with the ground material being below the defined threshold.
Austen discloses in at least Paragraphs 0120 – 0122 wherein the fill estimate is compared to a threshold volume, which may be defined as the maximum available volume of the tool, or a volume manually set by a human operator [i.e. determining if the predicted amount of how full the blade tool attachment is with the ground material satisfies a defined threshold] to determine if the threshold volume has been met. If the estimated quantity is less than the threshold volume, the digging routine may resume [i.e. determining to continue the at least one of the pushing mode or the cutting mode or the loading mode in response to one or more prior determinations of the at least one of the predicted degree or the predicted amount of how full the blade tool attachment is with the ground material being below the defined threshold]. Austen further discloses in at least Paragraph 0112 wherein the tool fill level is updated periodically in an automatic manner [i.e. wherein the determining of the at least one of the predicted degree or the predicted amount of how full the blade tool attachment is with the ground material is performed repeatedly during the motion of the bulldozer vehicle using the blade tool attachment to push the ground material].
Regarding Claim 11:
The autonomous vehicle steering system of claim 1 wherein the control system is configured to implement at least some automated operations of an earth-moving vehicle autonomous operations control system by executing software instructions of the earth-moving vehicle autonomous operations control system, and wherein the determining of the at least one of the predicted degree or the predicted amount of how full the blade tool attachment is with the ground material and the determining to initiate the end to the at least one of the pushing mode or the cutting mode or the loading mode and the initiating of the autonomous operations are performed autonomously without receiving human input and without receiving external signals other than GPS signals and real-time kinematic (RTK) correction signals.
Austen discloses in at least Paragraph 0077 wherein the control logic disclosed is implemented via software [i.e. the control system is configured to implement at least some automated operations of an earth-moving vehicle autonomous operations control system by executing software instructions of the earth-moving vehicle autonomous operations control system]. Austen further discloses in at least Paragraph 0112 wherein the tool fill level is updated periodically in an automatic manner [i.e. without receiving human input and without receiving external signals], said tool fill level evaluation being compared to a threshold to trigger ending the pushing mode as set forth above.
Claim(s) 6 - 8 is/are rejected under 35 U.S.C. 103 as being unpatentable over Austen (US 2022/0120060 A1) in view of Kikani (US 2022/0412051 A1) as applied to claim 1 above, and further in view of Milton (US 2021/0312725 A1).
Regarding Claim 6:
The autonomous vehicle steering system of claim 1 wherein the trained machine learning model is an LSTM (long short-term memory) recurrent neural network with a classifier output, and wherein the multiple data readings from the plurality of sensors on the bulldozer vehicle provide vehicle status data that includes multiple data types from a group including RPMs (revolutions per minute) of an engine of the bulldozer vehicle, and fuel consumption of the bulldozer vehicle, and engine torque of the bulldozer vehicle, and engine load of the bulldozer vehicle, and speed of the tracks of the bulldozer vehicle, and speed of the chassis of the bulldozer vehicle, and pitch pressure for the blade tool attachment, and a transmission gear ratio in use by the bulldozer vehicle.
Austen discloses in at least Paragraph 0208 wherein an engine RPM may be monitored as a vehicle performance metric [i.e. wherein the multiple data readings from the plurality of sensors on the bulldozer vehicle provide vehicle status data that includes engine revolutions per minute]. At least Paragraph 0050 of Austen further discloses wherein measurement sensors may measure the vehicle speed [i.e. wherein the multiple data readings from the plurality of sensors on the bulldozer vehicle provide vehicle status data that includes speed of the at least one of the tracks or wheels/chassis of the vehicle]. Austen however does not appear to specifically disclose wherein the trained machine learning model is an LSTM (long short-term memory) recurrent neural network with a classifier output.
While Kikani teaches in at least Paragraph 0054 wherein the machine learning model may be a neural network model, Kikani appears to be silent regarding wherein the trained machine learning model is an LSTM (long short-term memory) recurrent neural network with a classifier output.
However Milton teaches in at least Paragraphs 0093 & 0094 wherein a machine learning model implemented on the vehicle and configured to determine and label vehicle movement patterns may be embodied as a convolutional long-short-term memory (LSTM) neural network configured to classify time series data [i.e. wherein the trained machine learning model is an LSTM (long short-term memory) recurrent neural network with a classifier output].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Austen by incorporating the use of an LSTM (long short-term memory) recurrent neural network with a classifier output in the vehicle system as taught by Milton.
The motivation to do so is that, as acknowledged by Milton in at least Paragraphs 0093 & 0094, the interpretation of time series data by the learning model may be improved, improving the assessment of vehicle fill degree.
Regarding Claim 7:
The autonomous vehicle steering system of claim 6 wherein the multiple data readings further include at least one of a measure of an amount of the ground material that is spilling over at least one of a side or a top of the blade tool attachment, or a degree of tilt in a pitch of the bulldozer vehicle.
Austen discloses in at least Paragraph 0118 wherein the fill estimate may be based in part on the depth of the leading edge of the tool beneath the ground surface, which may be based in part on the angle of the implement with respect to the ground as disclosed in at least Paragraph 0127 [i.e. the data readings include a measure of a degree of tilt in a pitch of the bulldozer vehicle].
Regarding Claim 8:
The autonomous vehicle steering system of claim 6 wherein the multiple data readings further include one or more data readings from one or more additional sensors external to the bulldozer vehicle, the one or more additional sensors including at least one of an image sensor of a camera or a sensor of a LiDAR component.
Austen discloses in at least Paragraphs 0133 & 0134 wherein the fill estimate engine may acquire information from an imaging sensor, such as a point cloud or direct image, which may be used to determine a fill state of the tool by comparing the image or point cloud to an empty representation of the tool as disclosed in at least Paragraphs 0136 – 0138 of Austen [i.e. wherein the multiple data readings further include one or more data readings from an image sensor or LiDAR component].
Claim(s) 10 is/are rejected under 35 U.S.C. 103 as being unpatentable over Austen (US 2022/0120060 A1) in view of Kikani (US 2022/0412051 A1) as applied to claim 1 above, and further in view of Oasa (US 2023/0304255 A1).
Regarding Claim 10:
The autonomous vehicle steering system of claim 1 further comprising: a LiDAR component that is mounted on the bulldozer vehicle and configured to obtain LiDAR data indicating a plurality of three-dimensional (“3D”) points on surfaces of at least some of a job site on which the bulldozer vehicle is located; one or more GPS antennas mounted at one or more positions on the chassis and capable of receiving GPS signals for use in determining GPS coordinates of at least some of the chassis; one or more inertial navigation system units mounted at one or more positions on the chassis and capable of determining a current direction of the bulldozer vehicle; and one or more first position sensors mounted on the hydraulic arms and configured to detect one or more first angles between the chassis and the hydraulic arms, and one or more second position sensors mounted on the blade tool attachment and configured to detect one or more second angles between the blade tool attachment and at least one of the hydraulic arms, and wherein the automated operations include providing information for display about the determined at least one of the predicted degree or the predicted amount of how full the blade tool attachment is with the ground material.
Austen discloses in at least Paragraphs 0048 & 0049 wherein the earth-moving vehicle may include spatial and imaging sensors such as LIDAR sensors [i.e. a LiDAR component that is mounted on the bulldozer vehicle and configured to obtain LiDAR data indicating a plurality of three-dimensional (“3D”) points on surfaces of at least some of a job site on which the bulldozer vehicle is located]. At least Paragraph 0045 of Austen further discloses wherein the vehicle may include a GPS system as a position sensor for determining the position of the earth shaping vehicle [i.e. one or more GPS antennas mounted at one or more positions on the chassis and capable of receiving GPS signals for use in determining GPS coordinates of at least some of the chassis]. At least Paragraph 0050 of Austen discloses wherein a plurality of measurement sensors may be disposed on the vehicle, including inertial measurement sensors [i.e. one or more inertial navigation system units mounted at one or more positions on the chassis and capable of determining a current direction of the bulldozer vehicle] and end-effector sensors positioned at each joint of the earth shaping tool, the latter measuring a change in angle of the tool relative to each joint [i.e. one or more first position sensors mounted on the hydraulic arms and configured to detect one or more first angles between the chassis and the hydraulic arms, and one or more second position sensors mounted on the blade tool attachment and configured to detect one or more second angles between the blade tool attachment and at least one of the hydraulic arms]. Austen however appears to be silent regarding wherein the automated operations include providing information for display about the determined at least one of the predicted degree or the predicted amount of how full the blade tool attachment is with the ground material.
However Oasa teaches in at least Paragraphs 0213 & 0214 wherein a vehicle, such as a loading machine with associated bucket as taught in at least Paragraph 0006, may be provided with an output device, such as an LCD screen, through which the estimated excavated weight is output [i.e. providing information for display about the determined at least one of the predicted degree or the predicted amount of how full the blade tool attachment is with the ground material].
It would have been obvious to one of ordinary skill in the art before the effective filing date of the present claimed invention to have modified the disclosure of Austen by incorporating the output of estimated excavated weight through a display as taught by Oasa.
The motivation to do so is that, as acknowledged by Oasa in at least Paragraphs 0213 & 0214, the operator may be made aware of the predicted weight excavated, such that control to excavate a target weight may take place, improving the operator awareness of the excavation process.
Conclusion
The following prior art made of record but not relied upon is considered pertinent to the Applicant’s disclosure:
Maeda (US 2025/0277351 A1): Maeda recites wherein a reinforcement learning based approach may be used to determine if a slippage of the work vehicle is predicted to occur along a movement path based on if the predicted slip of the vehicle exceeds the shoe slip limit of the vehicle. Based on slip being determined to occur, another candidate movement path may be assessed such that the slip of the work vehicle does not occur, which may include the adjustment of excavation depth.
Oasa (US 2023/0304255 A1): Oasa recites wherein a vehicle, such as a loading machine with associated bucket, may be provided with an output device, such as an LCD screen, through which the estimated excavated weight is output. Oasa further recites wherein upon the difference between the estimated and target weight being below a specified amount, the output device is controlled to output notification data indicating such to the operator.
Wei (US 9,487,929 B2): Wei recites a method for controlling an earthmoving machine, including the determination of the profile of a work surface, such as bumps or raised portions of a field being worked, and determining a depth adjustment of the work implement based on such.
Verho (US 2023/0175232 A1): Verho recites a method of controlling a work machine, specifically including the boom of said work machine, on the basis of received driveline parameters. The vehicle may include load sensing and traction control systems, in order to determine if the boom/bucket should be actuated or not.
Shinkai (US 2019/0104675 A1): Shinkai recites a work vehicle system for working the ground upon which the vehicle passes, including the determination of slippage of the work vehicle relative to the ground surface. The vehicle state is then changed based on the engine load and slip determined.
Tobias (DE 102019214430 A1): Tobias recites a system for evaluating driving dynamics of a vehicle, based on assessment of vehicle characteristics, including inputs and loading levels. The evaluation may take place through a machine learning module, trained on relevant data, including a total mass and mass distribution of the vehicle. Responses of the vehicle to conditions such as slippage may be predicted and intervened in advance.
Armstrong (US 12,254,403 B2): Armstrong recites a container fill level determination system, including receiving images from sensors and training a machine learning model based on the collected image data. Multiple fill levels may be provided to the training program as images in order to train the machine learning model, which is then deployed for use.
Zhang (NPL: High Precision Control and Deep Learning-based Corn Stand Counting Algorithms for Agricultural Robot): Zhang recites a deep learning algorithm for determining a depth of an agricultural material, such as corn. The determined data may be used to predict the interactions of a robot with the field, including slips in the longitudinal and lateral directions.
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 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 CHRISTOPHER RYAN CARDIMINO whose telephone number is (571)272-2759. The examiner can normally be reached M-Th 8:30-5:00.
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, Ramya Burgess can be reached at (571)272-6011. 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.
/CHRISTOPHER R CARDIMINO/Examiner, Art Unit 3661
/RAMYA P BURGESS/Supervisory Patent Examiner, Art Unit 3661
