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 .
Response to Amendment
This action is in response to the amendment filed on October 17th, 2025. Claims 1, 3, 10, 13, 15, and 20 have been amended while claims 2 and 14 have been cancelled. The amended claims limitations have been fully considered but are not persuasive. Claims 1-20 remain rejected in the application.
Response to Arguments
In response to all of applicant’s arguments, which are all centered around applicant’s belief that Bai fails to disclose graph transformer networks. All arguments have been fully considered but are not persuasive. Bai explicitly discloses graph transformer models (Bai: Title “GRAPH TRANSFORMER NEURAL NETWORK”)[Bai: 0005 “graph transformer neural network”]. Claims 1-20 remain rejected in the application.
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.
Claims 1, 8, 9, 10, 11, 13, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Barr et. al (U.S. Patent Publication No. 2020/0387579), in view of Bai et. al (U.S. Patent Publication No. 2021/0081505).
Regarding claim 1, Barr discloses a method comprising: receiving a digital representation of a three-dimensional (3D) object (interpreted as the system gets a computer file that describes the objects shape) [Barr: 0051 “the method may include receiving input data that includes shape parameters of a geometric object and flow parameters associated with the geometric object.”] [Barr: 0057 “The shape information may include curves, lines, radii, width, length, depth information, and the like. The geometric object may be a two - dimensional object or a three-dimensional object.”] (receiving data about shape parameters of a geometric object corresponds to receiving a digital 3d object); learning, using a model, relationships between a plurality of points on a surface of the 3D object (interpreted as: train a fast stand-in ML model to understand how surface points interact) [Barr: 0019 “As a result, the deep learning surrogate model for CFD reduces the time it takes for a designer ( or CFD practitioner ) to get a flow solution given a grid ( representing a discretized model of the actual geometry of interest ).”] [Barr: 0004 “Accordingly, the predictive model described herein can be used to replace ( i.e. , a surrogate ) a traditional CFD simulation.”] (teaches identifying a network as CFD “surrogate” that learns from a discretized surface grid (points), thereby learning relationships among those surface points exactly as claimed); and generating, using the surrogate model, one or more predictions about fluid properties along the surface of the 3D object (interpreted as the surrogate then outputs pressures, velocities, etc. at those surface locations) [Barr: 0005 “output one or more attributes of the predicted CFD flow about the geometric object”][Barr: 0017 “These equations describe how the velocity, pressure, temperature, and density of a moving fluid are related”](surrogate produces fields (velocity, pressure, etc.) for the objects surface), but fails to explicitly disclose Graph Transformer Network (GTN), GTN.
However, Bai discloses Graph Transformer Network (GTN), GTN (Bai: Title “GRAPH TRANSFORMER NEURAL NETWORK”).
Barr and Bai are both considered to be analogous to the claimed invention because they are in the same field of machine learning surrogate models. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Barr to incorporate Bai’s teachings of utilizing graph transformers. The motivation for such a combination would provide the benefit of richer relational features with predictable accuracy and speed benefits.
Regarding claim 8, Barr and Bai disclose the method of claim 1, wherein the 3D object comprises one of: an aerial vehicle, a ground vehicle, a watercraft, a subsurface vehicle, a projectile [Barr: 0001 “assets can include, among other things, industrial manufacturing equipment on a production line, drilling equipment for use in mining operations, wind turbines that generate electricity on a wind farm, transportation vehicles (trains, subways, airplanes, etc.)”].
Regarding claim 9, Barr and Bai disclose the method of claim 1, wherein the one or more predictions comprise one of: heat, pressure, velocity, radio frequency and types thereof [Barr: 0017 “These equations describe how the velocity, pressure, temperature, and density of a moving fluid are related”].
Regarding claim 10, Barr discloses the method of claim 1, further comprising: receiving a digital representation of an environment near the 3D object [Barr: 0051 “the method may include receiving input data that includes shape parameters of a geometric object and flow parameters associated with the geometric object. The flow parameters may include external conditions around the geometric object.”](teaches receiving a 3D object and may include its external conditions which is the environment); learning, using the surrogate model, relationships between the plurality of points on the surface of the 3D object and/or a plurality of points near the surface of the 3D object (interpreted as the surrogate model trains on how surface points interact with nearby flow-field points) [Barr: 0018 “computers are used to perform the calculations required to simulate the free - stream flow of the fluid, and the interaction of the fluid (liquids and gases) with surfaces of an object which are defined by boundary conditions”](teaches the surrogate model learns the relationships between the points on the surface of the object); and generating, using the surrogate model, one or more predictions about fluid properties of the environment near the surface of the 3D object [Barr: 0050 “the output may include any information of the CFD flow over time including graphical displays of the object itself, visual patterns of the flow around the object, measurements taken from the flow including velocity in one or more directions, temperature, density, pressure, turbulence, and the like.”](teaches what the output may include which is the prediction), but fails to explicitly disclose GTN, GTN.
However, Bai discloses GTN, GTN (Bai: Title “GRAPH TRANSFORMER NEURAL NETWORK”).
Barr and Bai are both considered to be analogous to the claimed invention because they are in the same field of machine learning surrogate models. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Barr to incorporate Bai’s teachings of utilizing graph transformers. The motivation for such a combination would provide the benefit of richer relational features with predictable accuracy and speed benefits.
Regarding claim 11, Barr and Bai disclose the method of claim 10, wherein the digital representation of the 3D object is different from the digital representation of the environment near the 3D object [Barr: 0044 “The input file 310 may include shape parameters 311 which include a geometry of an object which is being designed. The object may be an airfoil, an engine, a gas turbine, a steam turbine, a wing, or any other body of mass. The shape parameters 311 may include boundary line measurements of the object plus an information about an environment where the object is located”](discloses the object shape parameters and then further teaches there is separate information about the environment, necessarily meaning that the environment is different than the object).
Claims 13 and 20 are system and non-transitory computer-readable media claims corresponding to the method claim 1 above. Barr further discloses processing circuitry [Barr: 0021 “a cloud computing system includes at least one processor circuit”]. Thus, claims 13 and 20 are rejected for the same reason as claim 1.
Claims 3 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Barr et. al (U.S. Patent Publication No. 2020/0387579), in view of Bai et. al (U.S. Patent Publication No. 2021/0081505), in further view of Rawlinson et. al (U.S. Patent No. 11,357,573).
Regarding claim 3, Barr in view of Bai disclose the method of claim 1, but fail to explicitly disclose wherein the digital representation comprises point cloud data.
However, Rawlinson discloses wherein the digital representation comprises point cloud data (Rawlinson: 606; Fig. 6).
Barr, Bai, and Rawlinson are considered to be analogous to the claimed invention because they are in the same field of addressing three dimensional objects for data driven flow. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Barr and Bai to incorporate Rawlinson’s teachings of utilizing point cloud data. The motivation for such a combination would provide a compact way to capture full 3D geometry for neural network pipelines and would provide the same information without unexpected technical hurdles.
Claim 15 is a system claim corresponding to the method claim 3 above. Thus, claim 15 is rejected for the same reason as claim 3.
Claims 4, 5, 6, 7, 16, 17, 18, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Barr et. al (U.S. Patent Publication No. 2020/0387579), in view of Bai et. al (U.S. Patent Publication No. 2021/0081505), in view of Rawlinson et. al (U.S. Patent No. 11,357,573), in further view of Fan et. al (U.S. Patent Publication No. 2019/0295266).
Regarding claim 4, Barr, Bai, and Rawlinson disclose the method of claim 3, but fail to explicitly disclose further comprising: converting the digital representation of the 3D object into graph-structured data prior to learning the relationships between the plurality of points on the surface of the 3D object.
However, Fan discloses further comprising: converting the digital representation of the 3D object into graph-structured data prior to learning the relationships between the plurality of points on the surface of the 3D object (Fan: 206; Fig. 2)(teaches transforming a point cloud model into a nearest neighbor graph of supervoxel nodes and edges before further processing).
Barr, Bai, Rawlinson, and Fan are considered to be analogous to the claimed invention because they process point cloud data for downstream machine learning analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Barr, Bai, and Rawlinson to incorporate Fan’s teachings of transforming a point cloud model into a nearest neighbor graph before further processing. The motivation for such a combination would enable neighborhood reasoning and yield the claimed preprocessed step with predictable success.
Regarding claim 5, Barr, Bai, and Rawlinson disclose the method of claim 4, but fail to explicitly disclose wherein converting the digital representation of the 3D object into the graph-structured data comprises generating a graph representing a shape of the 3D object using the point cloud data, and wherein the generated graph comprises a plurality of nodes representing the plurality of points on the surface of the 3D object in the point cloud and a plurality of edges representing spatial relationships between the plurality of points on the surface of the 3D object.
However, Fan discloses wherein converting the digital representation of the 3D object into the graph-structured data comprises generating a graph representing a shape of the 3D object using the point cloud data (interpreted as before training, turn the point cloud model into a graph that captures the objects shape) (Fan: 206; Fig. 2) (teaches converting the point cloud into a graph that encodes the objects 3D shape), and wherein the generated graph comprises a plurality of nodes representing the plurality of points on the surface of the 3D object in the point cloud (interpreted as each point on the surface becomes a node in the graph) (Fan: 206; Fig. 2 “the supervoxels represent nodes of the graphs”) (the constructed graph includes nodes representing the 3D object - Fan’s graph definition maps point cloud elements (supervoxels) to graph nodes) and a plurality of edges representing spatial relationships between the plurality of points on the surface of the 3D object (interpreted as the graph includes edges (connections) that encode which points are neighbors) (Fan: 206; Fig. 2 “adjacent relations of the supervoxels represents edges of the graphs”)(teaches edges that capture adjacency between nodes).
Barr, Bai, Rawlinson, and Fan are considered to be analogous to the claimed invention because they process point cloud data for downstream machine learning analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Barr, Bai, and Rawlinson to incorporate Fan’s teachings of multiple nodes representing multiple points. The motivation for such a combination would enable more accurate and predictable results.
Regarding claim 6, Barr, Rawlinson, and Fan disclose the method of claim 5, using one or more shared convolutional operations [Barr: 0038 “The convolutional layer ( s ) 220 may apply convolution operations to the input, passing the result to the next layer.”] but fail to explicitly disclose wherein learning the relationships between the plurality of points comprises learning, by the GTN model, the relationships between the plurality of nodes in the generated graph using one or more shared convolutional operations.
However, Bai discloses wherein learning the relationships between the plurality of points comprises learning, by the GTN model, the relationships between the plurality of nodes in the generated graph [Bai: 0003 “a computational method for simulating the motion of elements within a multi - element system using a graph transformer neural network ( GTFF ) includes converting a molecular dynamics snapshot of the elements within the multi - element system into a graph with atoms as nodes of the graph; defining a matrix such that each column of the matrix represents a node in the graph; defining a distance matrix according to a set of relative positions of each of the atoms; iterating through the GTFF using an attention mechanism, operating on the matrix and augmented by incorporating the distance matrix, to pass hidden state from a current layer of the GTFF to a next layer of the GTFF”](teaches the graph transformer learns node relations in the generated graph).
Barr, Bai, Rawlinson, and Fan are considered to be analogous to the claimed invention because they process point cloud data for downstream machine learning analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Barr, Rawlinson, and Fan to incorporate Bai’s teachings of having the graph transformer learn node relations. The motivation for such a combination would enable more accurate and predictable results.
Regarding claim 7, Barr, Rawlinson, and Fan disclose the method of claim 6, but fail to explicitly disclose wherein learning the relationships between the plurality of nodes comprises performing, by the GTN model, multi-head attention over neighbors.
However, Bai discloses wherein learning the relationships between the plurality of nodes comprises performing, by the GTN model, multi-head attention over neighbors [Bai: 0003 “a computational method for simulating the motion of elements within a multi - element system using a graph transformer neural network ( GTFF ) includes converting a molecular dynamics snapshot of the elements within the multi - element system into a graph with atoms as nodes of the graph; defining a matrix such that each column of the matrix represents a node in the graph; defining a distance matrix according to a set of relative positions of each of the atoms; iterating through the GTFF using an attention mechanism, operating on the matrix and augmented by incorporating the distance matrix, to pass hidden state from a current layer of the GTFF to a next layer of the GTFF”](teaches learning the node relationships using an attention mechanism).
Barr, Bai, Rawlinson, and Fan are considered to be analogous to the claimed invention because they process point cloud data for downstream machine learning analysis. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Barr, Rawlinson, and Fan to incorporate Bai’s teachings of using an attention mechanism to learn node relationships. The motivation for such a combination would enable more accurate and predictable results.
Claim 16 is a system claim corresponding to the method claim 4 above. Thus, claim 16 is rejected for the same reason as claim 4.
Claim 17 is a system claim corresponding to the method claim 5 above. Thus, claim 17 is rejected for the same reason as claim 5.
Claim 18 is a system claim corresponding to the method claim 6 above. Thus, claim 18 is rejected for the same reason as claim 6.
Claim 19 is a system claim corresponding to the method claim 7 above. Thus, claim 19 is rejected for the same reason as claim 7.
Claim 12 is rejected under 35 U.S.C. 103 as being unpatentable over Barr et. al (U.S. Patent Publication No. 2020/0387579), in view of Bai et. al (U.S. Patent Publication No. 2021/0081505), in further view of P R et. al (U.S. Patent Publication No. 2022/0026925).
Regarding claim 12, Barr and Bai disclose the method of claim 1, based on the one or more predictions about the fluid properties along the surface of the 3D object [Barr: 0050 “the output may include any information of the CFD flow over time including graphical displays of the object itself, visual patterns of the flow around the object, measurements taken from the flow including velocity in one or more directions, temperature, density, pressure, turbulence”] (teaches the output (prediction) is based on the properties (direction, temperature, etc.) around the object which corresponds to the surface of the object) but fail to explicitly disclose further comprising: performing an action to mitigate one or more negative effects of the fluid properties along the surface of the 3D object.
However, P R discloses further comprising: performing an action to mitigate one or more negative effects of the fluid properties along the surface of the 3D object [P R: 0013 “the ground effect information can be prepared to adjust flight controls in a fly by wire ( autopilot system ). In autopilot embodiments, the systems and methods can determine the amount of force which is to be generated based on, for example, predicted approach speed of the aircraft, entry point into ground effect or exit point of the ground effect, and the type of ground surface. This information along with the start point of ground effect region and end point of ground effect region combined allows the system to generate appropriate commands to suppress or utilize the ground effect forces in accordance with the phase of flight of the aircraft.”] (P R’s prediction model is used to suppress or adjust (mitigate) flight controls based on information it has to mitigate negative effects).
Barr, Bai, and P R are considered to be analogous to the claimed invention because they are in the same field of machine learning flow prediction models. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Barr and Bai to incorporate P R’s teachings of mitigating negative effects using the prediction model. The motivation for such a combination would provide the benefit of performance improvements and reduce negative consequences when using the prediction model.
Conclusion
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 AHMED TAHA whose telephone number is (571)272-6805. The examiner can normally be reached 8:30 am - 5 pm, Mon - Fri. 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, XIAO WU can be reached at (571)272-7761. 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.
/AHMED TAHA/Examiner, Art Unit 2613
/XIAO M WU/Supervisory Patent Examiner, Art Unit 2613