DETAILED CORRESPONDENCE
This action is in response to the filing of the Application on 06/27/2025. This application is a Continuation from Serial # 18/056,443.
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 .
Claim Objections
Claim 11 is objected to because of the following informalities: Claim 11 depends on Claim 11. Appropriate correction is required. The Examiner will examine with the understanding that Claim 11 depends on Claim 10.
Double Patenting
The non-statutory double patenting rejection is based on a judicially created doctrine grounded in public policy (a policy reflected in the statute) to prevent the unjustified or improper time wise extension of the “right to exclude” granted by a patent and to prevent possible harassment by multiple assignees. A non-statutory double patenting rejection is appropriate where the conflicting claims are not identical, but at least one examined application claim is not patentably distinct from the reference claim(s) because the examined application claim is either anticipated by, or would have been obvious over, the reference claim(s). See, e.g., In re Berg, 140 F.3d 1428, 46 USPQ2d 1226 (Fed. Cir. 1998); In re Goodman, 11 F.3d 1046, 29 USPQ2d 2010 (Fed. Cir. 1993); In re Longi, 759 F.2d 887, 225 USPQ 645 (Fed. Cir. 1985); In re Van Ornum, 686 F.2d 937, 214 USPQ 761 (CCPA 1982); In re Vogel, 422 F.2d 438, 164 USPQ 619 (CCPA 1970); In re Thorington, 418 F.2d 528, 163 USPQ 644 (CCPA 1969).
A timely filed terminal disclaimer in compliance with 37 CFR 1.321(c) or 1.321(d) may be used to overcome an actual or provisional rejection based on non-statutory double patenting provided the reference application or patent either is shown to be commonly owned with the examined application, or claims an invention made as a result of activities undertaken within the scope of a joint research agreement. See MPEP § 717.02 for applications subject to examination under the first inventor to file provisions of the AIA as explained in MPEP § 2159. See MPEP § 2146 et seq. for applications not subject to examination under the first inventor to file provisions of the AIA .
A terminal disclaimer must be signed in compliance with 37 CFR 1.321(b). The USPTO Internet website contains terminal disclaimer forms which may be used. Please visit www.uspto.gov/patent/patents-forms. The filing date of the application in which the form is filed determines what form (e.g., PTO/SB/25, PTO/SB/26, PTO/AIA /25, or PTO/AIA /26) should be used. A web-based eTerminal Disclaimer may be filled out completely online using web-screens. An eTerminal Disclaimer that meets all requirements is auto-processed and approved immediately upon submission. For more information about eTerminal Disclaimers, refer to www.uspto.gov/patents/process/file/efs/guidance/eTD-info-I.jsp.
Claims 8, 10, 11 and 17 of the instant application are rejected on the ground of non-statutory double patenting as being unpatentable over claims 1, 4, 2 and 11 of U.S. Patent No. 12,358,138. Although the claims at issue are not identical, they are not patentably distinct from each other because the claims generally cover, machine learning logic trained and implemented to learn from or adjust based on one or more operational characteristics associated with the manufacturing robot environment.
Claim limitations of the instant claims can be mapped claim limitations of US. Patent No. US 12,358,138 See Chart below:
INSTANT CASE
PATENT 12,358,138
Claim 8
A computer-implemented method operating a welding robot, the computer-
implemented method comprising:
estimating, using machine learning logic, a first weld profile based on one or more weld parameters, a reference weld profile, a location of a seam, a trajectory, or a
combination thereof;
generating one or more updated weld parameters based on the one or more weld parameters, a difference profile generated based on a comparison based on the first weld profile, and the reference weld profile;
determining a second weld profile based on sensor data received from one or more
sensors, the sensor data associated with a weld formed by the welding robot
based on the one or more updated welding parameters; and
updating the machine learning logic based on the first weld profile, the second weld
profile, the difference profile, the trajectory, or a combination thereof.
10. The computer-implemented method of claim 8, further comprising:
receiving data that includes or indicates the one or more weld parameters, the reference weld profile, the location of the seam, the trajectory, the reference weld profile, or a combination thereof.
11.The computer-implemented method of claim 11, wherein receiving includes:
receiving, from the one or more sensors, image data of a part that is to be welded; and determining the location based on the image data.
Claim 17
A computer-implemented method for operating a welding robot, the computer-
implemented method comprising:
generating a first difference profile based on a reference weld profile and a first weld profile of a first weld formed by the welding robot;
generating at least one updated welding parameter based on the first difference profile, one or more input parameters, or a combination thereof;
determining a second weld profile of a second weld formed by the welding robot based on the at least one updated welding parameter, the second weld profiled
determined based on first sensor data received from one or more sensors;
generating a second difference profile based on the reference weld profile and the second weld profile; and
updating machine learning logic based on the first difference profile, the one or more input parameters, the second difference profile, and the at least one updated welding parameter.
Claim 1
A computer-implemented method of operating a welding robot, the computer-implemented method comprising:
simulating, using machine learning logic, a difference profile generated based on a comparison of a first weld profile and a reference weld profile, wherein the first weld profile comprises a geometric representation of a shape of a first weld,
wherein the reference weld profile comprises a geometric representation of a shape of a reference weld; determining one or more derivatives of the difference profile with respect to one or more weld parameters; generating one or more updated weld parameters based upon the one or more derivatives;
transmitting, to the welding robot, control information to instruct the welding robot to perform a weld operation on a seam to form a second weld, wherein the control information indicates the one or more updated weld parameters; and performing, by the welding robot, the weld operation on the seam using the control information.
4. The computer-implemented method of claim 1, wherein the reference weld is based on sensor data received from one or more sensors.
2. The computer-implemented method of claim 1, wherein the difference profile comprises differences in coordinates or images of the first weld profile and the reference weld profile.
Claim 11
A computer-implemented method for operating a welding robot, the computer-implemented method comprising:
determining a first weld profile of a first weld formed by the welding robot based on first sensor data received from one or more sensors,
wherein the first weld profile comprises a geometric representation of a shape of the first weld; generating, using machine learning logic, a first difference profile based on the first weld profile and a reference weld profile comprising a geometric representation of a shape of a reference weld, and at least one updated welding parameter based on the first difference profile;
transmitting, to the welding robot, control information to instruct the welding robot to perform a weld operation to form a second weld, wherein the control information indicates the at least one updated welding parameter; performing, by the welding robot, the weld operation using the control information;
determining a second weld profile of the second weld based on second sensor data received from the one or more sensors, wherein the second weld profile comprises a geometric representation of a shape of the second weld; and generating,
using the machine learning logic, a second difference profile based on the second weld profile and the reference weld profile.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1 – 10, 12 and 14 – 16 are rejected under 35 U.S.C. 103 as being unpatentable over Pinsker (EP3812105A1) (as submitted by Applicant) in view of Pascanu (US 20190232489).
Claim 1, Pinsker discloses a computer-implemented. method for operating a manufacturing robot, the computer-implemented. method comprising: receiving, from one or more sensors, information regarding a first manufacturing output, the first manufacturing output generated by the manufacturing robot based on a first profile, a first trajectory, a first one or more manufacturing parameters, or a combination thereof [see p0004, p0008 and p0048 - The method comprises obtaining operating data of the welding robot during welding and obtaining image data of a welded workpiece. The image data is processed to determine parameters of the welded workpiece and the parameters are processed to determine a quality of the welded workpiece by rating the parameters using a scoring function. The results of the scoring function are input into a machine learning model; The parameters of the welded workpiece preferably comprise the type of welding and/or geometric parameters];
processing, by a processor, the information regarding the first manufacturing output
and the first profile, the first trajectory, the first one or more manufacturing parameters, or a combination thereof, to generate a second trajectory or a second one or more manufacturing parameters; [see p0041, p0049 – Figure 1, S103, the image data is processed to determine parameters of the welded workpiece. The parameters preferably refer to the welding seam and comprise at least one of width, penetration depth, notches, pores and time consumed for the welding process. In S203 (layer 2), a scenario classification is performed. Furthermore, simulated data, i.e. data generated by a virtual welding process, may also be added to the scenario classification data. The scenario classification data comprises information on the welding seam such as width, penetration depth, notches, pores (on the surface as well as within the weld) and the time consumed for the welding process. A single parameter or a combination of at least two of the parameters listed above may be used. Other parameters may also be appropriate]; and
Pinsker does not specifically teach initiating generation of a second manufacturing output by the manufacturing robot based on the second trajectory, the second one or more manufacturing parameters, or a combination thereof.
However, Pascanu discloses training a robot-trained DNN by receiving as input (i) a layer output generated by a layer preceding the output layer of the first robot-trained DNN, and (ii) a layer output generated by a layer preceding an output layer of the simulation-trained DNN; applying a first set of parameters to the layer output generated by a layer preceding the output layer the first robot-trained DNN; and applying a second set of parameters to the layer output generated by a layer preceding the output layer of the simulation-trained DNN [See summary of invention and Fig 2 teaching that the first layer of information is used to obtain the second output, and so on. No matter how many iterations is made, the output always includes the prior input].
Further disclosing, the one or more layers following the first layer 216 are each configured to process the respective input to generate a respective layer output. The layer output is then provided to the next layer as input, or if there are no layers following the current layer (i.e., the current layer is the output layer 220), as a second policy output 230. The second policy output 230 defines an action to be performed by the robotic agent to perform the second robotic task [see Fig 2 and p0048 – p0055].
Therefore, it would have been obvious before the effective date of the claimed invention to one of ordinary skill in the art to modify the device in Pinsker to include initiating generation of a second manufacturing output by the manufacturing robot based on the second trajectory, the second one or more manufacturing parameters, or a combination thereof. as suggested and taught by Pascanu, with a reasonable expectation of success, for the purpose of providing a current state of a real-world environment being interacted with by a robotic agent to perform a robotic task and to process the observation to generate a policy output that defines an action to be performed by the robotic agent in response to the inputs, thereby eliminating the need to process an output without new inputs and data.
Claim 2, Pinsker discloses the computer-implemented method of claim 1, further comprising: receiving the first profile, the first trajectory, the first one or more manufacturing parameters, or a combination thereof [see p0004, p0008 – method comprises obtaining operating data of the welding robot during welding and obtaining image data of a welded workpiece. The image data is processed to determine parameters of the welded workpiece and the parameters are processed to determine a quality of the welded workpiece by rating the parameters using a scoring function; The parameters of the welded workpiece preferably comprise the type of welding and/or geometric parameters].
Claim 3, Pinsker discloses the computer-implemented method of claim 2, but not specifically further comprising: generating first control information based on the first profile, the first trajectory, the first one or more manufacturing parameters, or a combination thereof.
However, Pinsker does disclose in S206 (see Fig 2), the results of the scoring function are input into a machine learning algorithm (hereinafter also referred to as "RLAgent"). The machine learning algorithm may comprise an initial training set obtained under laboratory conditions or by conducting the welding process in the production site. The machine learning algorithm may for example be a reinforcement learning (RL) algorithm. Depending on the result of the scoring function, a machine learning agent modifies the robot parameters, i.e. the operating data, such as, according to a preferred embodiment, movement trajectories, atmosphere, temperature, CO2 content, power and/or angle with respect to the workpiece. Movement trajectories may be defined by translatory movements as well as rotational movements. After classification of the type weld, the quality of the weld may subsequently be classified. Therefore, at least one of the parameters indicated above may be used. model may also be capable of detecting the type of weld by analyzing the image data. In S207, the modified operating data is used to control the welding robot. The process with the modified parameters is then again monitored and the obtained image and movement data is fed into the machine learning algorithm again to further improve the welding results [see p0053 – p0056].
It would have been obvious before the effective date of the claimed invention to one of ordinary skill in the art to modify the device in Pinsker, to include generating first control information based on the first profile, the first trajectory, the first one or more manufacturing parameters, or a combination thereof, with a reasonable expectation of success, for the purpose of providing more quality and aims at optimizing the robot parameters and hence the welding process.
Claim 4, Pinsker discloses the computer-implemented method of claim 3, wherein the first control information includes one or more weld instructions [see p0058 - Scoring function is then input into a machine learning agent (S305). The function is then used to modify the instructions for the welding environment (S306). The new instructions are then used to control and instruct the welding robot (S307)].
Claim 5, Pinsker discloses the computer-implemented method of claim 3, further comprising: transmitting the first control information to the manufacturing robot to initiate a first manufacturing task by the manufacturing robot on a first part positioned in a
manufacturing workspace to generate the first manufacturing output [see p0022, p0058 - Simulated data generated by a virtual welding process may preferably also be used in S302 to obtain clean data (scenario classification according to Fig. 2). The data is then used for tracking in S303b, i.e. record states, actors and rewards, and for the classification of a weld in S303a preferably using computer vision. The computer vision may deploy artificial intelligence (Al). The results are then fed into a scoring function (S304a). Scoring function is then input into a machine learning agent (S305). In addition, the tracking data might also input into the machine learning agent (S304b). The agent comprises an initial model and uses the newly fed data for an approximation of an optimal action value function. The function is then used to modify the instructions for the welding environment (S306). The new instructions are then used to control and instruct the welding robot (S307)].
Claim 6, Pinsker discloses the computer-implemented method of claim 1, further comprising: generating the second trajectory, the second one or more manufacturing parameters, or a combination thereof.
However, Pinsker does disclose in S206 (see Fig 2), the results of the scoring function are input into a machine learning algorithm (hereinafter also referred to as "RLAgent"). The machine learning algorithm may comprise an initial training set obtained under laboratory conditions or by conducting the welding process in the production site. The machine learning algorithm may for example be a reinforcement learning (RL) algorithm. Depending on the result of the scoring function, a machine learning agent modifies the robot parameters, i.e. the operating data, such as, according to a preferred embodiment, movement trajectories, atmosphere, temperature, CO2 content, power and/or angle with respect to the workpiece. Movement trajectories may be defined by translatory movements as well as rotational movements. After classification of the type weld, the quality of the weld may subsequently be classified. Therefore, at least one of the parameters indicated above may be used. model may also be capable of detecting the type of weld by analyzing the image data. In S207, the modified operating data is used to control the welding robot. The process with the modified parameters is then again monitored and the obtained image and movement data is fed into the machine learning algorithm again to further improve the welding results [see p0053 – p0056].
Therefore, it would have been obvious before the effective date of the claimed invention to one of ordinary skill in the art to modify the device in Pinsker, to include generating the second trajectory, the second one or more manufacturing parameters, or a combination thereof, with a reasonable expectation of success, for the purpose of providing more quality and aims at optimizing the robot parameters and hence the welding process.
Claim 7, Pinsker discloses the computer-implemented method of claim 1, further comprising: generating second control information based on the second trajectory, the second one or more manufacturing parameters, or a combination thereof; and transmitting the second control information to the manufacturing robot to initiate a second manufacturing task by the manufacturing robot on a second part positioned in a manufacturing workspace to generate the second manufacturing output [see p0022, p0058 - Simulated data generated by a virtual welding process may preferably also be used in S302 to obtain clean data (scenario classification according to Fig. 2). The data is then used for tracking in S303b, i.e. record states, actors and rewards, and for the classification of a weld in S303a preferably using computer vision. The computer vision may deploy artificial intelligence (Al). The results are then fed into a scoring function (S304a). Figure 2, S103, the image data is processed to determine parameters of the welded workpiece. The parameters preferably refer to the welding seam and comprise at least one of width, penetration depth, notches, pores and time consumed for the welding process].
Claim 8, Pinsker discloses a computer-implemented method operating a welding robot, the computer-implemented method comprising: estimating, using machine learning logic, a first weld profile based on one or more weld parameters, a reference weld profile, a location of a seam, a trajectory, or a combination thereof [see p0004, p0041 – p0049, p0076 - a computer-implemented method for controlling welding performed by a welding robot. The method comprises obtaining operating data of the welding robot during welding and obtaining image data of a welded workpiece; performing the welding and taking an image of the resulting weld. The quality of the welding will be estimated or predicted by passing the image to a trained Deep Learning neural network. The quality can be a set of parameters as mentioned above, for example, welding width, penetration depth and existence of notches or pores; Figures 1, S103, the image data is processed to determine parameters of the welded workpiece. The parameters preferably refer to the welding seam and comprise at least one of width, penetration depth, notches, pores and time consumed for the welding process];
generating one or more updated weld parameters based on the one or more weld parameters, a difference profile generated based on a comparison based on the first weld profile, and the reference weld profile [p0044, S106 the scoring function applied to the obtained operating data to modify the operating data. A preferred embodiment of a generalized weighted function will be described below. The determination of the quality of the welded workpiece is preferably determined based on the occurrence and/or quantity of defects of the welding. In S205, the classifiers are then fed into a scoring function (layer 4). The scoring function may be a generalized weighted scoring function. Generalized weights allow implementing not acceptable (hereinafter also referred to as "no go") and (strongly) favorable results in contrast to a fixed weighted average. A notch in the weld for example may not be rated and weighted accordingly but directly lead to rejection of the workpiece].
Pinsker does not specifically teach determining a second weld profile based on sensor data received from one or more sensors, the sensor data associated with a weld formed by the welding robot based on the one or more updated welding parameters; and updating the machine learning logic based on the first weld profile, the second weld profile, the difference profile, the trajectory, or a combination thereof.
However, Pinsker does disclose using the scoring function, parameters of the welded workpiece preferably comprise the type of welding and/or geometric parameters.
More preferably, the parameters of the welded workpiece comprise at least one of width of the weld, penetration depth of the weld, notches on the surface and/or in the weld, pores on the surface or in the weld and time consumed for the welding process. In other words, at least one of the following is used as a classifier for the machine learning model to classify the score of the welding process: width of the weld, penetration depth of the weld, notches on the surface and/or in the weld, pores on the surface or in the weld and time consumed for the welding process. The sensors may be used in order to monitor parameters of the robotic arm in addition to controlling of the arm itself. ” [see p0074].
The process with the modified parameters is then again monitored and the obtained image and movement data is fed into the machine learning algorithm again to further improve the welding results (here second weld) [see p0053 – p0056].
It would have been obvious before the effective date of the claimed invention to one of ordinary skill in the art to modify the device in Pinsker, to include determining a second weld profile based on sensor data received from one or more sensors, the sensor data associated with a weld formed by the welding robot based on the one or more updated welding parameters; and updating the machine learning logic based on the first weld profile, the second weld profile, the difference profile, the trajectory, or a combination thereof, with a reasonable expectation of success, for the purpose of providing more quality and aims at optimizing the robot parameters and hence the welding process.
Claim 9, Pinsker discloses the computer-implemented. method of claim 8, wherein the machine learning logic is configured to be executed by a process to perform one or more operations [see p0004 - results of the scoring function are input into a machine learning model. The results of the scoring function are, using the machine learning model, applied to the obtained operating data to modify said operating data to improve the welding. The modified operating data is used to control the welding robot].
Claim 10, Pinsker discloses the computer-implemented method of claim 8, further comprising: receiving data that includes or indicates the one or more weld parameters, the reference weld profile, the location of the seam, the trajectory, the reference weld profile, or a combination thereof [see p0041, p0049 – Figure 2, n S103, the image data is processed to determine parameters of the welded workpiece. The parameters preferably refer to the welding seam and comprise at least one of width, penetration depth, notches, pores and time consumed for the welding process. In S203 (layer 2), a scenario classification is performed. Furthermore, simulated data, i.e. data generated by a virtual welding process, may also be added to the scenario classification data. The scenario classification data comprises information on the welding seam such as width, penetration depth, notches, pores (on the surface as well as within the weld) and the time consumed for the welding process. A single parameter or a combination of at least two of the parameters listed above may be used. Other parameters may also be appropriate].
Claim 12, Pinsker discloses the computer-implemented method of claim 12, wherein receiving include: receiving, via user interface, at least one weld parameter of the one or more weld parameters; and determining, using a path planning logic, the trajectory that a robot is configured to follow to weld the seam [see p0004, p0008, p0049 - The method comprises obtaining operating data of the welding robot during welding and obtaining image data of a welded workpiece. The image data is processed to determine parameters of the welded workpiece and the parameters are processed to determine a quality of the welded workpiece by rating the parameters using a scoring function. The results of the scoring function are input into a machine learning model; The parameters of the welded workpiece preferably comprise the type of welding and/or geometric parameters; The scenario classification data comprises information on the welding seam such as width, penetration depth, notches, pores (on the surface as well as within the weld) and the time consumed for the welding process. A single parameter or a combination of at least two of the parameters listed above may be used. Other parameters may also be appropriate].
Claim 14, Pinsker discloses the computer-implemented. method of claim 8, further comprising: transmitting, to the weld robot, control information to instruct the weld robot to perform a weld operation on the scam to form a weld, where the control information
indicates the one or more updated welding parameters; and receiving, from the one or more sensors, the sensor data [see p0022, p0058 - Simulated data generated by a virtual welding process may preferably also be used in S302 to obtain clean data (scenario classification according to Fig. 2). The data is then used for tracking in S303b, i.e. record states, actors and rewards, and for the classification of a weld in S303a preferably using computer vision. The computer vision may deploy artificial intelligence (Al). The results are then fed into a scoring function (S304a). Figure 2, S103, the image data is processed to determine parameters of the welded workpiece. The parameters preferably refer to the welding seam and comprise at least one of width, penetration depth, notches, pores and time consumed for the welding process; The agent comprises an initial model and uses the newly fed data for an approximation of an optimal action value function. The function is then used to modify the instructions for the welding environment (S306). The new instructions are then used to control and instruct the welding robot (S307)].
Claim 15, Pinsker discloses the computer-implemented method of claim 8, further comprising providing, to the machine learning logic, learned operational characteristics specific to the welding robot [see p0004 – Pinsker discloses obtaining operating data of the welding robot during welding and obtaining image data of a welded workpiece. The image data is processed to determine parameters of the welded workpiece and the parameters are processed to determine a quality of the welded workpiece by rating the parameters using a scoring function. The results of the scoring function are input into a machine learning model. The results of the scoring function are, using the machine learning model, applied to the obtained operating data to modify said operating data to improve the welding. The modified operating data is used to control the welding robot].
Claim 16, Pinsker discloses the computer-implemented method of claim 8, does not specifically teach further comprising providing, to the machine learning logic, learned operational characteristics of multiple welding robots, the multiple welding robots including the welding robot.
However, Pinsker does teach the results of the scoring function are input into a machine learning model. The results of the scoring function are, using the machine learning model, applied to the obtained operating data to modify said operating data to improve the welding. The modified operating data is used to control the welding robot [see p0004].
Therefore, it would have been obvious to one of ordinary skill in the art at the time the invention was made to control more than one robot with the machine learning logic, since a mere duplication of essential working parts of device involves only routine skill in the art. Since Pinsker can control the machine learning algorithm of one robot, it certainly can control many or multiple robots.
Claim(s) 11 is rejected under 35 U.S.C. 103 as being unpatentable over Pinsker (EP3812105A1) (as submitted by Applicant) in view of Pascanu (US 20190232489) and Daniel (US 20140042136).
Claim 11, Pinsker discloses the computer-implemented method of claim 11, wherein receiving includes: receiving, from the one or more sensors, image data of a part that is to be welded; [p0004, p0007- The method comprises obtaining operating data of the welding robot during welding and obtaining image data of a welded workpiece. The image data is processed to determine parameters of the welded workpiece and the parameters are processed to determine a quality of the welded workpiece by rating the parameters using a scoring function];
Pinsker does not specifically teach and determining the location based on the image data.
However, Daniel discloses a welding system is provided that includes a welding job sequencer component that is configured to identify a welding sequence for a welding work cell. Further disclosing, the quality assurance collection can be a data capture of at least one of welding equipment (e.g., settings, physical location, consumable amount, and the like), the operator (e.g., physical location of person, location of arms, location of hands, and the like), or a weld (e.g., dimensions, size, location of weld, among others). It is to appreciate that the data collection can be video, audio, images, settings, welding equipment settings, physical location of the operator, movements from the operator, and the like [see p0103].
Therefore, it would have been obvious before the effective date of the claimed invention to one of ordinary skill in the art to modify the device in Pinsker, to include and determining the location based on the image data, as suggested and taught by Daniel, with a reasonable expectation of success, for the purpose of providing more quality weld and aims at optimizing the robot parameters and hence the welding process, increasing productivity and efficiency.
Claim(s) 13 is rejected under 35 U.S.C. 103 as being unpatentable over Pinsker (EP3812105A1) (as submitted by Applicant) in view of Pascanu (US 20190232489) and Ma (US 20120234812).
Claim 13, Pinsker discloses the computer-implemented method of claim 8, but is silent to further comprising: performing the comparison based on the first weld profile and the reference weld profile; and generating the difference profile based on the comparison, and wherein generating the one or more updated weld parameters includes processing the difference profile.
However, Ma discloses an apparatus to maintain welding current to compensate for deterioration of a welding contact tip includes a monitor for monitoring at least one of welding current and welding voltage during welding production, a comparator for comparing the at least one of the welding current and the welding voltage to a reference value for a given set of parameters, and a controller for automatically adjusting welding parameters in real time to increase energy output in response to the comparison of the comparator so that consistent energy is consumed across the welding arc. Further teaching, See Fig 4, actual welding parameters such as the welding current and welding voltage are monitored in real-time at S106 by the controller or associated electronics during operation of the welding cell. The measured welding parameter data is analyzed in real-time. For example, contact tip deterioration may be quantified by comparing the measured values of welding parameters such as the average welding current (I.sub.t) and the standard deviation (.sigma..sub.t) of the welding current acquired at S108 with stored reference values (I.sub.0 and .sigma..sub.0) for a new contact tip under a given set of parameter [see Fig 4 and p0007 -p0011, p0029, and p0031].
Therefore, it would have been obvious before the effective date of the claimed invention to one of ordinary skill in the art to modify the device in Pinsker, to include performing the comparison based on the first weld profile and the reference weld profile; and generating the difference profile based on the comparison, and wherein generating the one or more updated weld parameters includes processing the difference profile, as suggested and taught by Ma, with a reasonable expectation of success, for the purpose of providing adjusting one or more welding parameters in real time to increase energy output in response to the comparison so that consistent energy is consumed across the welding arc, whereby the quality of the welding production is maintained as the welding contact tip is consumed.
Claim(s) 17 – 20 are rejected under 35 U.S.C. 103 as being unpatentable over Pinsker (EP3812105A1) (as submitted by Applicant) in view Ma (US 20120234812).
Claim 17, Pinsker discloses a computer-implemented method for operating a welding robot, the computer- implemented method comprising [see p0004 - a computer-implemented method for controlling welding performed by a welding robot. The method comprises obtaining operating data of the welding robot during welding and obtaining image data of a welded workpiece]:
Pinsker does not specifically teach generating a first difference profile based on a reference weld profile and a first weld profile of a first weld formed by the welding robot; generating at least one updated welding parameter based on the first difference profile, one or more input parameters, or a combination thereof; generating a second difference profile based on the reference weld profile and the second weld profile; updating machine learning logic based on the first difference profile, the one or more input parameters, the second difference profile, and the at least one updated welding parameter; determining a second weld profile of a second weld formed by the welding robot based on the at least one updated welding parameter, the second weld profiled determined based on first sensor data received from one or more sensors.
However, Ma discloses an apparatus to maintain welding current to compensate for deterioration of a welding contact tip includes a monitor for monitoring at least one of welding current and welding voltage during welding production, a comparator for comparing the at least one of the welding current and the welding voltage to a reference value for a given set of parameters, and a controller for automatically adjusting welding parameters in real time to increase energy output in response to the comparison of the comparator so that consistent energy is consumed across the welding arc. Further teaching, See Fig 4, actual welding parameters such as the welding current and welding voltage are monitored in real-time at S106 by the controller or associated electronics during operation of the welding cell. The measured welding parameter data is analyzed in real-time. For example, contact tip deterioration may be quantified by comparing the measured values of welding parameters such as the average welding current (I.sub.t) and the standard deviation (.sigma..sub.t) of the welding current acquired at S108 with stored reference values (I.sub.0 and .sigma..sub.0) for a new contact tip under a given set of parameter [see Fig 4 and p0007 - p0011, p0029, and p0031].
Therefore, it would have been obvious before the effective date of the claimed invention to one of ordinary skill in the art to modify the device in Pinsker, to include generating a first difference profile based on a reference weld profile and a first weld profile of a first weld formed by the welding robot; generating at least one updated welding parameter based on the first difference profile, one or more input parameters, or a combination thereof; generating a second difference profile based on the reference weld profile and the second weld profile; updating machine learning logic based on the first difference profile, the one or more input parameters, the second difference profile, and the at least one updated welding parameter; determining a second weld profile of a second weld formed by the welding robot based on the at least one updated welding parameter, the second weld profiled determined based on first sensor data received from one or more sensors, as suggested and taught by Ma, with a reasonable expectation of success, for the purpose of providing
Claim 18, Pinsker discloses the computer-implemented method of claim 17, further comprising: receiving first information that indicates the first weld profile, the reference weld profile, a location of a seam to be welded, the one or more input parameters, a planned trajectory, or a combination thereof; [see p0041, p0049 – Figure 2, n S103, the image data is processed to determine parameters of the welded workpiece. The parameters preferably refer to the welding seam and comprise at least one of width, penetration depth, notches, pores and time consumed for the welding process. In S203 (layer 2), a scenario classification is performed. Furthermore, simulated data, i.e. data generated by a virtual welding process, may also be added to the scenario classification data. The scenario classification data comprises information on the welding seam such as width, penetration depth, notches, pores (on the surface as well as within the weld) and the time consumed for the welding process. A single parameter or a combination of at least two of the parameters listed above may be used. Other parameters may also be appropriate].
and transmitting, to the weld robot, control information to instruct the weld robot to perform a weld operation to form the second weld, where the control information indicates the at least one updated welding parameter [see p0022, p0058 - Simulated data generated by a virtual welding process may preferably also be used in S302 to obtain clean data (scenario classification according to Fig. 2). The data is then used for tracking in S303b, i.e. record states, actors and rewards, and for the classification of a weld in S303a preferably using computer vision. The computer vision may deploy artificial intelligence (Al). The results are then fed into a scoring function (S304a). Figure 2, S103, the image data is processed to determine parameters of the welded workpiece. The parameters preferably refer to the welding seam and comprise at least one of width, penetration depth, notches, pores and time consumed for the welding process; and teaches a simulated weld by scoring function is a first difference profile. The results are then fed into a scoring function (S304a). Scoring function is then input into a machine learning agent (S305). In addition, the tracking data might also input into the machine learning agent (S304b). The agent comprises an initial model and uses the newly fed data for an approximation of an optimal action value function. The function is then used to modify the instructions for the welding environment (S306). The new instructions are then used to control and instruct the welding robot (S307). Hence, a loop is created which allows optimizing the welding parameters by machine learning];
Claim 19, Pinsker discloses the computer-implemented method of claim 18, further comprising receiving the first sensor data from the one or more sensors, the first sensor data generated based on the second weld [see p0009 - The parameters of the welded workpiece preferably comprise information on the occurrence of defects of welding, particularly on welding splash, arc strike, cold cracking, crater crack, hat crack, underbead crack, longitudinal crack, reheat cracking, root and toe cracks, transverse cracks and/or inclusions. The parameters may also comprise temperature or temperature changes of the workpiece and other parameters that may be measured by optical and/or sensoric means. The new result of the scoring function is applied, using the machine learning model, to the newly obtained operating data to remodify the new operating data and the remodified new operating data are used to control the welding robot. Thus, an iterative loop is preferably created].
Claim 20, Pinsker discloses the computer-implemented. method of claim 17, further comprising, prior to generating the first difference profile: estimating, using the machine learning logic, a third weld profile based on one or more weld parameters, the reference weld profile, a location of a seam to be welded, a planned trajectory, or a combination there of; [see p0041 – p0049, p0076 -performing the welding and taking an image of the resulting weld. The quality of the welding will be estimated or predicted by passing the image to a trained Deep Learning neural network. The quality can be a set of parameters as mentioned above, for example, welding width, penetration depth and existence of notches or pores; Figure 2, S103, the image data is processed to determine parameters of the welded workpiece. The parameters preferably refer to the welding seam and comprise at least one of width, penetration depth, notches, pores and time consumed for the welding process];
generating one or more updated weld parameters based on the one or more weld parameters, a third difference profiled generated based on a comparison based on the third weld profile and the reference weld profile; [p0042, p0051 – p0052 - In S104, the parameters are processed to determine a quality of the welded workpiece by rating the parameters using a scoring function. The scoring function preferably is a generalized weighted function. A preferred embodiment of a generalized weighted function will be described below. The determination of the quality of the welded workpiece is preferably determined based on the occurrence and/or quantity of defects of the welding. In S205, the classifiers are then fed into a scoring function (layer 4). The scoring function may be a generalized weighted scoring function. Generalized weights allow implementing not acceptable (hereinafter also referred to as "no go") and (strongly) favorable results in contrast to a fixed weighted average. A notch in the weld for example may not be rated and weighted accordingly but directly lead to rejection of the workpiece];
Pinsker does not specifically teach transmitting, to the weld robot, control information to instruct the weld robot to perform a weld operation to form a third weld, where the control information indicates the one or more updated welding parameters; and determining the first weld profile based on second sensor data received from the one or more sensors, the second sensor data associated with the third weld formed by the welding robot based on the one or more updated welding parameters.
However, Pinsker does disclose using the scoring function, parameters of the welded workpiece preferably comprise the type of welding and/or geometric parameters.
More preferably, the parameters of the welded workpiece comprise at least one of width of the weld, penetration depth of the weld, notches on the surface and/or in the weld, pores on the surface or in the weld and time consumed for the welding process. In other words, at least one of the following is used as a classifier for the machine learning model to classify the score of the welding process: width of the weld, penetration depth of the weld, notches on the surface and/or in the weld, pores on the surface or in the weld and time consumed for the welding process.
The parameters of the welded workpiece preferably comprise information on the occurrence of defects of welding, particularly on welding splash, arc strike, cold cracking, crater crack, hat crack, underbead crack, longitudinal crack, reheat cracking, root and toe cracks, transverse cracks and/or inclusions. The parameters may also comprise temperature or temperature changes of the workpiece and other parameters that may be measured by optical and/or sensoric means. The determination of the quality of the welded workpiece is preferably determined based on the occurrence and/or quantity of defects of the welding. The new result of the scoring function is inputted into the machine learning model. The new result of the scoring function is applied, using the machine learning model, to the newly obtained operating data to remodify the new operating data and the remodified new operating data are used to control the welding robot. Thus, an iterative loop is preferably created [see p0004 – p0009].
The process with the modified parameters is then again monitored and the obtained image and movement data is fed into the machine learning algorithm again to further improve the welding results (here second weld) [see p0053 – p0056].
It would have been obvious before the effective date of the claimed invention to one of ordinary skill in the art to modify the device in Pinsker, to include transmitting, to the weld robot, control information to instruct the weld robot to perform a weld operation to form a third weld, where the control information indicates the one or more updated welding parameters; and determining the first weld profile based on second sensor data received from the one or more sensors, the second sensor data associated with the third weld formed by the welding robot based on the one or more updated welding parameters, with a reasonable expectation of success, for the purpose of providing more quality and aims at optimizing the robot parameters and hence the welding process.
Overall for Claim 20, the Examiner would like to note that sense Pinsker teaches, either implicitly or explicitly a “third weld”, “a third parameter”, etc. it remains true that duplication whether in method steps or in multiple parts, it is obvious to one having ordinary skill in the art, to include an “x number of weld”, “many welding robots” and a controller and/or algorithm that is used for such a task, is a mere duplication of the essential working aspects of the device in Pinsker, is only routine skill in the art.
Conclusion
The examiner has pointed out particular references contained in the prior art of record in the body of this action for the convenience of the applicant. Although the specified citations are representative of the teachings in the art and are applied to the specific limitations within the individual claim, other passages and figures may apply as well. Applicant should consider the entire prior art as applicable as to the limitations of the claims. It is respectfully requested from the applicant, in preparing the response, to consider fully the entire references as potentially teaching all or part of the claimed invention, as well as the context of the passage as taught by the prior art or disclosed by the examiner.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to RENEE LAROSE whose telephone number is (313)446-4856. The examiner can normally be reached on Monday - Friday 8:30am - 5:00pm EST.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Abby Lin can be reached on (571) 270-3976. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/Renee LaRose/Examiner, Art Unit 3657
/SOHANA TANJU KHAYER/Primary Examiner, Art Unit 3657
Prosecution Insights