DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 1, 3, 5, 7, 14, 15, 17, and 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Watanabe (US 20020173878 A1) in view of Chen (IDS: “Evolutionary Robot Calibration and Nonlinear Compensation Methodology Based on GA-DNN and an Extra Compliance Error Model”) and Pickett (US 20230000565 A1).
Regarding Claim 1,
Watanabe teaches
A (“a robot controller capable of operating an articulated flexible arm robot at a high speed with a high accuracy by correcting bendings produced in the robot.” See at least [0006])
a robotic arm comprising a plurality of links and joints; a (“a robot controller for a robot whose respective links are driven by respective actuators and which executes a job by pressing a robot hand distal end against an object includes a sensor for detecting a reaction force produced when the robot hand distal end is pressed against the object; bending calculation means for determining the bendings produced in the respective joints at the target position/attitude of the robot hand distal end based on a teaching program;” See at least [0009])
obtain, based on kinematic data from the robotic arm, a pose of the (“a rotational angle .theta.' is determined from the target position/attitude x read at step S1 by executing inverse transformation using the corrected D-H parameters .alpha.s, .beta.s, and .theta.0s where no twist is produced, and the rotational angle .theta.' is stored as a rotational angle .theta.st at the beginning of correction (step S6).” See at least [0041])
obtain, from the sensing system, the interaction force; (“The torques Tx, Ty, and Tz about the axes of the respective joints are determined using the Newton-Euler method based on the angles and actual rotational speeds of the respective joints which have been read (step S3).” See at least [0040]; “However, the torques Tx, Ty, and Tz about the respective axes may be detected using a force sensor or a torque sensor. In this case, the processings at steps S2 and S3 shown in FIG. 5 are executed using the forces or torques obtained by these sensors.” See at least [0061])
input the pose of the (“a position/attitude x' is determined from the rotational angle .theta.' determined at step S6 by executing forward transformation using the corrected D-H parameters .alpha., .beta., and .theta.0 determined at step S5 (step S8). Then, the target position/attitude x read at step S1 is subtracted from the thus determined position/attitude x' to thereby calculate the difference .DELTA.x therebetween (step S9).” See at least [0042], wherein the rotational angle .theta.' is calculated based on the pose and torque (steps S4, S5, S6) and the difference .DELTA.x is the deflection.)
and control the robotic arm and/or the (“Next, it is determined whether or not the absolute value of the difference between the final rotational angle .theta. and the rotational angle .theta.st at the beginning of correction, determined at step S6, is equal to or larger than a set threshold value .theta.max (step S16). When the absolute value is equal to or larger than the threshold value .theta.max, alarm is issued and alarm processing is executed to abruptly stop the robot by applying brake thereto and outputting speed reduction torque commands to the servo motors of the respective axes (step S18). In contrast, when the absolute value is less than the set value .theta. max, a rotational angle j on the teaching pendant (rotational angle to be taught and commanded) is determined by subtracting the initial rotational angle .theta.0, determined at step S5, from the final rotational angle .theta. and stored (step S17).” See at least [0047]; “the offset of the distal end position of the robot hand made by bending due to external force is also compensated.” See at least [0062])
Watanabe does not explicitly teach, but Chen teaches
estimate a deflection of the robotic arm based on an output of the machine learning model; (“To realise the prediction and compensation of the robot positioning error, the output of the DNN is the predicted position error with a three-dimensional (3D) vector Er (Er = [Δx, Δy, Δz,]).” See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of Watanabe to further include the teachings of Chen with a reasonable expectation of success because “given the complexity of the sources of the robot’s residual errors (i.e., elastic deformation of links and joints caused by the link self-gravity of the robot and the change of external payload, gear backlash, hysteresis, and friction) after model-based calibration, these residual errors generally have strong nonlinearity” and a “DNN can achieve complex function approximation by learning a deep nonlinear network structure.” (See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
Chen also does not explicitly teach, but Pickett teaches
A surgical system comprising: … a surgical tool supported and moveable by the robotic arm and being configured to interact with an anatomy; (“a system is provided for enabling autonomous or semi-autonomous surgical operations. … and generate one or more control signals for controlling i) a movement of the tool based on the optimal path and ii) a tension force applied to the tissue by the tool during the surgical procedure.” See at least [0004]; “The tool 103 may optionally couple to a sensor for sensing stitch tension or tissue tension for the force control. In optional embodiments, a sensor may be operably coupled to the tool for measuring a force or tension applied to the tissue.” See at least [0049])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of Watanabe and Chen to further include the teachings of Picket with a reasonable expectation of success to “automate surgical procedures without or with little human intervention” and because “the provided autonomous robotic system may be utilized in a minimal access surgery (also known as minimally invasive surgery) which minimizes trauma to soft tissue, reduces post-operative pain, promotes earlier mobilization, shortens hospital stays, and speeds rehabilitation.” (See at least [0003])
Regarding Claim 3,
Watanabe further teaches
wherein: the sensing system is further configured to detect a disturbance force applied to one or more of the links of the robotic arm; and the one or more controllers are further configured to: obtain, from the sensing system, the disturbance force; (“The torques Tx, Ty, and Tz about the axes of the respective joints are determined using the Newton-Euler method based on the angles and actual rotational speeds of the respective joints which have been read (step S3).” See at least [0040]; “In the second embodiment, the following points are different from the processing shown by the processing flowchart shown in FIG. 5. 1. At step S4, the bending angle .gamma. of the entire link as shown in FIG. 3 is calculated, in addition to .DELTA..alpha., .DELTA..beta., and .DELTA..theta.. This bending angle .gamma. is determined by dividing the torque Tz about the Z-axis determined at step S3 by the preset coefficient kz' for determining the bending of the link. .gamma.=Tz/kz' … The second embodiment is the same as the case in which only the bendings of the joints are taken into consideration as described above except the above points.” See at least [0049-0053])
input the disturbance force to the (“a position/attitude x' is determined from the rotational angle .theta.' determined at step S6 by executing forward transformation using the corrected D-H parameters .alpha., .beta., and .theta.0 determined at step S5 (step S8). Then, the target position/attitude x read at step S1 is subtracted from the thus determined position/attitude x' to thereby calculate the difference .DELTA.x therebetween (step S9).” See at least [0042], wherein the rotational angle .theta.' is calculated based on the pose and torque (steps S4, S5, S6) and the difference .DELTA.x is the deflection.)
and control the robotic arm and/or the (“Next, it is determined whether or not the absolute value of the difference between the final rotational angle .theta. and the rotational angle .theta.st at the beginning of correction, determined at step S6, is equal to or larger than a set threshold value .theta.max (step S16). When the absolute value is equal to or larger than the threshold value .theta.max, alarm is issued and alarm processing is executed to abruptly stop the robot by applying brake thereto and outputting speed reduction torque commands to the servo motors of the respective axes (step S18). In contrast, when the absolute value is less than the set value .theta. max, a rotational angle j on the teaching pendant (rotational angle to be taught and commanded) is determined by subtracting the initial rotational angle .theta.0, determined at step S5, from the final rotational angle .theta. and stored (step S17).” See at least [0047]; “the offset of the distal end position of the robot hand made by bending due to external force is also compensated.” See at least [0062])
Watanabe does not explicitly teach, but Chen teaches
estimate the deflection of the robotic arm based on the output of the machine learning model; (“To realise the prediction and compensation of the robot positioning error, the output of the DNN is the predicted position error with a three-dimensional (3D) vector Er (Er = [Δx, Δy, Δz,]).” See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe to further include the teachings of Chen with a reasonable expectation of success because “given the complexity of the sources of the robot’s residual errors (i.e., elastic deformation of links and joints caused by the link self-gravity of the robot and the change of external payload, gear backlash, hysteresis, and friction) after model-based calibration, these residual errors generally have strong nonlinearity” and a “DNN can achieve complex function approximation by learning a deep nonlinear network structure.” (See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
Chen also does not explicitly teach, but Pickett teaches
the surgical tool; (“a surgical tool” See at least [0042])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe and Chen to further include the teachings of Picket with a reasonable expectation of success to “automate surgical procedures without or with little human intervention” and because “the provided autonomous robotic system may be utilized in a minimal access surgery (also known as minimally invasive surgery) which minimizes trauma to soft tissue, reduces post-operative pain, promotes earlier mobilization, shortens hospital stays, and speeds rehabilitation.” (See at least [0003])
Regarding Claim 5,
Watanabe does not explicitly teach, but Pickett teaches
wherein the one or more controllers are configured to: control the robotic arm to move the surgical tool along a tool path to interact with the anatomy; and control the robotic arm and/or the surgical tool based on the estimated deflection by being configured to modify the tool path. (“determine an optimal path for automatically moving the tool to perform a surgical procedure at the surgical site; and generate one or more control signals for controlling i) a movement of the tool based on the optimal path and ii) a tension force applied to the tissue by the tool during the surgical procedure.” See at least [0004]; “the provided control algorithm may be capable of accounting for changes in the dynamic environment such as correcting tool position errors caused by external forces.” See at least [0059]; Also see at least [0070])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe to further include the teachings of Picket with a reasonable expectation of success to “automate surgical procedures without or with little human intervention” and because “the provided autonomous robotic system may be utilized in a minimal access surgery (also known as minimally invasive surgery) which minimizes trauma to soft tissue, reduces post-operative pain, promotes earlier mobilization, shortens hospital stays, and speeds rehabilitation.” (See at least [0003])
Regarding Claim 7,
Watanabe further teaches
wherein the one or more controllers are configured to: control the robotic arm to move the surgical tool to a commanded pose; (“First, a target position/attitude x commanded at this time, that is, a teaching point or an interpolating point x is read (step S1) by executing a teaching program.” See at least [0039])
and control the robotic arm and/or the (“it is intended to take the bendings in the joints and links of the robot into consideration and to drive the servo motors M1 to M6 acting as the actuators of the respective axes by correcting the bending components so that the robot hand distal end (tool center point) is accurately driven and controlled according to a target position/attitude.” See at least [0034]; Also see at least [0047])
Regarding Claim 14,
Watanabe does not explicitly teach, but Chen teaches
wherein the machine learning model is a deep learning model including at least two hidden layers. (“DNN can achieve complex function approximation by learning a deep nonlinear network structure (5, 6, or even 10 hidden layers are usually present) and how a strong capacity to learn the effective features of data from a small number of samples.” See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe to further include the teachings of Chen with a reasonable expectation of success because “given the complexity of the sources of the robot’s residual errors (i.e., elastic deformation of links and joints caused by the link self-gravity of the robot and the change of external payload, gear backlash, hysteresis, and friction) after model-based calibration, these residual errors generally have strong nonlinearity” and a “DNN can achieve complex function approximation by learning a deep nonlinear network structure.” (See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
Regarding Claim 15,
Watanabe does not explicitly teach, but Chen teaches
wherein the estimated deflection is a non-geometric deflection. (“the nonlinear errors including the intrinsic errors (i.e., elastic deformation of the link and joint caused by the link self-gravity of the robot, gear backlash, hysteresis, and friction) … a GA-DNN prediction model is proposed to achieve the optimal nonlinear error prediction compensation and to further enhance the positioning accuracy of the robot.” See at least pg. 2, Section 1. Introduction)
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe to further include the teachings of Chen with a reasonable expectation of success because “given the complexity of the sources of the robot’s residual errors (i.e., elastic deformation of links and joints caused by the link self-gravity of the robot and the change of external payload, gear backlash, hysteresis, and friction) after model-based calibration, these residual errors generally have strong nonlinearity” and a “DNN can achieve complex function approximation by learning a deep nonlinear network structure.” (See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
Regarding Claim 17,
Watanabe teaches
A method of operating a (“the principle and method of an embodiment of the present invention will be described with reference to a robot controller for controlling a six-axis vertical articulated robot as an example. This principle and method are also similarly applicable to other types of robots.” See at least [0017])
that includes a robotic arm comprising a plurality of links and joints, a (“a robot controller for a robot whose respective links are driven by respective actuators and which executes a job by pressing a robot hand distal end against an object includes a sensor for detecting a reaction force produced when the robot hand distal end is pressed against the object; bending calculation means for determining the bendings produced in the respective joints at the target position/attitude of the robot hand distal end based on a teaching program;” See at least [0009])
the method comprising the one or more controllers performing the following: obtaining, based on kinematic data from the robotic arm, a pose of the (“a rotational angle .theta.' is determined from the target position/attitude x read at step S1 by executing inverse transformation using the corrected D-H parameters .alpha.s, .beta.s, and .theta.0s where no twist is produced, and the rotational angle .theta.' is stored as a rotational angle .theta.st at the beginning of correction (step S6).” See at least [0041])
obtaining, from the sensing system, the interaction force; (“The torques Tx, Ty, and Tz about the axes of the respective joints are determined using the Newton-Euler method based on the angles and actual rotational speeds of the respective joints which have been read (step S3).” See at least [0040]; “However, the torques Tx, Ty, and Tz about the respective axes may be detected using a force sensor or a torque sensor. In this case, the processings at steps S2 and S3 shown in FIG. 5 are executed using the forces or torques obtained by these sensors.” See at least [0061])
inputting the pose of the (“a position/attitude x' is determined from the rotational angle .theta.' determined at step S6 by executing forward transformation using the corrected D-H parameters .alpha., .beta., and .theta.0 determined at step S5 (step S8). Then, the target position/attitude x read at step S1 is subtracted from the thus determined position/attitude x' to thereby calculate the difference .DELTA.x therebetween (step S9).” See at least [0042], wherein the rotational angle .theta.' is calculated based on the pose and torque (steps S4, S5, S6) and the difference .DELTA.x is the deflection.)
and controlling the robotic arm based on the estimated deflection. (“Next, it is determined whether or not the absolute value of the difference between the final rotational angle .theta. and the rotational angle .theta.st at the beginning of correction, determined at step S6, is equal to or larger than a set threshold value .theta.max (step S16). When the absolute value is equal to or larger than the threshold value .theta.max, alarm is issued and alarm processing is executed to abruptly stop the robot by applying brake thereto and outputting speed reduction torque commands to the servo motors of the respective axes (step S18). In contrast, when the absolute value is less than the set value .theta. max, a rotational angle j on the teaching pendant (rotational angle to be taught and commanded) is determined by subtracting the initial rotational angle .theta.0, determined at step S5, from the final rotational angle .theta. and stored (step S17).” See at least [0047]; “the offset of the distal end position of the robot hand made by bending due to external force is also compensated.” See at least [0062])
Watanabe does not explicitly teach, but Chen teaches
estimating a deflection of the robotic arm based on an output of the machine learning model; (“To realise the prediction and compensation of the robot positioning error, the output of the DNN is the predicted position error with a three-dimensional (3D) vector Er (Er = [Δx, Δy, Δz,]).” See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of Watanabe to further include the teachings of Chen with a reasonable expectation of success because “given the complexity of the sources of the robot’s residual errors (i.e., elastic deformation of links and joints caused by the link self-gravity of the robot and the change of external payload, gear backlash, hysteresis, and friction) after model-based calibration, these residual errors generally have strong nonlinearity” and a “DNN can achieve complex function approximation by learning a deep nonlinear network structure.” (See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
Chen also does not explicitly teach, but Pickett teaches
a surgical system … a surgical tool supported and moveable by the robotic arm and being configured to interact with an anatomy; (“a system is provided for enabling autonomous or semi-autonomous surgical operations. … and generate one or more control signals for controlling i) a movement of the tool based on the optimal path and ii) a tension force applied to the tissue by the tool during the surgical procedure.” See at least [0004]; “The tool 103 may optionally couple to a sensor for sensing stitch tension or tissue tension for the force control. In optional embodiments, a sensor may be operably coupled to the tool for measuring a force or tension applied to the tissue.” See at least [0049])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of Watanabe and Chen to further include the teachings of Picket with a reasonable expectation of success to “automate surgical procedures without or with little human intervention” and because “the provided autonomous robotic system may be utilized in a minimal access surgery (also known as minimally invasive surgery) which minimizes trauma to soft tissue, reduces post-operative pain, promotes earlier mobilization, shortens hospital stays, and speeds rehabilitation.” (See at least [0003])
Regarding Claim 19,
Watanabe teaches
A non-transitory computer-readable medium for use with a (“A system program for supporting the basic function of the robot and the robot controller is stored in the ROM of the memory 2. Further, the operation programs and relating set data of the robot that is taught according to an application are stored in the non-volatile memory of the memory 2. The RAM of the memory 2 is used as a memory region in which data used in various types of calculation processing executed by the main processor 1 is temporarily stored.” See at least [0032])
that includes a robotic arm with a plurality of links and joints, a (“a robot controller for a robot whose respective links are driven by respective actuators and which executes a job by pressing a robot hand distal end against an object includes a sensor for detecting a reaction force produced when the robot hand distal end is pressed against the object; bending calculation means for determining the bendings produced in the respective joints at the target position/attitude of the robot hand distal end based on a teaching program;” See at least [0009])
obtain, based on kinematic data from the robotic arm, a pose of the (“a rotational angle .theta.' is determined from the target position/attitude x read at step S1 by executing inverse transformation using the corrected D-H parameters .alpha.s, .beta.s, and .theta.0s where no twist is produced, and the rotational angle .theta.' is stored as a rotational angle .theta.st at the beginning of correction (step S6).” See at least [0041])
obtain, from the sensing system, the interaction force; (“The torques Tx, Ty, and Tz about the axes of the respective joints are determined using the Newton-Euler method based on the angles and actual rotational speeds of the respective joints which have been read (step S3).” See at least [0040]; “However, the torques Tx, Ty, and Tz about the respective axes may be detected using a force sensor or a torque sensor. In this case, the processings at steps S2 and S3 shown in FIG. 5 are executed using the forces or torques obtained by these sensors.” See at least [0061])
input the pose of the (“a position/attitude x' is determined from the rotational angle .theta.' determined at step S6 by executing forward transformation using the corrected D-H parameters .alpha., .beta., and .theta.0 determined at step S5 (step S8). Then, the target position/attitude x read at step S1 is subtracted from the thus determined position/attitude x' to thereby calculate the difference .DELTA.x therebetween (step S9).” See at least [0042], wherein the rotational angle .theta.' is calculated based on the pose and torque (steps S4, S5, S6) and the difference .DELTA.x is the deflection.)
and control the robotic arm and/or the (“Next, it is determined whether or not the absolute value of the difference between the final rotational angle .theta. and the rotational angle .theta.st at the beginning of correction, determined at step S6, is equal to or larger than a set threshold value .theta.max (step S16). When the absolute value is equal to or larger than the threshold value .theta.max, alarm is issued and alarm processing is executed to abruptly stop the robot by applying brake thereto and outputting speed reduction torque commands to the servo motors of the respective axes (step S18). In contrast, when the absolute value is less than the set value .theta. max, a rotational angle j on the teaching pendant (rotational angle to be taught and commanded) is determined by subtracting the initial rotational angle .theta.0, determined at step S5, from the final rotational angle .theta. and stored (step S17).” See at least [0047]; “the offset of the distal end position of the robot hand made by bending due to external force is also compensated.” See at least [0062])
Watanabe does not explicitly teach, but Chen teaches
estimate a deflection of the robotic arm based on an output of the machine learning model; (“To realise the prediction and compensation of the robot positioning error, the output of the DNN is the predicted position error with a three-dimensional (3D) vector Er (Er = [Δx, Δy, Δz,]).” See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of Watanabe to further include the teachings of Chen with a reasonable expectation of success because “given the complexity of the sources of the robot’s residual errors (i.e., elastic deformation of links and joints caused by the link self-gravity of the robot and the change of external payload, gear backlash, hysteresis, and friction) after model-based calibration, these residual errors generally have strong nonlinearity” and a “DNN can achieve complex function approximation by learning a deep nonlinear network structure.” (See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
Chen also does not explicitly teach, but Pickett teaches
a surgical system that includes a surgical tool supported and moveable by the robotic arm and being configured to interact with an anatomy, (“a system is provided for enabling autonomous or semi-autonomous surgical operations. … and generate one or more control signals for controlling i) a movement of the tool based on the optimal path and ii) a tension force applied to the tissue by the tool during the surgical procedure.” See at least [0004]; “The tool 103 may optionally couple to a sensor for sensing stitch tension or tissue tension for the force control. In optional embodiments, a sensor may be operably coupled to the tool for measuring a force or tension applied to the tissue.” See at least [0049])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of Watanabe and Chen to further include the teachings of Picket with a reasonable expectation of success to “automate surgical procedures without or with little human intervention” and because “the provided autonomous robotic system may be utilized in a minimal access surgery (also known as minimally invasive surgery) which minimizes trauma to soft tissue, reduces post-operative pain, promotes earlier mobilization, shortens hospital stays, and speeds rehabilitation.” (See at least [0003])
Claim(s) 2 and 4 is/are rejected under 35 U.S.C. 103 as being unpatentable over Watanabe (US 20020173878 A1) in view of Chen (IDS: “Evolutionary Robot Calibration and Nonlinear Compensation Methodology Based on GA-DNN and an Extra Compliance Error Model”), Pickett (US 20230000565 A1), and Gayet (US 20250186142 A1).
Regarding Claim 2,
Watanabe further teaches
wherein: the sensing system is further configured to detect a (“detecting a reaction force produced when the robot hand distal end is pressed against the object;” See at least [0009]; “The torques Tx, Ty, and Tz about the axes of the respective joints are determined using the Newton-Euler method based on the angles and actual rotational speeds of the respective joints which have been read (step S3).” See at least [0040]; “However, the torques Tx, Ty, and Tz about the respective axes may be detected using a force sensor or a torque sensor. In this case, the processings at steps S2 and S3 shown in FIG. 5 are executed using the forces or torques obtained by these sensors.” See at least [0061])
input the (“a position/attitude x' is determined from the rotational angle .theta.' determined at step S6 by executing forward transformation using the corrected D-H parameters .alpha., .beta., and .theta.0 determined at step S5 (step S8). Then, the target position/attitude x read at step S1 is subtracted from the thus determined position/attitude x' to thereby calculate the difference .DELTA.x therebetween (step S9).” See at least [0042], wherein the rotational angle .theta.' is calculated based on the pose and torque (steps S4, S5, S6) and the difference .DELTA.x is the deflection.)
and control the robotic arm and/or the (“Next, it is determined whether or not the absolute value of the difference between the final rotational angle .theta. and the rotational angle .theta.st at the beginning of correction, determined at step S6, is equal to or larger than a set threshold value .theta.max (step S16). When the absolute value is equal to or larger than the threshold value .theta.max, alarm is issued and alarm processing is executed to abruptly stop the robot by applying brake thereto and outputting speed reduction torque commands to the servo motors of the respective axes (step S18). In contrast, when the absolute value is less than the set value .theta. max, a rotational angle j on the teaching pendant (rotational angle to be taught and commanded) is determined by subtracting the initial rotational angle .theta.0, determined at step S5, from the final rotational angle .theta. and stored (step S17).” See at least [0047]; “the offset of the distal end position of the robot hand made by bending due to external force is also compensated.” See at least [0062])
Watanabe does not explicitly teach, but Chen teaches
estimate the deflection of the robotic arm based on the output of the machine learning model; (“To realise the prediction and compensation of the robot positioning error, the output of the DNN is the predicted position error with a three-dimensional (3D) vector Er (Er = [Δx, Δy, Δz,]).” See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe to further include the teachings of Chen with a reasonable expectation of success because “given the complexity of the sources of the robot’s residual errors (i.e., elastic deformation of links and joints caused by the link self-gravity of the robot and the change of external payload, gear backlash, hysteresis, and friction) after model-based calibration, these residual errors generally have strong nonlinearity” and a “DNN can achieve complex function approximation by learning a deep nonlinear network structure.” (See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
Chen also does not explicitly teach, but Gayet teaches
a user-applied force on the surgical tool; (“At step 1605, the system compensates for the gravity of the surgical instrument and the force applied by the hand of the operator, e.g., by measuring the force applied to the distal end of robot arm 300 due to the mass of the surgical instrument.” See at least [0161])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe and Chen to further include the teachings of Gayet with a reasonable expectation of success to provide “the surgeon with the ability to seamlessly position and manipulate various surgical instruments as needed, thus avoiding the workflow limitations inherent to both human and mechanical solutions.” (See at least [0008]; Also see at least [0060])
Regarding Claim 4,
Watanabe further teaches
wherein: the sensing system is further configured to detect: a disturbance force applied to one or more of the links of the robotic arm; and a (“detecting a reaction force produced when the robot hand distal end is pressed against the object;” See at least [0009]; “The torques Tx, Ty, and Tz about the axes of the respective joints are determined using the Newton-Euler method based on the angles and actual rotational speeds of the respective joints which have been read (step S3).” See at least [0040]; “In the second embodiment, the following points are different from the processing shown by the processing flowchart shown in FIG. 5. 1. At step S4, the bending angle .gamma. of the entire link as shown in FIG. 3 is calculated, in addition to .DELTA..alpha., .DELTA..beta., and .DELTA..theta.. This bending angle .gamma. is determined by dividing the torque Tz about the Z-axis determined at step S3 by the preset coefficient kz' for determining the bending of the link. .gamma.=Tz/kz' … The second embodiment is the same as the case in which only the bendings of the joints are taken into consideration as described above except the above points.” See at least [0049-0053])
input the disturbance force and the user-applied force to the (“a position/attitude x' is determined from the rotational angle .theta.' determined at step S6 by executing forward transformation using the corrected D-H parameters .alpha., .beta., and .theta.0 determined at step S5 (step S8). Then, the target position/attitude x read at step S1 is subtracted from the thus determined position/attitude x' to thereby calculate the difference .DELTA.x therebetween (step S9).” See at least [0042], wherein the rotational angle .theta.' is calculated based on the pose and torque (steps S4, S5, S6) and the difference .DELTA.x is the deflection.)
and control the robotic arm and/or the (“Next, it is determined whether or not the absolute value of the difference between the final rotational angle .theta. and the rotational angle .theta.st at the beginning of correction, determined at step S6, is equal to or larger than a set threshold value .theta.max (step S16). When the absolute value is equal to or larger than the threshold value .theta.max, alarm is issued and alarm processing is executed to abruptly stop the robot by applying brake thereto and outputting speed reduction torque commands to the servo motors of the respective axes (step S18). In contrast, when the absolute value is less than the set value .theta. max, a rotational angle j on the teaching pendant (rotational angle to be taught and commanded) is determined by subtracting the initial rotational angle .theta.0, determined at step S5, from the final rotational angle .theta. and stored (step S17).” See at least [0047]; “the offset of the distal end position of the robot hand made by bending due to external force is also compensated.” See at least [0062])
Watanabe does not explicitly teach, but Chen teaches
estimate the deflection of the robotic arm based on the output of the machine learning model; (“To realise the prediction and compensation of the robot positioning error, the output of the DNN is the predicted position error with a three-dimensional (3D) vector Er (Er = [Δx, Δy, Δz,]).” See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe to further include the teachings of Chen with a reasonable expectation of success because “given the complexity of the sources of the robot’s residual errors (i.e., elastic deformation of links and joints caused by the link self-gravity of the robot and the change of external payload, gear backlash, hysteresis, and friction) after model-based calibration, these residual errors generally have strong nonlinearity” and a “DNN can achieve complex function approximation by learning a deep nonlinear network structure.” (See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
Chen also does not explicitly teach, but Gayet teaches
wherein: the sensing system is further configured to detect: a disturbance force applied to one or more of the links of the robotic arm; and a user-applied force on the surgical tool; (“Force detection module 1422 may be executed by processor 1402 for detecting forces applied on robot arm 300, e.g., at the joints or links of robot arm 300 or along the surgical instrument, as well as applied on the trocar, e.g., body wall forces.” See at least [0118]; “At step 1605, the system compensates for the gravity of the surgical instrument and the force applied by the hand of the operator, e.g., by measuring the force applied to the distal end of robot arm 300 due to the mass of the surgical instrument.” See at least [0161])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe and Chen to further include the teachings of Gayet with a reasonable expectation of success to provide “the surgeon with the ability to seamlessly position and manipulate various surgical instruments as needed, thus avoiding the workflow limitations inherent to both human and mechanical solutions.” (See at least [0008]; Also see at least [0060])
Claim(s) 6 and 12 is/are rejected under 35 U.S.C. 103 as being unpatentable over Watanabe (US 20020173878 A1) in view of Chen (IDS: “Evolutionary Robot Calibration and Nonlinear Compensation Methodology Based on GA-DNN and an Extra Compliance Error Model”), Pickett (US 20230000565 A1), and Henderson (US 20230157773 A1).
Regarding Claim 6,
Modified Watanabe does not explicitly teach, but Henderson teaches
wherein the one or more controllers are configured to: control the robotic arm to move the surgical tool at a feed rate to interact with the anatomy; (“The surgical plan may have instructions for the surgical robot 100 in the form of a cut-file. The cut-file contains instructions (e.g., cut paths, orientations, velocities, feed rates, spindle speeds, etc.) that direct the end-effector 106 to cut the bone according the plan.” See at least [0007])
and control the robotic arm and/or the surgical tool based on the estimated deflection by being configured to modify the feed rate. (“For a robotic surgical system, trajectory deviations are introduced by factors that might include robot actuation biases and vibration, movement of the target anatomy, or unexpected forces imparted by the anatomy. … In general, the larger the cutting forces being imparted to the tissue, the more the anatomy moves thereby enhancing deviations. The deviations are also expected to increase with the translational speed of the end-effector. Higher translational speed gives the system less-time to correct actuation biases and also increase cutting forces that may tend to move anatomy.” See at least [0010]; “If after cutting one or more regions, the robotic system determines that the deviation is higher than desired for a given tissue region being cut, the surgical robot revisits that region in a second (or subsequent) pass, cutting the surface again. The second pass may be performed in a manner such that the deviations are statistically uncorrelated with the deviations of the first pass, and possibly with different operating conditions, such as a lower speed, to reduce the deviations.” See at least [0038])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe to further include the teachings of Henderson with a reasonable expectation of success to improve accuracy of the robotic surgery. (See at least [0010] and [0037-0038])
Regarding Claim 12,
Modified Watanabe does not explicitly teach, but Henderson teaches
wherein: the surgical tool is configured to cut tissue of the anatomy; (“The surgical plan may have instructions for the surgical robot 100 in the form of a cut-file. The cut-file contains instructions (e.g., cut paths, orientations, velocities, feed rates, spindle speeds, etc.) that direct the end-effector 106 to cut the bone according the plan.” See at least [0007])
and the interaction force is based on the interaction of the surgical tool with the tissue during cutting of the tissue. (“For a robotic surgical system, trajectory deviations are introduced by factors that might include robot actuation biases and vibration, movement of the target anatomy, or unexpected forces imparted by the anatomy. … In general, the larger the cutting forces being imparted to the tissue, the more the anatomy moves thereby enhancing deviations.” See at least [0010])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe to further include the teachings of Henderson with a reasonable expectation of success to improve accuracy of the robotic surgery. (See at least [0010] and [0037-0038])
Claim(s) 8-11, 18, and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Watanabe (US 20020173878 A1) in view of Chen (IDS: “Evolutionary Robot Calibration and Nonlinear Compensation Methodology Based on GA-DNN and an Extra Compliance Error Model”), Pickett (US 20230000565 A1), and Barral (US 20180055577 A1).
Regarding Claim 8,
Modified Watanabe does not explicitly teach, but Barral teaches
wherein the one or more controllers are configured to: obtain an operating parameter of the surgical tool; input the operating parameter of the surgical tool and the estimated deflection into a second machine learning model; and characterize the interaction of the surgical tool with the anatomy based on an output of the second machine learning model. (“adjusting the actual position of the surgical instrument includes using a machine learning algorithm to control surgical robot 103 to compensate for discrepancies in the force needed to cut through specific biological media. In this and other embodiments, the machine learning algorithm is trained to recognize a composition of a medium (e.g., skin, muscle, tendon, etc.) and output a force sufficient to move the surgical instrument from the actual position to the desired position through the medium. For example, the machine learning algorithm may recognize that a scalpel held by surgical robot 103 is trying to cut through tendon. Accordingly, the machine learning algorithm may apply additional force to the scalpel in order to make a smooth direct cut. In one embodiment, the medium (e.g., biological tissue) is recognized by the machine learning algorithm via at least one of optical appearance, the force sufficient to move the surgical instrument through the medium, input from a user of the surgical instrument, and/or the difference between the actual and desired positions of the surgical instrument.” See at least [0020])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe to further include the teachings of Barral with a reasonable expectation of success to improve surgical outcomes by facilitating accurate positioning and compliance compensating for mechanical imperfections of the robotic system, variable payloads, and contact forces. (See at least [0003])
Regarding Claim 9,
Modified Watanabe does not explicitly teach, but Barral teaches
wherein the one or more controllers characterize the interaction by characterizing a type of the anatomy. (“the machine learning algorithm is trained to recognize a composition of a medium (e.g., skin, muscle, tendon, etc.) and output a force sufficient to move the surgical instrument from the actual position to the desired position through the medium. For example, the machine learning algorithm may recognize that a scalpel held by surgical robot 103 is trying to cut through tendon.” See at least [0020])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe to further include the teachings of Barral with a reasonable expectation of success to improve surgical outcomes by facilitating accurate positioning and compliance compensating for mechanical imperfections of the robotic system, variable payloads, and contact forces. (See at least [0003])
Regarding Claim 10,
Modified Watanabe does not explicitly teach, but Barral teaches
wherein the one or more controllers characterize the interaction by characterizing a workload of the surgical tool. (“the machine learning algorithm is trained to recognize a composition of a medium (e.g., skin, muscle, tendon, etc.) and output a force sufficient to move the surgical instrument from the actual position to the desired position through the medium. For example, the machine learning algorithm may recognize that a scalpel held by surgical robot 103 is trying to cut through tendon. Accordingly, the machine learning algorithm may apply additional force to the scalpel in order to make a smooth direct cut.” See at least [0020])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe to further include the teachings of Barral with a reasonable expectation of success to improve surgical outcomes by facilitating accurate positioning and compliance compensating for mechanical imperfections of the robotic system, variable payloads, and contact forces. (See at least [0003])
Regarding Claim 11,
Modified Watanabe does not explicitly teach, but Barral teaches
wherein the one or more controllers control the robotic arm and/or the surgical tool based on the characterized interaction by being configured to perform one or more of: modify a tool path of the surgical tool; modify a feed rate of the surgical tool; modify a cutting speed of the surgical tool; adjust a commanded pose of the surgical tool; halt movement of the surgical tool; move the surgical tool away from the anatomy; and/or generate or modify a virtual haptic setting for the surgical tool. (“adjusting the actual position of the surgical instrument includes using a machine learning algorithm to control surgical robot 103 to compensate for discrepancies in the force needed to cut through specific biological media. … In another or the same embodiment, when moving the surgical instrument to the desired position, the difference between actual and desired position could be converted into haptic feedback which is provided to the user (e.g., desired position wasn't achieved, but the user now knows it because of a controller vibration).” See at least [0020])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe to further include the teachings of Barral with a reasonable expectation of success to improve surgical outcomes by facilitating accurate positioning and compliance compensating for mechanical imperfections of the robotic system, variable payloads, and contact forces. (See at least [0003])
Regarding Claim 18,
Modified Watanabe does not explicitly teach, but Barral teaches
comprising the one or more controllers: obtaining an operating parameter of the surgical tool; inputting the operating parameter of the surgical tool and the estimated deflection into a second machine learning model; characterizing the interaction of the surgical tool with the anatomy based on an output of the second machine learning model; and controlling the robotic arm and/or the surgical tool based on the characterized interaction. (“adjusting the actual position of the surgical instrument includes using a machine learning algorithm to control surgical robot 103 to compensate for discrepancies in the force needed to cut through specific biological media. In this and other embodiments, the machine learning algorithm is trained to recognize a composition of a medium (e.g., skin, muscle, tendon, etc.) and output a force sufficient to move the surgical instrument from the actual position to the desired position through the medium. For example, the machine learning algorithm may recognize that a scalpel held by surgical robot 103 is trying to cut through tendon. Accordingly, the machine learning algorithm may apply additional force to the scalpel in order to make a smooth direct cut. In one embodiment, the medium (e.g., biological tissue) is recognized by the machine learning algorithm via at least one of optical appearance, the force sufficient to move the surgical instrument through the medium, input from a user of the surgical instrument, and/or the difference between the actual and desired positions of the surgical instrument.” See at least [0020])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe to further include the teachings of Barral with a reasonable expectation of success to improve surgical outcomes by facilitating accurate positioning and compliance compensating for mechanical imperfections of the robotic system, variable payloads, and contact forces. (See at least [0003])
Regarding Claim 20,
Modified Watanabe does not explicitly teach, but Barral teaches
wherein the instructions, when executed by the one or more processors, are configured to: obtain an operating parameter of the surgical tool; input the operating parameter of the surgical tool and the estimated deflection into a second machine learning model; characterize the interaction of the surgical tool with the anatomy based on an output of the second machine learning model; and control the robotic arm and/or the surgical tool based on the characterized interaction. (“adjusting the actual position of the surgical instrument includes using a machine learning algorithm to control surgical robot 103 to compensate for discrepancies in the force needed to cut through specific biological media. In this and other embodiments, the machine learning algorithm is trained to recognize a composition of a medium (e.g., skin, muscle, tendon, etc.) and output a force sufficient to move the surgical instrument from the actual position to the desired position through the medium. For example, the machine learning algorithm may recognize that a scalpel held by surgical robot 103 is trying to cut through tendon. Accordingly, the machine learning algorithm may apply additional force to the scalpel in order to make a smooth direct cut. In one embodiment, the medium (e.g., biological tissue) is recognized by the machine learning algorithm via at least one of optical appearance, the force sufficient to move the surgical instrument through the medium, input from a user of the surgical instrument, and/or the difference between the actual and desired positions of the surgical instrument.” See at least [0020])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe to further include the teachings of Barral with a reasonable expectation of success to improve surgical outcomes by facilitating accurate positioning and compliance compensating for mechanical imperfections of the robotic system, variable payloads, and contact forces. (See at least [0003])
Claim(s) 13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Watanabe (US 20020173878 A1) in view of Chen (IDS: “Evolutionary Robot Calibration and Nonlinear Compensation Methodology Based on GA-DNN and an Extra Compliance Error Model”), Pickett (US 20230000565 A1), and Post (US 20170128136 A1).
Regarding Claim 13,
Modified Watanabe does not explicitly teach, but Post teaches
wherein the sensing system comprises at least a force/torque sensor coupled to a distal link of the robotic arm. (“A sensor 34, such as a force-torque sensor, may be mounted between the distal link 20e and the tool 30.” See at least [0039])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe to further include the teachings of Post with a reasonable expectation of success to “allow the user to grossly position the manipulator with ease.” (See at least [0010]; Also see at least [0039])
Claim(s) 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Watanabe (US 20020173878 A1) in view of Chen (IDS: “Evolutionary Robot Calibration and Nonlinear Compensation Methodology Based on GA-DNN and an Extra Compliance Error Model”), Pickett (US 20230000565 A1), and Danilchenko (US 20220134558 A1).
Regarding Claim 16,
Watanabe further teaches
and wherein in the deflection monitoring mode the one or more controllers are configured to: input the pose of the (“a position/attitude x' is determined from the rotational angle .theta.' determined at step S6 by executing forward transformation using the corrected D-H parameters .alpha., .beta., and .theta.0 determined at step S5 (step S8). Then, the target position/attitude x read at step S1 is subtracted from the thus determined position/attitude x' to thereby calculate the difference .DELTA.x therebetween (step S9).” See at least [0042], wherein the rotational angle .theta.' is calculated based on the pose and torque (steps S4, S5, S6) and the difference .DELTA.x is the deflection.)
Watanabe does not explicitly teach, but Chen teaches
estimate the deflection of the robotic arm based on the output of the machine learning model; (“To realise the prediction and compensation of the robot positioning error, the output of the DNN is the predicted position error with a three-dimensional (3D) vector Er (Er = [Δx, Δy, Δz,]).” See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe to further include the teachings of Chen with a reasonable expectation of success because “given the complexity of the sources of the robot’s residual errors (i.e., elastic deformation of links and joints caused by the link self-gravity of the robot and the change of external payload, gear backlash, hysteresis, and friction) after model-based calibration, these residual errors generally have strong nonlinearity” and a “DNN can achieve complex function approximation by learning a deep nonlinear network structure.” (See at least pg. 4, Section 3. DNN-Based Regression Prediction Method for the Nonlinear Errors)
Chen also does not explicitly teach, but Pickett teaches
the surgical tool; (“a surgical tool” See at least [0042])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe and Chen to further include the teachings of Picket with a reasonable expectation of success to “automate surgical procedures without or with little human intervention” and because “the provided autonomous robotic system may be utilized in a minimal access surgery (also known as minimally invasive surgery) which minimizes trauma to soft tissue, reduces post-operative pain, promotes earlier mobilization, shortens hospital stays, and speeds rehabilitation.” (See at least [0003])
Pickett also does not explicitly teach, but Danilchenko teaches
wherein the one or more controllers are configured to trigger a deflection monitoring mode in response to detection of a certain condition, (“At step 510, by monitoring a tracked position of the surgical tool relative to the planned path, the processing circuit 260 can calculate an error between the actual position of the surgical tool and the planned path. In some cases, the error is a distance between the actual position of the surgical tool and closest point in space on the planned path. In other cases, the error is a distance between the actual position of the surgical tool and a point in space where the surgical tool should have been if it had followed the planned path according to a planned rate of motion. The error can be compared to a threshold value over which a deviation from the planned path is considered to have occurred. That is, if the error exceeds the threshold value, a deviation is detected under step 510. … At step 512, in response to detection of a deviation of the surgical tool from the planned path in step 510, the automated motion of the robotic arm is stopped. … Step 512 may include switching control of the robotic device to a manual mode in which the user is allowed to manually reposition the surgical tool by directly manipulating the robotic device.” See at least [0089-0092])
It would have been obvious to a person having ordinary skill in the art before the effective filing date of the invention to modify the teachings of modified Watanabe, Chen, and Pickett to further include the teachings of Danilchenko with a reasonable expectation of success to facilitate safer control of the robot by preventing an undesirable slingshot or springing behavior. (See at least [0077])
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Oga (US 20110060460 A1) is pertinent because it discusses calculating error of a robot based on a tip end position and external force calculation.
Siravuru (US 20260263180 A1) is pertinent because it discusses enabling a compliance mode in response to an external force.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Karston G Evans whose telephone number is (571)272-8480. The examiner can normally be reached Mon-Fri 9:00-5:00.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Abby Lin can be reached at (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 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.
/KARSTON G. EVANS/Examiner, Art Unit 3657
