Detailed Action
This Office Action is taken in response to Applicant’s Amendment and Remarks filed on 03/02/2026 regarding Application No. 18/093,761 originally filed on 01/05/2023. Claims 1-8, 10-17, 19-22, 24-26 as filed are currently pending and have been considered as follows:
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments
Applicant’s arguments with respect to the claim(s) have been considered but are moot because the new ground of rejection does not rely on any reference applied in the prior rejection of record for any teaching or matter specifically challenged in the argument.
Claim Objections
Claim 26 is objected to because of the following informalities:
“load-displacement samples” should read “load-displacement samples.”
Appropriate correction is required.
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) 1-8, 10-17, 19-22, 24-26 are rejected under 35 U.S.C. 103 as being unpatentable over Bacher (NPL Title: Design and Control of Soft Robots Using Differentiable Simulation, Year 2021) in view of DisneyResearchHub (Youtube Video: “A Computational Design Tool for Compliant Mechanisms”, Year 2017) in further view of Hafner (NPL Title: X-CAD: optimizing CAD models with extended finite elements, Year 2019).
As per Claim 1, Bacher discloses of design and control of soft robots using differentiable simulation, comprising:
with a computing device, receiving user input including a definition of a robot mechanism with a plurality of rigid links, (as per “A soft robot made of passive materials, and actuated through external interactions, has a rest shape with an assigned material at every point, displaying a locally or globally homogenous or inhomogenous; isotropic or anisotropic behavior. Both the rest shape and the material assignment influence the deformation behavior of the robot. Representing these shape and material parameters with a parameter vector pdes, we seek to solve for their optimal values to, for example, maximize the contact surface or forces between the soft robot and objects of varying shape” in P3, Mathematical Tools for Optimal Design and Control, as per “important to be able to formulate design and control optimizations for robots consisting of rigid and flexible components,” in 2.4 Rigid-Flexible Systems)
with a simulator running on the computing device, simulating operation of the robot mechanism with the one or more flexible links to generate a simulated operation of the robot mechanism; (as per “To simulate a soft robot, we first discretize these entities using, for example, the finite element method (FEM).” in P3, 2.1 Differentiable Simulation, as per “For characterization, sensing, actuation, or rest shape optimzation, one typically has a target state
x
-
given, and seeks to find the optimal set of parameters…” in P4, 2.1 Differentiable Simulation)
with an optimizer running on the computing device, comparing the simulated operation of the robot mechanism to user-defined operations of the robot mechanism; (as per “For characterization, sensing, actuation, or rest shape optimzation, one typically has a target state
x
-
given, and seeks to find the optimal set of parameters… , and seeks to find the optimal set of parameters such that the difference between the simulated and target state… is minimal” in P4, 2.1 Differentiable Simulation, as per “a natural approach is to compare the simulated behavior to captured data, under matching actuation and external forces (see Fig. 2 top). This characterization can either be performed on material specimens [10] and individual sensors and actuators [11], or on the full robot [12].” in P2, Mathematical Tools for Optimal Design and Control)
based on the comparing, modifying a shape of at least one of the one or more flexible links; (as per “For rest shape or material distribution optimization, topology optimization has seen widespread use [31–36]. We can interpret traditional topology optimization as an application of differentiable simulation. However, because a linear elastic behavior is often assumed, it is, in its simplest form, not an ideal tool for soft robotics applications. To formulate shape optimizations, combinations of eXtended Finite Elements (XFE), level-set approaches that enable the continuous movement of material-material or material-void boundaries through elements, and analytical sensitivity analysis has seen use [15, 20, 37, 38]. This is again a variant of differentiable simulation that enables the optimization of soft robots consisting of composite materials [39]. Shape derivative techniques are also common [40].” in P6, 2.6 Alternative Optimization Techniques)
optimizing the shape of the at least one of the one or more flexible links by repeating the simulating and the comparing steps. (as per “Whenever we evaluate the objective g or its gradient after updating the set of parameters, we first solve for the state of the robot, represented by x and its time derivatives
x
˙
and
x
¨
, by integrating the dynamics system (Eq. 4) forward in time.” in P4, 2.1 Differentiable Simulation, as per Fig. 3)
PNG
media_image1.png
715
529
media_image1.png
Greyscale
Bacher fails to expressly disclose:
wherein the user input further includes a selection of one or more of the rigid links for replacement with one or more flexible links;
performing a remeshing-free parameterization of a volumetric rest shape of the one or more flexible links using splines;
DisneyResearchHub discloses of a computational design tool for compliant mechanisms, wherein the user input further includes a selection of one or more of the rigid links for replacement with one or more flexible links. (as per 0:05-0:20 of video, as per 1:55-2:20 of video)
PNG
media_image2.png
1092
1458
media_image2.png
Greyscale
PNG
media_image3.png
968
1443
media_image3.png
Greyscale
PNG
media_image4.png
1044
1427
media_image4.png
Greyscale
PNG
media_image5.png
1002
1442
media_image5.png
Greyscale
In this way, DisneyResearchHub operates to take a rigidly-articulated mechanism and are automatically replaces components with parameterized flexures (Video Description). Like Bacher, DisneyResearchHub is concerned with robotics.
It would have been obvious for one of ordinary skill in the art before the effective filing date to have
modified the design and control of soft robots simulation of Bacher with the computational design tool for compliant mechanisms as taught by DisneyResearchHub to enable another standard means of allowing the user to modify one or more of the rigid links to be realized as compliant. (Video Description)
Such modification also allows the system to allow the designer’s input to specify which of Bacher’s rigid links are to be replaced, so that those links can be modeled as compliant elements in the subsequent simulation.
Bacher and DisneyResearchHub fail to expressly disclose:
performing a remeshing-free parameterization of a volumetric rest shape of the one or more flexible links using splines;
Hafner discloses of optimizing CAD models with extended finite elements, comprising:
performing a remeshing-free parameterization of a volumetric rest shape of the one or more flexible links using splines; (as per “To dodge remeshing discontinuities and avoid a dependence of shape derivatives on the simulation mesh, we intersect the CADmodel with a regular hexahedral grid that we keep constant throughout optimizations” in P2, 1 INTRODUCTION, as per “In modern CAD systems, a boundary representation (B-rep), predominantly composed of Non-Uniform Rational Basis Spline (NURBS) patches, is used to describe solid models.” in P1, 1 INTRODUCTION, as per “We rely on projective coordinates to represent NURBS patches, where points [x,y,z]T in Euclidean coordinates are represented with points [wx,wy,wz,w]T in projective space P3. We therefore assume a NURBS patch with control points qi,j ∈ P3 and polynomialbasis functions Bi,j : R2 → R to be a parametric mapping” in P3, 3.1 CAD Model Representation, as per “first generic type of objective we seek to optimize integrates a function д that depends on the elastic response of the model over the volume enclosed by the B-rep” in P4, 3.3 Optimizing CAD Models, as per Eqn. 4)
In this way, Hafner operates to embed the CAD model in a regular hexahedral simulation grid that remains constant throughout optimizations, thereby removing the need to mesh or remesh during shape optimization (P2, 1 INTRODUCTION). Like Bacher and DisneyResearchHub, Hafner is concerned with differentiable simulation and shape optimization of deformable bodies.
It would have been obvious for one of ordinary skill in the art before the effective filing date to have modified the design and control of soft robots simulation of Bacher and the computational design tool for compliant mechanisms of DisneyResearchHub with the extended finite element optimization of CAD models as taught by Hafner to dodge remeshing discontinuities during rest shape optimization of the flexible links, as expressly suggested by Bacher itself at P5, 2.3 Differentiability of Simulation Representations ("Alternatively, the soft robot can be embedded in a simulation grid which remains constant throughout optimizations [20], enabling soft robot design optimizations directly on CAD representations," where reference [20] is Hafner).
Such modification also allows the system to parameterize the volumetric rest shape of Bacher's flexible links using NURBS spline control points on the CAD representation, so that the rest shape of those flexible links can be optimized directly on the CAD model without remeshing the simulation mesh as shape parameters change.
PNG
media_image6.png
275
487
media_image6.png
Greyscale
As per Claim 2, the combination of Bacher, DisneyResearchHub, and Hafner or suggests all limitations of Claim 1. Bacher further discloses wherein the user-defined operations comprise load-displacement samples for one or more points of interest on the robot mechanism. (as per “The important question is then how to find optimal values for the set of material parameters pmat that best explain the captured behavior (e.g., by comparing the simulated to the observed behavior at a few marker locations on the surface of the robot).” in P2, Mathematical Tools for Optimal Design and Control, as per Fig. 2)
As per Claim 3, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 2. Bacher further discloses wherein the simulating step comprises setting one or more actuators of the robot mechanism to a particular configuration and forces to user-specified forces in the load-displacement samples. (as per “In the process of developing accurate simulation models, it is key to be able to quantify if a simulator provides us with the desired prediction accuracy, and to detect where it fails to model the underlying physical behavior. To this end, a natural approach is to compare the simulated behavior to captured data, under matching actuation and external forces (see Fig. 2 top). This characterization can either be performed on material specimens [10] and individual sensors and actuators [11], or on the full robot [12].” in P2, Mathematical Tools for Optimal Design and Control)
As per Claim 4, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 2. Bacher further discloses wherein the simulated operation comprises simulated displacements and comparing the simulated operation to the user-defined operations comprises comparing the simulated displacements to user-specified displacements in the load-displacement samples. (as per “For characterization, sensing, actuation, or rest shape optimzation, one typically has a target state
x
-
given, and seeks to find the optimal set of parameters such that the difference between the simulated and target state, namely
1
2
|
x
p
-
x
-
|
2 or similar, is minimal” in P4, 2.1 Differentiable Simulation, as per Fig. 2)
PNG
media_image6.png
275
487
media_image6.png
Greyscale
As per Claim 5, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 1. Bacher further discloses wherein the definition of the robot mechanism further comprises a user selection of a material to be assigned to the one or more flexible links for the simulating step. , (as per “A soft robot made of passive materials, and actuated through external interactions, has a rest shape with an assigned material at every point, displaying a locally or globally homogenous or inhomogenous; isotropic or anisotropic behavior. Both the rest shape and the material assignment influence the deformation behavior of the robot. Representing these shape and material parameters with a parameter vector pdes, we seek to solve for their optimal values to, for example, maximize the contact surface or forces between the soft robot and objects of varying shape” in P3, Mathematical Tools for Optimal Design and Control)
PNG
media_image7.png
735
515
media_image7.png
Greyscale
As per Claim 6, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 1. Bacher further discloses wherein the simulator comprises a differentiable quasi-static simulator. (as per Fig. 3)
As per Claim 7, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 6. Bacher further discloses wherein the differentiable quasistatic simulator is configured to enforce coupling between flexible-flexible and flexible-rigid link pairs in the robot mechanism after inclusion of the one or more flexible links using constraints in a Lagrangian formulation. (as per “important to be able to formulate design and control optimizations for robots consisting of rigid and flexible components, coupled to one another with constraints. Flexible multibody simulation is a well-studied problem. However, its differentiability has received less attention. We point the interested reader to design and control approaches (quasi-statics [24]; dynamics [12, 25]) for examples on how to expand the mathematical tools for soft bodied models (Sec. 2.1) to handle systems with rigid bodies and coupling constraints between flexible and rigid components, or other constraint types.” In P5, 2.4 Rigid-Flexible Systems, as per “Embedded actuators or sensors can lead to another source of discontinuities. Due to the discontinuity of the deformation gradient at element boundaries for various standard finite elements (e.g., ones that reply on Lagrange shape functions), the strain field is not differentiable everywhere within the volume enclosed by the soft robot.” in P5, 2.3 Differentiability of Simulation Representations)
As per Claim 8, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 1. Bacher further discloses wherein the optimizing step is performed until the comparing step indicates a match between the simulated operation and a load-displacement profile at user-specified points of interest on the robot mechanism. (as per “Using g as a placeholder for an arbitrary objective that depends on the deformed state of the robot, we seek to solve the minimization problem
min
g
p
,
x
p
s
.
t
.
f
p
,
x
p
=
0
… constraining the deformed configuration to be an equilibrium.” in P4, 2.1 Differentiable Simulation, as per “The important question is then how to find optimal values for the set of material parameters pmat that best explain the captured behavior (e.g., by comparing the simulated to the observed behavior at a few marker locations on the surface of the robot). If the prediction accuracy is insufficient for the task at hand, the optimal characterization can help
us with the identification of inaccuracies in the simulation model (sim-to-real gap).” in P2, Mathematical Tools for Optimal Design and Control)
As per Claim 10, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 1. Bacher further discloses wherein the shape is an at rest configuration of the one or more flexible links. (as per “A soft robot made of passive materials, and actuated through external interactions, has a rest shape with an assigned material at every point, displaying a locally or globally homogenous or inhomogenous; isotropic or anisotropic behavior. Both the rest shape and the material assignment influence the deformation behavior of the robot. Representing these shape and material parameters with a parameter vector pdes, we seek to solve for their optimal values to, for example, maximize the contact surface or forces between the soft robot and objects of varying shape” in P3, Mathematical Tools for Optimal Design and Control, as per “for rest shape optimization, the topology of the simulation mesh is changing over time (remeshing), constituting a discontinuous operation. Meshfree methods [19] can remedy this problem. Alternatively, the soft robot can be embedded in a simulation grid which remains constant throughout optimizations” in P5, 2.3 Differentiability of Simulation Representations)
As per Claim 11, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 1. Bacher further discloses a robot mechanism fabricated based on the definition of the robot mechanism of claim 1 (as per “Moreover, we discuss how design optimization can help to optimally place soft actuators and sensors” in Abstract) and including the one or more flexible links fabricated based on the optimized shape generated by performance of the method of claim 1. (as per “We also list the optimization variables considered: actuation (or control) variables, shape, material/meta-material design, or actuation or sensor layout parameters” in P6, State of the Art)
As per Claim 12, Bacher discloses of design and control of soft robots using differentiable simulation, comprising:
with a computing device, accessing user input including a definition of a robot mechanism with a plurality of rigid links, (as per “A soft robot made of passive materials, and actuated through external interactions, has a rest shape with an assigned material at every point, displaying a locally or globally homogenous or inhomogenous; isotropic or anisotropic behavior. Both the rest shape and the material assignment influence the deformation behavior of the robot. Representing these shape and material parameters with a parameter vector pdes, we seek to solve for their optimal values to, for example, maximize the contact surface or forces between the soft robot and objects of varying shape” in P3, Mathematical Tools for Optimal Design and Control, as per “important to be able to formulate design and control optimizations for robots consisting of rigid and flexible components, coupled to one another with constraints.” in 2.4 Rigid-Flexible Systems) and
load-displacement samples for one or more points of interest on the robot mechanism; (as per “The important question is then how to find optimal values for the set of material parameters pmat that best explain the captured behavior (e.g., by comparing the simulated to the observed behavior at a few marker locations on the surface of the robot).” in P2, Mathematical Tools for Optimal Design and Control, as per Fig. 2)
with a simulator running on the computing device, simulating operation of the robot mechanism with the one or more flexible links to generate a simulated operation of the robot mechanism; (as per “To simulate a soft robot, we first discretize these entities using, for example, the finite element method (FEM).” in P3, 2.1 Differentiable Simulation, as per “For characterization, sensing, actuation, or rest shape optimzation, one typically has a target state
x
-
given, and seeks to find the optimal set of parameters…” in P4, 2.1 Differentiable Simulation)
with an optimizer running on the computing device, comparing the simulated operation of the robot mechanism to the load-displacement samples; (as per “For characterization, sensing, actuation, or rest shape optimzation, one typically has a target state
x
-
given, and seeks to find the optimal set of parameters… , and seeks to find the optimal set of parameters such that the difference between the simulated and target state… is minimal” in P4, 2.1 Differentiable Simulation, as per “a natural approach is to compare the simulated behavior to captured data, under matching actuation and external forces (see Fig. 2 top). This characterization can either be performed on material specimens [10] and individual sensors and actuators [11], or on the full robot [12].” in P2, Mathematical Tools for Optimal Design and Control, as per “The important question is then how to find optimal values for the set of material parameters pmat that best explain the captured behavior (e.g., by comparing the simulated to the observed behavior at a few marker locations on the surface of the robot).” in P2, Mathematical Tools for Optimal Design and Control, as per Fig. 2))
based on the comparing, modifying a shape of the one or more flexible links with the optimizer, wherein the shape is a rest configuration of the one or more flexible links. (as per “For rest shape or material distribution optimization, topology optimization has seen widespread use [31–36]. We can interpret traditional topology optimization as an application of differentiable simulation. However, because a linear elastic behavior is often assumed, it is, in its simplest form, not an ideal tool for soft robotics applications. To formulate shape optimizations, combinations of eXtended Finite Elements (XFE), level-set approaches that enable the continuous movement of material-material or material-void boundaries through elements, and analytical sensitivity analysis has seen use [15, 20, 37, 38]. This is again a variant of differentiable simulation that enables the optimization of soft robots consisting of composite materials [39]. Shape derivative techniques are also common [40].” in P6, 2.6 Alternative Optimization Techniques)
Bacher fails to expressly disclose:
wherein the user input further includes a selection of one or more of the rigid links for replacement with one or more flexible links;
performing a remeshing-free parameterization of a volumetric rest shape of the one or more flexible links using splines;
DisneyResearchHub discloses of a computational design tool for compliant mechanisms, wherein the user input further includes a selection of one or more of the rigid links for replacement with one or more flexible links. (as per 0:05-0:20 of video, as per 1:55-2:20 of video)
PNG
media_image2.png
1092
1458
media_image2.png
Greyscale
PNG
media_image3.png
968
1443
media_image3.png
Greyscale
PNG
media_image4.png
1044
1427
media_image4.png
Greyscale
PNG
media_image5.png
1002
1442
media_image5.png
Greyscale
In this way, DisneyResearchHub operates to take a rigidly-articulated mechanism and are automatically replaces components with parameterized flexures (Video Description). Like Bacher, DisneyResearchHub is concerned with robotics.
It would have been obvious for one of ordinary skill in the art before the effective filing date to have
modified the design and control of soft robots simulation of Bacher with the computational design tool for compliant mechanisms as taught by DisneyResearchHub to enable another standard means of allowing the user to modify one or more of the rigid links to be realized as compliant (Video Description). Such modification also allows the system to allow the designer’s input to specify which of Bacher’s rigid links are to be replaced, so that those links can be modeled as compliant elements in the subsequent simulation.
Bacher and DisneyResearchHub fail to expressly disclose:
performing a remeshing-free parameterization of a volumetric rest shape of the one or more flexible links using splines;
Hafner discloses of optimizing CAD models with extended finite elements, comprising:
performing a remeshing-free parameterization of a volumetric rest shape of the one or more flexible links using splines; (as per “To dodge remeshing discontinuities and avoid a dependence of shape derivatives on the simulation mesh, we intersect the CADmodel with a regular hexahedral grid that we keep constant throughout optimizations” in P2, 1 INTRODUCTION, as per “In modern CAD systems, a boundary representation (B-rep), predominantly composed of Non-Uniform Rational Basis Spline (NURBS) patches, is used to describe solid models.” in P1, 1 INTRODUCTION, as per “We rely on projective coordinates to represent NURBS patches, where points [x,y,z]T in Euclidean coordinates are represented with points [wx,wy,wz,w]T in projective space P3. We therefore assume a NURBS patch with control points qi,j ∈ P3 and polynomialbasis functions Bi,j : R2 → R to be a parametric mapping” in P3, 3.1 CAD Model Representation, as per “first generic type of objective we seek to optimize integrates a function д that depends on the elastic response of the model over the volume enclosed by the B-rep” in P4, 3.3 Optimizing CAD Models, as per Eqn. 4)
In this way, Hafner operates to embed the CAD model in a regular hexahedral simulation grid that remains constant throughout optimizations, thereby removing the need to mesh or remesh during shape optimization (P2, 1 INTRODUCTION). Like Bacher and DisneyResearchHub, Hafner is concerned with differentiable simulation and shape optimization of deformable bodies.
It would have been obvious for one of ordinary skill in the art before the effective filing date to have modified the design and control of soft robots simulation of Bacher and the computational design tool for compliant mechanisms of DisneyResearchHub with the extended finite element optimization of CAD models as taught by Hafner to dodge remeshing discontinuities during rest shape optimization of the flexible links, as expressly suggested by Bacher itself at P5, 2.3 Differentiability of Simulation Representations ("Alternatively, the soft robot can be embedded in a simulation grid which remains constant throughout optimizations [20], enabling soft robot design optimizations directly on CAD representations," where reference [20] is Hafner).
Such modification also allows the system to parameterize the volumetric rest shape of Bacher's flexible links using NURBS spline control points on the CAD representation, so that the rest shape of those flexible links can be optimized directly on the CAD model without remeshing the simulation mesh as shape parameters change.
As per Claim 13, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 12. Bacher further discloses wherein the simulating step comprises setting one or more actuators of the robot mechanism to a particular configuration and forces to user-specified forces in the load-displacement samples. (as per “it is key to be able to quantify if a simulator provides us with the desired prediction accuracy, and to detect where it fails to model the underlying physical behavior. To this end, a natural approach is to compare the simulated behavior to captured data, under matching actuation and external forces (see Fig. 2 top). This characterization can either be performed on material specimens [10] and individual sensors and actuators [11], or on the full robot [12].” in P2, Mathematical Tools for Optimal Design and Control, as per “we are then interested in how well the deformed configuration x, that results from solving the sensing problem to first-order optimality, matches the user-specified target.” in P4, 2.2 Nesting through First-Order Optimality Constraints)
PNG
media_image6.png
275
487
media_image6.png
Greyscale
As per Claim 14, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 12. Bacher further discloses wherein the simulated operation comprises simulated displacements and comparing the simulated operation to the load displacement samples comprises comparing the simulated displacements to user-specified displacements in the load-displacement samples. (as per “For characterization, sensing, actuation, or rest shape optimzation, one typically has a target state
x
-
given, and seeks to find the optimal set of parameters such that the difference between the simulated and target state, namely
1
2
|
x
p
-
x
-
|
2 or similar, is minimal” in P4, 2.1 Differentiable Simulation, as per
PNG
media_image6.png
275
487
media_image6.png
Greyscale
Fig. 2)
PNG
media_image7.png
735
515
media_image7.png
Greyscale
As per Claim 15, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 12. Bacher further discloses wherein the simulator comprises a differentiable quasi-static simulator. (as per Fig. 3)
As per Claim 16, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 15. Bacher further discloses wherein the differentiable quasistatic simulator is configured to enforce coupling between flexible-flexible and flexible-rigid link pairs in the robot mechanism after inclusion of the one or more flexible links using constraints in a Lagrangian formulation. (as per “important to be able to formulate design and control optimizations for robots consisting of rigid and flexible components, coupled to one another with constraints. Flexible multibody simulation is a well-studied problem. However, its differentiability has received less attention. We point the interested reader to design and control approaches (quasi-statics [24]; dynamics [12, 25]) for examples on how to expand the mathematical tools for soft bodied models (Sec. 2.1) to handle systems with rigid bodies and coupling constraints between flexible and rigid components, or other constraint types.” In P5, 2.4 Rigid-Flexible Systems, as per “Embedded actuators or sensors can lead to another source of discontinuities. Due to the discontinuity of the deformation gradient at element boundaries for various standard finite elements (e.g., ones that reply on Lagrange shape functions), the strain field is not differentiable everywhere within the volume enclosed by the soft robot.” in P5, 2.3 Differentiability of Simulation Representations)
As per Claim 17, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 12. Bacher further discloses optimizing the shape of the one or more flexible links by repeating the simulating step, the comparing step, and the modifying step (as per “Whenever we evaluate the objective g or its gradient after updating the set of parameters, we first solve for the state of the robot, represented by x and its time derivatives
x
˙
and
x
¨
, by integrating the dynamics system (Eq. 4) forward in time.” in P4, 2.1 Differentiable Simulation, as per Fig. 3) until the comparing step indicates a match between the simulated operation and a load-displacement profile at user-specified points of interest on the robot mechanism. (as per “Using g as a placeholder for an arbitrary objective that depends on the deformed state of the robot, we seek to solve the minimization problem
min
g
p
,
x
p
s
.
t
.
f
p
,
x
p
=
0
… constraining the deformed configuration to be an equilibrium.” in P4, 2.1 Differentiable Simulation, as per “The important question is then how to find optimal values for the set of material parameters pmat that best explain the captured behavior (e.g., by comparing the simulated to the observed behavior at a few marker locations on the surface of the robot). If the prediction accuracy is insufficient for the task at hand, the optimal characterization can help
us with the identification of inaccuracies in the simulation model (sim-to-real gap).” in P2, Mathematical Tools for Optimal Design and Control)
As per Claim 19, Bacher discloses of design and control of soft robots using differentiable simulation, comprising:
memory storing user input including a definition of a robot mechanism with a plurality of rigid links, (as per “A soft robot made of passive materials, and actuated through external interactions, has a rest shape with an assigned material at every point, displaying a locally or globally homogenous or inhomogenous; isotropic or anisotropic behavior. Both the rest shape and the material assignment influence the deformation behavior of the robot. Representing these shape and material parameters with a parameter vector pdes, we seek to solve for their optimal values to, for example, maximize the contact surface or forces between the soft robot and objects of varying shape” in P3, Mathematical Tools for Optimal Design and Control, as per “important to be able to formulate design and control optimizations for robots consisting of rigid and flexible components, coupled to one another with constraints.” in 2.4 Rigid-Flexible Systems)
a simulator provided by a processor executing instructions and configured to simulate operation of the robot mechanism with the one or more flexible links to generate a simulated operation of the robot mechanism; (as per “To simulate a soft robot, we first discretize these entities using, for example, the finite element method (FEM).” in P3, 2.1 Differentiable Simulation, as per “For characterization, sensing, actuation, or rest shape optimzation, one typically has a target state
x
-
given, and seeks to find the optimal set of parameters…” in P4, 2.1 Differentiable Simulation)
wherein the at-rest configuration (as per “has a rest shape with an assigned material at every point, displaying a locally or globally homogenous or inhomogenous; isotropic or anisotropic behavior. Both the rest shape and the material assignment influence the deformation behavior of the robot” in P3, Mathematical Tools for Optimal Design and Control) comprises the volumetric rest shape; (as per “A general soft robot may consist of one-dimensional entities such as artificial muscle fibers or stretch sensors, two-dimensional entities such as layers of active materials, and three-dimensional entities such as passive materials surrounding other entities.” in P3, 2.1 Differentiable Simulation, as per To simulate a soft robot, we first discretize these entities using, for example, the finite element method (FEM). Representing the deformed state with a set of discrete degrees of freedom x,” in P3, 2.1 Differentiable Simulation)
an optimizer provided by the processor executing the instructions and configured to compare the simulated operation of the robot mechanism to user-defined operations of the robot mechanism (as per “For characterization, sensing, actuation, or rest shape optimzation, one typically has a target state
x
-
given, and seeks to find the optimal set of parameters… , and seeks to find the optimal set of parameters such that the difference between the simulated and target state… is minimal” in P4, 2.1 Differentiable Simulation, as per “a natural approach is to compare the simulated behavior to captured data, under matching actuation and external forces (see Fig. 2 top). This characterization can either be performed on material specimens [10] and individual sensors and actuators [11], or on the full robot [12].” in P2, Mathematical Tools for Optimal Design and Control) and to modify an at-rest configuration of the one or more flexible links. (as per “For rest shape or material distribution optimization, topology optimization has seen widespread use [31–36]. We can interpret traditional topology optimization as an application of differentiable simulation. However, because a linear elastic behavior is often assumed, it is, in its simplest form, not an ideal tool for soft robotics applications. To formulate shape optimizations, combinations of eXtended Finite Elements (XFE), level-set approaches that enable the continuous movement of material-material or material-void boundaries through elements, and analytical sensitivity analysis has seen use [15, 20, 37, 38]. This is again a variant of differentiable simulation that enables the optimization of soft robots consisting of composite materials [39]. Shape derivative techniques are also common [40].” in P6, 2.6 Alternative Optimization Techniques)
Bacher fails to expressly disclose:
wherein the user input further includes a selection of one or more of the rigid links for replacement with one or more flexible links;
performing a remeshing-free parameterization of a volumetric rest shape of the one or more flexible links using splines;
DisneyResearchHub discloses of a computational design tool for compliant mechanisms, wherein the user input further includes a selection of one or more of the rigid links for replacement with one or more flexible links. (as per 0:05-0:20 of video, as per 1:55-2:20 of video)
PNG
media_image2.png
1092
1458
media_image2.png
Greyscale
PNG
media_image3.png
968
1443
media_image3.png
Greyscale
PNG
media_image4.png
1044
1427
media_image4.png
Greyscale
PNG
media_image5.png
1002
1442
media_image5.png
Greyscale
In this way, DisneyResearchHub operates to take a rigidly-articulated mechanism and are automatically replaces components with parameterized flexures (Video Description). Like Bacher, DisneyResearchHub is concerned with robotics.
It would have been obvious for one of ordinary skill in the art before the effective filing date to have
modified the design and control of soft robots simulation of Bacher with the computational design tool for compliant mechanisms as taught by DisneyResearchHub to enable another standard means of allowing the user to modify one or more of the rigid links to be realized as compliant (Video Description). Such modification also allows the system to allow the designer’s input to specify which of Bacher’s rigid links are to be replaced, so that those links can be modeled as compliant elements in the subsequent simulation.
Bacher and DisneyResearchHub fail to expressly disclose:
performing a remeshing-free parameterization of a volumetric rest shape of the one or more flexible links using splines;
Hafner discloses of optimizing CAD models with extended finite elements, comprising:
performing a remeshing-free parameterization of a volumetric rest shape of the one or more flexible links using splines; (as per “To dodge remeshing discontinuities and avoid a dependence of shape derivatives on the simulation mesh, we intersect the CADmodel with a regular hexahedral grid that we keep constant throughout optimizations” in P2, 1 INTRODUCTION, as per “In modern CAD systems, a boundary representation (B-rep), predominantly composed of Non-Uniform Rational Basis Spline (NURBS) patches, is used to describe solid models.” in P1, 1 INTRODUCTION, as per “We rely on projective coordinates to represent NURBS patches, where points [x,y,z]T in Euclidean coordinates are represented with points [wx,wy,wz,w]T in projective space P3. We therefore assume a NURBS patch with control points qi,j ∈ P3 and polynomialbasis functions Bi,j : R2 → R to be a parametric mapping” in P3, 3.1 CAD Model Representation, as per “first generic type of objective we seek to optimize integrates a function д that depends on the elastic response of the model over the volume enclosed by the B-rep” in P4, 3.3 Optimizing CAD Models, as per Eqn. 4)
In this way, Hafner operates to embed the CAD model in a regular hexahedral simulation grid that remains constant throughout optimizations, thereby removing the need to mesh or remesh during shape optimization (P2, 1 INTRODUCTION). Like Bacher and DisneyResearchHub, Hafner is concerned with differentiable simulation and shape optimization of deformable bodies.
It would have been obvious for one of ordinary skill in the art before the effective filing date to have modified the design and control of soft robots simulation of Bacher and the computational design tool for compliant mechanisms of DisneyResearchHub with the extended finite element optimization of CAD models as taught by Hafner to dodge remeshing discontinuities during rest shape optimization of the flexible links, as expressly suggested by Bacher itself at P5, 2.3 Differentiability of Simulation Representations ("Alternatively, the soft robot can be embedded in a simulation grid which remains constant throughout optimizations [20], enabling soft robot design optimizations directly on CAD representations," where reference [20] is Hafner).
Such modification also allows the system to parameterize the volumetric rest shape of Bacher's flexible links using NURBS spline control points on the CAD representation, so that the rest shape of those flexible links can be optimized directly on the CAD model without remeshing the simulation mesh as shape parameters change.
As per Claim 20, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 19. Bacher further discloses wherein the user-defined operations comprise load-displacement samples for one or more points of interest on the robot mechanism. (as per “The important question is then how to find optimal values for the set of material parameters pmat that best explain the captured behavior (e.g., by comparing the simulated to the observed behavior at a few marker locations on the surface of the robot).” in P2, Mathematical Tools for Optimal Design and Control, as per Fig. 2)
As per Claim 21, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 19. Bacher further discloses wherein the simulator comprises a differentiable quasi-static simulator. (as per Fig. 3)
PNG
media_image7.png
735
515
media_image7.png
Greyscale
As per Claim 22, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 21. Bacher further discloses wherein the differentiable quasistatic simulator is configured to enforce coupling between flexible-flexible and flexible-rigid link pairs in the robot mechanism after inclusion of the one or more flexible links using constraints in a Lagrangian formulation. (as per “important to be able to formulate design and control optimizations for robots consisting of rigid and flexible components, coupled to one another with constraints. Flexible multibody simulation is a well-studied problem. However, its differentiability has received less attention. We point the interested reader to design and control approaches (quasi-statics [24]; dynamics [12, 25]) for examples on how to expand the mathematical tools for soft bodied models (Sec. 2.1) to handle systems with rigid bodies and coupling constraints between flexible and rigid components, or other constraint types.” In P5, 2.4 Rigid-Flexible Systems, as per “Embedded actuators or sensors can lead to another source of discontinuities. Due to the discontinuity of the deformation gradient at element boundaries for various standard finite elements (e.g., ones that reply on Lagrange shape functions), the strain field is not differentiable everywhere within the volume enclosed by the soft robot.” in P5, 2.3 Differentiability of Simulation Representations)
As per Claim 24, the combination of Bacher, DisneyResearchHub, and Hafner teaches or suggests all limitations of Claim 20. Bacher further discloses wherein the simulator is configured to set one or more actuators of the robot mechanism to a particular configuration and forces to user- specified forces in the load-displacement samples to simulate the operation of the robot mechanism. (as per "it is key to be able to quantify if a simulator provides us with the desired prediction accuracy, and to detect where it fails to model the underlying physical behavior. To this end, a natural approach is to compare the simulated behavior to captured data, under matching actuation and external forces (see Fig. 2 top). This characterization can either be performed on material specimens [10] and individual sensors and actuators [11], or on the full robot [12]." in P2, Mathematical Tools for Optimal Design and Control, as per "we are then interested in how well the deformed configuration x, that results from solving the sensing problem to first-order optimality, matches the user-specified target." in P5, 2.2 Nesting through First-Order Optimality Constraints)
As per Claim 25, the combination of Bacher and DisneyResearchHub teaches or suggests all limitations of Claim 24. Bacher further discloses wherein the simulated operation comprises simulated displacements. (as per "For characterization, sensing, actuation, or rest shape optimzation, one typically has a target state x given, and seeks to find the optimal set of parameters such that the difference between the simulated and target state, namely ½‖x(p)−x̄‖² or similar, is minimal" in P4, 2.1 Differentiable Simulation, as per Fig. 2, as per “Simulations typically take external contact forces, body forces such as gravity, and actuation parameters as input, and output a deformed configuration of the robot under these forces. In state estimation, external contact forces are assumed to be unknown, and one seeks to estimate these forces such that the simulated sensor readings, that depend on the resulting deformed configuration, explain the measured read ings well [14]” in P3, 2 Mathematical Tools for Optimal Design and Control)
As per Claim 26, the combination of Bacher and DisneyResearchHub teaches or suggests all limitations of Claim 25. Bacher further discloses wherein the optimizer is configured to compare the simulated operation to the user-defined operations by comparing the simulated displacements to user-specified displacements in the load-displacement samples. (as per "For characterization, sensing, actuation, or rest shape optimzation, one typically has a target state x given, and seeks to find the optimal set of parameters such that the difference between the simulated and target state, namely ½‖x(p)−x̄‖² or similar, is minimal" in P4, 2.1 Differentiable Simulation, as per "The important question is then how to find optimal values for the set of material parameters pmat that best explain the captured behavior (e.g., by comparing the simulated to the observed behavior at a few marker locations on the surface of the robot)." in P2, Mathematical Tools for Optimal Design and Control, as per Fig. 2)
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Megaro (NPL Title: A computational design tool for compliant mechanisms., Year 2017) discloses a computational design tool for compliant mechanisms.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to TYLER R ROBARGE whose telephone number is (703)756-5872. The examiner can normally be reached Monday - Friday, 8:00 am - 5:00 pm EST.
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, Ramon Mercado can be reached at (571) 270-5744. 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.
/T.R.R./Examiner, Art Unit 3658
/Ramon A. Mercado/Supervisory Patent Examiner, Art Unit 3658
Prosecution Insights