WO2015143281A1 - Monolithic, multi-axis force sensor - Google Patents

Abstract

A monolithic, multi-axis force sensor can be in the form of a laminated structure including a scaffold; a plurality of arms extending within and across the scaffold at distinct angles, wherein the arms include a structural support layer, a sensor layer including a strain-gauge alloy, and a flexible and electrically insulating polymer layer sandwiched between the structural support layer and the sensor layer in a monolithic, multi-layered laminate structure; and electrically conductive pathways positioned to deliver a voltage through the strain-gauge alloy in the arms.

Description

MONOLITHIC, MULTI-AXIS FORCE SENSOR

BACKGROUND

The advent of minimally invasive surgical (MIS) techniques has cultivated a paradigm shift in surgery. Procedures once requiring a large incision, resulting in significant morbidity and recovery times, can now be performed through a discrete number of millimeter-sized ports, which can be quickly stitched or patched. Robotic MIS surgery, in particular, is experiencing increasing commercialization and adaptation for a limited subset of procedures. However, in light of substantial advancements in robotic control and dexterity, a haptic chasm still separates the surgeon from his or her anatomical workspace. When interacting with delicate anatomy, the lack of haptic feedback can lead to numerous complications including intraoperative hemorrhage, tissue damage, and suture breaking.

Haptic feedback has been explored in many areas of surgery, including laparoscopic surgery, microsurgery, and vitreoretinal surgery. Haptic feedback implementations can be broken into kinesthetic (force sensing) and cutaneous

(tactile sensing) modalities. Although cutaneous implementations necessitate distal sensing, as shown in FIG. 2, to accurately reconstruct the tactile profile, such a requirement is relaxed for kinesthetic sensing where only the magnitude of the applied force is desired. Coupling these two modalities can offer a full reconstruction of the applied force over the regime of interest.

Numerous examples of force transduction elements for haptics and MIS exist in literature. These elements range from optical fiber Bragg grating (FBG)

modalities, microfabricated lead zirconate titanate (PZT), semiconductor-based strain gages, and soft microchannels filled with liquid metal to name a few. These approaches often suffer from prohibitively high costs of manufacturing and assembly, exceedingly complicated or expensive signal conditioning infrastructures, susceptibility to thermal drift, limited range of applied forces, and limitations on achievable linearity and resolution. In addition, due to strict size constraints imposed by MIS, the force sensors are often placed proximally (or remotely with respect to the tool/end-effector), as shown in FIG. 1. As such, measured forces at the tool tip are contaminated by friction and reaction forces at the point of entry, as well as by actuation forces and by the mechanics of the tool itself, which makes this proximal placement less desirable than distal placement of the sensor, as shown in FIG. 2.

SUMMARY

A monolithic, multi-axis force sensor and methods for its fabrication and use are described herein. Various embodiments of the apparatus and methods may include some or all of the elements, features, and steps described below.

A monolithic, multi-axis force sensor can be in the form of a laminated structure including a scaffold; a plurality of arms extending within and across the scaffold at distinct angles, wherein the arms include a structural support layer, a sensor layer including a strain-gauge alloy, and a flexible and electrically insulating polymer layer sandwiched between the structural support layer and the sensor layer in a monolithic, multi-layered laminate structure; and electrically conductive pathways positioned to deliver a voltage through the strain-gauge alloy in the arms.

The multi-axis force sensor can be fabricated by aligning a polymer layer comprising a flexible and electrically insulating polymer on a substrate layer;

aligning a sensor layer comprising a strain-gauge alloy on the polymer layer; and joining the substrate layer, the polymer layer, and the sensor layer to form a monolithic multi-layered laminate structure, wherein the substrate layer forms a scaffold, and wherein the sensor layer is included in a plurality of arms extending within and across the scaffold and is configured to produce a change in electrical resistance through the sensor layer as the arms are displaced with strain.

The monolithic sensor can be used for multi-axis force sensing by contacting an object with a remote-controlled tool coupled with the monolithic force sensor. In the method, an electrical current is passed through the strain-gauge alloy as the tool contacts the object. The resistivity of the strain-gauge alloy is tracked as the tool contacts the object. A determination is made of the force vectors applied to the tool as it contacts the object as a function of changes in the resistivity of the strain-gauge alloy. Those force vectors are communicated to an operator {e.g., a surgeon) operating the tool via remote control. The inventors have rapidly prototyped customized, highly sensitive, mm-scale multi-axis force sensors for medical applications. Using a composite laminate batch fabrication process with biocompatible constituent materials, the inventors have fabricated a fully-integrated, 10 x 10 mm three-axis force sensor with up to 5 V/N sensitivity and root-mean-square (RMS) noise on the order of -1.6 mN, operational over a range of -500 to 500 mN in the x- and /-axes, and -2.5 to 2.5 N in the ^-axis. Custom foil-based strain sensors were fabricated in parallel with the mechanical structure, obviating the need for post-manufacturing alignment and assembly. The sensor and its custom-fabricated signal conditioning circuitry fit within a 1 x 1 x 2 cm volume to realize a fully integrated force transduction platform with potential haptics and control applications in minimally invasive surgical tools. The form factor, bio compatibility, and cost of the sensor and signal conditioning makes this method advantageous for the rapid prototyping of low-cost, mm-scale distal kinesthetic force sensors. Sensor performance is validated in a simulated tissue palpation task using a robotic master-slave platform.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGS. 1 and 2 show proximal and distal sensing, respectfully with a sensor 12 mounted on a surgical tool, wherein the sensor 12 is outside the skin in FIG. 1 and under the skin in FIG. 2.

FIG. 3 shows a triaxial force sensor 12 with a Maltese cross morphology.

FIG. 4 shows the deformation modes (exaggerated for clarity) and

characteristic flexure surface strain profiles of the Maltese cross of FIG. 3 due to loads in the ^-direction, and loads in the x- and y- directions. The accompanying plots show the (normalized) characteristic strain profiles within the flexural beams (arms) 16 as functions of position x along the arm 16 (where x = 0 at the attachment and x = 1 at the probe).

FIG. 5 illustrates a finite element analysis of sensor performance under loading in the ^-direction.

FIG. 6 illustrates fine element analysis of sensor performance under loading in the ^-direction. FIG. 7 shows a signal conditioning circuit schematic, wherein a tunable half- bridge feeds an instrumentation amplifier with a gain of 1000 and a voltage offset of

-₂Vcc

FIG. 8 shows uniaxial deformation for an infinitesimal element of a strain gage 18.

FIG. 9 shows an optimization analysis ofx- and /-direction sensitivities, where the inputs are arm width, w, and arm thickness, t, and the output is the resulting voltage normalized for a 1 N load. The labeled contours give mechanical factors-of- safety.

FIG. 10 includes an exploded view of a sensor, showing structural, sensing, and encapsulation sublaminates.

FIG. 11 shows millimeter-scale triaxial force sensors 12 as fabricated in a batch-manufacturing process. At left, sensors 12 are shown pre-release, while still attached to an alignment scaffold 14. At center bottom is a flat sensor 12 post-release. At bottom right, a sensor 12 is shown with stiffening struts folded up and locked into place with a ball of solder.

FIG. 12 is a top view of a sensor 12 with callouts to various wiring and assembly features; and a magnified image (obtained from a scanning electron microscope) of the manufactured sensor before encapsulation.

FIG. 13 shows a test setup for obtaining sensor calibration data.

FIG. 14 shows a calibration curve obtained for differential output voltage using the test setup of FIG. 13 as a function of force.

FIG. 15 shows the dynamic response of the sensor in the test setup of FIG. 13 to a step load in the ^-direction is measured and compared with predicted values, where the time constant, τ = 9 ms; the rise time, t_r = 18 ms; and the 95% settling time, t_ss = 31 ms.

FIG. 16 shows a quantitative palpation evaluation obtained via an

experimental setup.

FIG. 17 shows a measurement using the experimental setup of FIG. 16, wherein a characteristic palpation force profile demonstrating a dominant z- component is measured. FIG. 18 shows an experimental curve produced from an Ecoflex-0010 indentation test with the setup of FIG. 16 compared with analytical Hertzian and Mooney-Rivlin models.

FIGS. 19-21 illustrate a high-bandwidth, 2-axis, arm force sensor formed by folding a monolithic laminar structure, wherein FIG. 19 is a top view, preassembly; FIG. 20 is a front view, after folding and locking; and FIG. 21 is an isometric view showing sensitive axes.

FIGS. 22 and 23 illustrate stacked triaxial force sensors 12, which can be fabricated via folding, as shown in FIGS 19-21. When the layers are "popped up," they enable 6-axis force sensing because strain gages 18 on perpendicular faces allow for torque sensing; the six axes of sensing are as follows: F_x, F_y, F_z, ¾ r , and τ_ζ.

FIG. 24 shows the fabrication, assembly and release steps in a schematic representation of the fabrication of a pop-up MEMS device.

FIG. 25 illustrates four stages (a)-(d) in the formation of folding joints 21. FIGS. 26-28 provide schematic illustrations of folding and locking steps with brass plates 47 and solder 48.

In the accompanying drawings, like reference characters refer to the same or similar parts throughout the different views; and apostrophes are used to

differentiate multiple instances of the same or similar items sharing the same reference numeral. The drawings are not necessarily to scale; instead emphasis is placed upon illustrating particular principles in the exemplifications discussed below.

DETAILED DESCRIPTION

The foregoing and other features and advantages of various aspects of the invention(s) will be apparent from the following, more -particular description of various concepts and specific embodiments within the broader bounds of the invention(s). Various aspects of the subject matter introduced above and discussed in greater detail below may be implemented in any of numerous ways, as the subject matter is not limited to any particular manner of implementation. Examples of specific implementations and applications are provided primarily for illustrative purposes. Unless otherwise defined, used or characterized herein, terms that are used herein (including technical and scientific terms) are to be interpreted as having a meaning that is consistent with their accepted meaning in the context of the relevant art and are not to be interpreted in an idealized or overly formal sense unless expressly so defined herein. For example, if a particular composition is referenced, the composition may be substantially, though not perfectly pure, as practical and imperfect realities may apply; e.g., the potential presence of at least trace impurities {e.g., at less than 1 or 2%) can be understood as being within the scope of the description; likewise, if a particular shape is referenced, the shape is intended to include imperfect variations from ideal shapes, e.g., due to manufacturing

tolerances. Percentages or concentrations expressed herein can represent either by weight or by volume. Processes, procedures and phenomena described below can occur at ambient pressure {e.g., about 50-120 kPa— for example, about 90-110 kPa) and temperature {e.g., -20 to 50°C— for example, about 10-35°C).

Although the terms, first, second, third, etc., may be used herein to describe various elements, these elements are not to be limited by these terms. These terms are simply used to distinguish one element from another. Thus, a first element, discussed below, could be termed a second element without departing from the teachings of the exemplary embodiments.

Spatially relative terms, such as "above," "below," "left," "right," "in front,"

"behind," and the like, may be used herein for ease of description to describe the relationship of one element to another element, as illustrated in the figures. It will be understood that the spatially relative terms, as well as the illustrated configurations, are intended to encompass different orientations of the apparatus in use or operation in addition to the orientations described herein and depicted in the figures. For example, if the apparatus in the figures is turned over, elements described as "below" or "beneath" other elements or features would then be oriented "above" the other elements or features. Thus, the exemplary term, "above," may encompass both an orientation of above and below. The apparatus may be otherwise oriented {e.g., rotated 90 degrees or at other orientations) and the spatially relative descriptors used herein interpreted accordingly. Further still, in this disclosure, when an element is referred to as being "on," "connected to," "coupled to," "in contact with," etc., another element, it may be directly on, connected to, coupled to, or in contact with the other element or intervening elements may be present unless otherwise specified.

The terminology used herein is for the purpose of describing particular embodiments and is not intended to be limiting of exemplary embodiments. As used herein, singular forms, such as "a" and "an," are intended to include the plural forms as well, unless the context indicates otherwise. Additionally, the terms, "includes," "including," "comprises" and "comprising," specify the presence of the stated elements or steps but do not preclude the presence or addition of one or more other elements or steps.

Additionally, the various components identified herein can be provided in an assembled and finished form; or some or all of the components can be packaged together and marketed as a kit with instructions {e.g., in written, video or audio form) for assembly and/or modification by a customer to produce a finished product.

The inventors have developed a new class of cost-effective, mm-scale sensors that can be implemented distally to generate a pure measurement of the applied force at the tissue interface. Herein, an approach is presented that can be used to batch fabricate low-cost, high-resolution three-axis force sensors. Embodiments of the sensor can be constructed out of biocompatible materials and can fit within the size constraints imposed by MIS procedures.

These sensors can be manufactured using a process called pop-up MEMS, as described in published PCT Application No. WO 2012/109559 Al and in PCT

Application No. WO 2015/022952 Al, filed on 4 August 2014, wherein the mechanical structure, sensing elements, and signal conditioning printed circuit board (PCB) are all fabricated via stacking and bonding of layers in one integrated, monolithic manufacturing process. Proper arrangement of passive and active hinges 24, as shown in FIG. 12, in the laminate structure {e.g., by cutting rigid layers, while leaving a remaining connection via a flexible layer to form a hinge 24 in the laminate and then folding at the hinge 24) allows for out-of-plane features {e.g., oriented orthogonal to the remainder of the scaffold 14) to stiffen the overall structure and localize deformation to the sensitive elements.

Such a sensor 12 can be used, in particular, for applications in which the applied force is relatively low {e.g., less than 1 N), including tissue palpation and characterization, microsurgery, and vitreoretinal surgery, to name a few. The level of force detection can extend down to the limit of resolution, which can be 1.6 mN in the x- and y- axes and 16 mN in the z- axis. When the sensor 12 is used in a surgical tool, surgery on a patient can be performed using robotic tools, wherein the human surgeon (or other operator) controls the tool {e.g., by manually manipulating an input device, such as a "joystick") from a remote location. Readings of the force vectors (and/ or torque) sensed by the tool 22 (along three orthogonal axes) via the sensor 12 are processed and communicated {e.g., electronically and/or wirelessly) to the surgeon's input device; and the input device can be displaced and/or can change in stiffness via a vector/torque-force feedback mechanism in accord with the force vector and/ or torque measurements provided by the sensor 12 so that the surgeon can haptically "feel" a representation of the patient's internal tissues and other internal structures that contact the tool 22 via the surgeon's remote input device. The forces that are presented to the operator can be processed and/or amplified to improve human perception. The surgeon can also manipulate the input device, and the surgeon's movements can be recorded by the input device and communicated to actuators coupled with the tool 22 to manipulate the tool 22 in accord with the surgeon's motions.

In addition to surgical applications and other medical applications, such as biological tissue manipulation, the sensors 12 can also be used in a wide variety of applications, such as in portable electronics, where the sensors 12 can, for example, measure forces against the display screen of the device {e.g., when a user accidentally sits on the device) and may issue a warning to the user if the force is high enough to potentially damage the display screen. The sensors 12 can also be used in various other forms of instrumentation and metrology in a manner similar to the use of commercially available load cells and in other applications, where less-expensive, more-sensitive, lower-profile and lightweight sensors are used. For example, to provide force sensing for buttons and triggers in consumer products to give the user a range of input levels for a simple button interface {e.g., light, medium, or large forces cause three different behaviors in the device). Additionally, the sensor electronics [e.g., data acquisition, signal conditioning, signal processing, and communication {e.g., wireless)] can be an integral part of the sensor 12 so the electronics can operate as standalone measurement units individually or as part of an easily reconfigurable sensor network with modular units, thus saving the need for expensive and potentially bulky off-board signal processing and transmission electronics. In another example, the sensors 12 can be embedded in or placed on shelves, where they can provide a force reading to indicate how full the shelf is. The sensors 12 can also be used in a variety of assembly processes involving manipulation of fine and/or delicate components {e.g., watch assembly). In yet another

embodiment, the sensors 12 can used for viscous force measurement.

With a materials cost on the order of a few dollars for the sensor and electronics, such a platform presents an ultra-low-cost (<$10) alternative to commercial sensors (which often cost up to $l,000/axis) to realize smart, disposable sensors for some of the above-mentioned applications. For example, such a sensor can be used to create a disposable surgical tool 22 that can provide haptic feedback or can be tagged to items or shelves in a retail or industrial setting.

To leverage the capabilities of the pop-up MEMS process, a thin-arm "Maltese

Cross" topology is adopted, including a suspended platform ('probe') attached to a structural ground via axisymmetric flexures that transmit motion in the desired sensing degrees of freedom. This morphology is shown in FIG. 3, where four arms 16, each with a strain gage 18 layered thereon, extend across the sensor 12 (at orthogonal angles from one another) from the scaffold 14 to the central mount 20 for the tool 22. While the tool 22 here is shown as a probe, the tool 22 in other

embodiments {e.g., for surgical applications) can take other forms, such as a grasper or cutter. Signals generated by the strain gages 18 bonded to these flexures can be combined to determine the magnitude and direction of the force applied at the probe 22 in N axes so long as the number of gages 18, n_gages > N. Here, a three-axis sensor 12 is presented, where each arm 16 has two strain gages 18 (one in tension and one in compression). The combination of signals generated by each arm 16 under load can be reconstructed into a three-dimensional applied force vector.

Loading conditions experienced by the sensor 12 are shown in FIGS. 3 and 4. When a load in the ^-direction is applied axially to the probe tip, each flexure is in a state of pure bending with a vertical displacement u(x_b) (where x_b = 0 at the attachment to the structural ground). The resulting surface strain, ε_{ζ 1}, given as a function of position, x_b, along each beam (arm) 16, is described by Equation 1:

(1) where F_z is the applied load; h is the distance between strain and the arm's neutral axis; Eis the Young's Modulus; /is the second moment of area; and l_b is the length of the beam (arm) 16.

When a load is applied in the x- or ^-direction, a moment about the central platform produces an angular displacement Θ at the platform as given by Equation 2. The resulting surface strain s_x>i is given by Equation 3.

cx,l -^A-b ~ ,2 TH) ⁺ <> (IH) (3)

In the above equations, F_x is the applied load; ris the distance defined in FIG. 4; and L is the probe length.

These strain profiles, normalized and plotted as a function of distance along the arml6 (also normalized to arm length, l_b), are shown in the plots of FIG. 4 for each loading condition. In the case of ^--loading, the strain profile is antisymmetric about the midpoint of the arml6, whereas for x- or ^-loading, the zero strain crossing is slightly offset from the midpoint.

At the millimeter scale, the strain gage coverage area comprises a significant portion of the arm surface area, so it is inaccurate to consider only the maximum strain in gage analysis. Instead, the average value of strain integrated is considered over the strain gage length. Thus, the average strain, ε, is defined as follows: ε — ~_j

l9,f - ^l _T9~,i i, ^L9^g,i'^{f £}x,l (^xb)dx_b , where l_g>i and l_{g j} are the strain gage's start and end position, respectively, along length, x_b, of the beam (arm) 16.

After integration, the average strain, £_x, in the case of ^-loading (and y- loading) is given by the following equation:

Likewise, for loads in the ^-direction, the average strain, ε_ζ, is:

As the inventors were primarily interested in average strain, the length of the strain gage 18 was limited to half of the arm length (noting that if the strain gage length exceeds half of the arm length, desensitization occurs due to a reversal in the direction of strain as shown in the normalized strain plots in FIG. 4).

Methodology for Designing Custom Strain Gages:

Using a pop-up fabrication process using laminate structures, metal foil strain gages 18 can be custom-designed on an application-specific basis using a thorough understanding of strain-gage mechanics and an anticipation of the strain levels that the strain gage 18 is predicted to endure. Accurately predicting strain-gage performance pre-fabrication is advantageous to ensure that sensitivity requirements are satisfied in light of mechanical constraints.

When computing strain gage sensitivity, we consider a volume of strain gage material 18 with infinitesimal length, dx, in uniaxial tension, as shown in FIG. 8. Θ is defined as the resistivity of this volume (the resistance normalized by unit length). Under an applied uniaxial strain, which is a function of x, the resistance of this element ( )dx) is given by:

where w_g is the width; t_g is thickness; p is the resistivity of the strain gage material 18; and v is the Poisson ratio of the material.

Anticipating integration, Equation 7 can be simplified by linearizing the right- hand side about ε(χ) = 0 without significant loss of generality. For reasonable strains (i.e., those experienced during purely elastic deformation), any errors imposed by the linearized system can be shown to be negligible (<< 1%). The linearized

resistance can then be expressed as:

Integrating both sides over the length of the strain gage 18, l_g (assuming strain gage coverage beings at x = 0): ft θά^χ = ft(l + (1 + 2ν)ε)ά^χ; (9) u ^wg ^tg ^υ

·^'· ^R = ^wg ^tg i \^{1 + + 2ν}^ τ ^Lg ίο^{3 εάχ and} (¹⁰)

" R = R_mm + - Wg tg- l + 2v)e, (11) where the gage factor is given by S_e = (1 + 2v).

Thus, the strain gage resistance under loading can be expressed as a linear function of the nominal (unloaded) resistance and the average strain, which was computed previously. This methodology can be used to predict strain gage behavior when designing custom strain gages on an application-specific basis.

From a sensitivity standpoint, making the strain gage 18 as short as possible is advantageous. However, as the characteristic strain gage length is diminished, the effects of Joule heating become substantial (Qaiss ^K ^-T¹); resulting in thermal expansion. Although the geometric axisymmetry of the mechanical structure helps to cancel out some thermal expansion, significant gradients could still compromise sensor stability in terms of thermal drifting.

The previously derived analysis was used to guide a brute-force optimization routine to determine beam geometry where the objective function is to maximize sensitivity, (defined here as V_out/F_app). As the length of the arms 16 are roughly constrained by the outer dimensions of the sensor 12 (10 mm x 10 mm), arm width, w, and arm thickness, t, were chosen as free parameters in the optimization. The results of such an analysis for x and /loading is shown in FIG. 9, where a design sensitivity of approximately 5 V/N (satisfied by w= 1.25 mm and t = 0.20 mm) is chosen. Note that this design sensitivity corresponds to a mechanical factor of safety (defined as <J_y/a_max) of around 2-3 for a 1 N load, as shown by the contour lines. These dimensions result in a ^-sensitivity (analysis not shown) of roughly 0.6 V/N. Note that mechanical yield defines the geometry limit.

SENSOR DESIGN AND FABRICATION

The monolithic structure of the sensor 12 can be fabricated in parallel with the sensing elements to realize a fully-integrated multi-axis sensor 12 with no need for post-manufacturing alignment, bonding or assembly. Manufacturing:

The embodiment of the sensor 12 shown in FIG. 10 comprises a multi-material laminate composed of several functional sub-laminates. Rigid 304 Stainless steel shim stock 38 (four layers, where each layer is 50 μηι thick) forms the core of a structural support layer that dominates the mechanical behavior of the sensor 12. Kapton polyimide 36 (25 μηι thick) is a polymer used as (1) a flexible layer allowing active hinges to transmit motion and (2) an electrically insulating and encapsulating layer to isolate the strain gage 18. Constantan 40 (a 45% Cu / 55% Ni alloy, 5 μηι thick) is used as the strain gage alloy (in a sensor layer) due to its versatility and similarity to 304SS in terms of thermal expansion; and it offers an electrical resistivity that is substantially constant over a wide range of temperatures. DuPont Pyralux FR1500 sheet adhesive 34 (12.5 μηι thick) is used to bond subsequent layers. All materials with the exception of the adhesive 34 are biocompatible; a thin parylene [poly(p-xylylene polymer) coating can be deposited onto the sensor structure to cover the other layers for complete bio compatibility. Each of these layers can be laser machined to form the illustrated shapes and profiles. Additional layers include copper pads 32 and solder 42, wherein the copper pads 32 and solder 42 provide electrically conductive pathways and are positioned to deliver a voltage (electric current) through the strain-gage alloy 40 in each of the arms 16.

Of all modern strain-gauge alloys, constantan 40 is the oldest and still the most widely used. The acceptance of constantan 40 reflects the fact that constantan 40 has the best overall combination of properties needed for many strain gauge applications. This alloy has, for example, an adequately high strain sensitivity or gauge factor, which is relatively insensitive to strain level and temperature. Its resistivity is high enough to achieve suitable resistance values in even very small grids, and its temperature coefficient of resistance is not excessive. In addition, constantan 40 is characterized by good fatigue life and relatively high elongation capability. In other embodiments, an alternative strain-gage alloy, such as manganin (Cu86Mni2Ni₂) can be used.

The manufacturing process of the sensor 12 can be similar to that outlined in J. Gafford, et al, "Force-Sensing Grasper Enabled by Pop-Up Book MEMS" IEEE IROS, Tokyo, Japan (2013). Layers are individually machined using a diode-pumped solid-state (DPSS) laser, plasma etched with Argon gas to promote surface adhesion, and laminated in a heat press at 50 psi and 200°C for 2 hours. After lamination, the strain gage pattern is cut using the laser, and the sensor 12 is encapsulated in an additional Kapton layer (with breakout contacts) prior to the final release cuts. After release cuts are made, the stiffening struts are folded, locked, and soldered; and the sensor 12 is wired to the signal conditioning circuitry.

The fabrication of millimeter-scale triaxial force sensors 12 in a batch- manufacturing process is shown in FIG. 11 {i.e., many sensors 12 can be

manufactured in parallel across a large scaffold 14). In FIG. 11, at left, the sensors 12 are shown pre-release, while still attached to the alignment scaffold 14. At center bottom is a flat sensor 12 post-release. At bottom right, a sensor 12 is shown with stiffening struts folded up and locked into place with a ball of solder. A US penny provides scale.

Various assembly, wiring, and folding features are pre-programmed into the laminate. A combination of active hinges 24 (castellated hinges approximating pin- joint motion), plastic hinges (serrated material that plastically deforms when folded), and snapfit features allow for precision folding into a robust, 3-dimensional assembly which stiffens the structure and localizes deformation to the flexural elements. A microscope image of the sensor prototype, pre-encapsulation, is shown via the call- out in FIG. 12; the gage arm width here is shown to be 30 μηι.

Meanwhile, the main image in FIG. 12 provides a top view of the sensor 12, with the scaffold 14, arms 16, strain gage 18, central mount 20, copper 32 and solder 42 conductive pathways, and the mounting holes 26 in the scaffold 14. The mounting holes 26, which are provided in each layer, are aligned with fixed mounting pins that feed through the holes 26 as the layers are stacked so that each layer is precisely aligned with the others via the alignment of its mounting holes 26 with the mounting pins.

Finite Element Analysis:

Although the previously derived analytical model can be used to approximate the strain behavior of the sensor 12, the underlying assumption is that the flexure beam (arm) attachment points are themselves perfectly rigid. Commercial off-the- shelf (COTS) sensors typically satisfy this assumption by substantially increasing material thickness at the attachment. Using pop-up laminate structures, where structural rigidity is approximated using plastic hinges and features that fold out-of- plane and lock into place, the elastic behavior becomes more complicated at these interfaces. As such, finite element analyses (using Solidworks Simulation from

Dassault Systems) were performed to determine the location of maximum strain, and therefore, optimum strain gage placement.

Note that the arms 16 are attached to the scaffold 14 at corners for increased stiffness. As the finite element analysis results elucidate in FIGS. 5 and 6, which represent the forces illustrated in FIGS. 3 and 4, respectively, this attachment point is not perfectly rigid; and a small amount of local deformation occurs at the corners due to strain diffusion. Thus, placing the strain gages 18 on the probe half of the arms 16 (for optimum sensitivity) becomes beneficial, where the attachment of the probe 22 more closely approximates a rigid connection point. Due to symmetry, loading in the /-direction results in behavior that is substantially identical (beyond the directional difference) to behavior due to loading in the ^-direction.

Signal Conditioning:

An on-board signal conditioning circuit was designed to amplify the strain gage output such that reasonable transistor-transistor logic (TTL) voltage levels can be processed by a data acquisition system (DAQ). On-board sensing can ensure that resistive/inductive contributions from leadwires are minimized pre-amplification, so the design challenge is to minimize the footprint of the signal conditioning printed circuit board (PCB) to dimensions similar to those of the sensor 12, itself. As shown in FIG. 7, a half bridge with precision (0.05% tolerance) reference resistors feeds an AD8221 instrument amplifier with tunable gain. The amplifier gain was set to 1,000 such that a 500 mN load results in an ~2.5 V voltage swing. The half-bridge configuration ensures adequate temperature compensation and long-term stability. A high-impedance tuning potentiometer allows for manual zero-offset calibration coverage for up to 5% strain gage resistance mismatch with negligible effect on linearity and sensitivity. The sensor 12 is designed for a single supply, so a mid-level voltage reference is established by a buffered on-board voltage divider so that positive and negative forces can both be measured. The custom PCB, measuring 10 x 20 mm, was fabricated in-house via direct-write photolithography on a copper- cladded FR4 dielectric layer. TESTING AND VALIDATION

Sensor Calibration:

To reconstruct the applied force vector, a calibration matrix is defined that expresses each force component as a linear combination of the signals generated by each half-bridge. The objective is to formulate the calibration matrix [C]_3x4 that satisfies the following:

where f_x, f_y, and _z are the resolved forces; and s₁— s₄ are the bridge outputs.

The sensor 12 was calibrated in a benchtop setting using precision weights of known mass, as shown in FIG. 13. An example calibration curve for the -axis is shown in FIG. 14. As can be seen, the on-axis signals are linear and dwarf the off-axis signals for a sufficiently pure measurement of the ^-directional force. The on-axis sensitivity is roughly 5 V/N, as designed. Similar calibration curves were generated for the y- and z- axes to obtain the inverse calibration matrix satisfying s = C'_4x3 :

Note that the sensor 12 is significantly more sensitive in the x- and y- axes than in the z- axis due to the mechanical amplification of the probe. Sensitivity matching can be achieved simply by shortening the length of the probe 22.

To obtain the calibration matrix to convert from signal to source, the Moore- Penrose pseudoinverse is computed to our overconstrained system of equations. This entails the use of a least-squares algorithm to numerically compute the pseudoinverse matrix, C, which is given in the following equation:

Sample performance data for the -axis is given in Table I, below. The sensor 12 is designed to withstand mechanical loads of up to 2 N so the gain can simply be adjusted to accommodate a higher operational range given the 5 V supply. The root- mean-square (RMS) noise of the sensor 12, measured by integrating the power spectral density of a null signal, is roughly 8 mV (corresponding to 1.6 mN in the x- and /-axes). Note that, due to increased stiffness, the RMS noise is a factor of 10 higher in the ^-direction (roughly, 16 mN).

TABLE I: Performance data for the sensor in the ^-direction.

Dynamic Response: The dynamic response of the system is damped and of the second order, as shown in FIG. 15, where a step load of 50 mN was applied in the ^-direction, and the response of the system was measured. As the sensor 12 is most compliant in the x- and /-directions, dynamic behavior in these axes will limit the operational bandwidth. As can be seen, the dynamic behavior is accurately mapped by a second- order model of the form, x + 2ζνν^χ + WQ X = 0, where ζ = /l jmk is the damping coefficient; w₀ = ^jk/m is the natural frequency; A is the analytical stiffness of the sensor 12; m is the combined mass of the sensor 12 and probe; and β is damping added by the adhesive 34 and polyimide 36 layers (empirically determined). An interesting property is that the sensor 12 is almost critically damped (ζ = 0:88). The measured 10%-90% rise time is roughly 18 ms, limiting the bandwidth to about 20 Hz (using a first-order approximation for bandwidth given by f-_3dB = 0.35/t_r). Since there is a tradeoff between bandwidth (f-_{3 c}iB ^K ^²) ^and sensitivity (5 oc t^-2), increasing the thickness of the steel substrate can greatly improve the dynamic range at the cost of sensitivity. Tissue Palpa don Sim ula don:

A tissue palpation experiment was simulated, wherein the sensor 12 was attached to the end effector of a 3-degree-of-freedom (DoF) linear stage equipped with a 3-DoF wrist to realize a 6-DoF micromanipulation platform. An image of the test setup is shown in FIG 16. Three ball-screw linear stages (ATSlOO-100, Aerotech, Pittsburgh, PA) are mounted orthogonally to create the 3-DoF linear stage. The linear stages have 100 mm travel with a 0.5 μτη resolution and are connected to a control box (A3200 Npaq Drive Rack, Aerotech, Pittsburgh, PA) that runs an internal servo loop on the stages at 8 kHz. The 3-DoF wrist has a gimbal design with three direct drive rotary joints. Three 12-Watt 22-mm brushless DC motors (EC-max

283840, Maxon, Switzerland) are used to actuate the wrist. The motors are controlled using three digital positioning controllers (EPOS2 24/2, Maxon, Switzerland) that run an internal servo loop at 1 kHz. The micromanipulation stage is controlled through a graphical user interface (GUI) that handles communications with the Aerotech and Maxon motors through provided API (Application Programming

Interface) library calls. The GUI was programmed in C++ using the Qt application framework (Digia Pic, Helsinki, Finland).

A tissue analog was molded out of Ecoflex-0010 (Smooth-On), with high- stiffness intrusions (steel balls) embedded at different depths to simulate metastatic or cancerous tissue regimes. As a preliminary quantitative evaluation, a solid block of Ecoflex-0010 was palpated with a spherical sensor probe. The force profile generated by the sensor 12 was compared to an analytical force profile based on (1) a simple model assuming linear-elastic Hertzian contact mechanics and (2) a hyperelastic, neo-Hookean Mooney-Rivlin model given an empirical Young's

Modulus of E= 30 kPa for the Ecoflex-0010. The individual force components, as well as the stiffness curve, are shown in FIGS. 17 and 18. The sensor 12 accurately captures the nonlinear hyperelastic behavior of the elastomer, as well as the hysteretic behavior that is not captured in the model.

For the palpation task, the tissue analog was probed in an 8 x 8 grid with each subsequent probe separated from the previous probe by 9.2 mm. Steel balls were buried at depths of 1 mm (ball 1), 3 mm (ball 2), and 5 mm (ball 3). The indentation speed was 1 mm/ s, with a nominal indentation depth of 2 mm. A normalized stiffness contour plot of the tissue, as measured by the sensor 12, was obtained; and the sensor 12 was able to accurately reconstruct the stiffness map of the tissue and localize high-stiffness intrusions, with the exception of the deepest intrusion, which was 5 mm deep (most likely due to the shallow indentation depth). As described herein, the monolithic fabrication of a fully integrated, highly sensitive triaxial sensor 12 is demonstrated using a pop-up MEMS approach. The performance of the sensor 12 was tested both in benchtop calibration experiments as well as in a simulated tissue palpation task to demonstrate the sensor's efficacy at evaluating unknown forces with high precision. Preliminary characterization methods have shown that the sensor 12 can resolve forces with ~1 mN resolution at a frequency of 20 Hz. Additionally, the sensor 12 was able to accurately detect subtle stiffness changes in a simulated tissue.

Further discussion of the pop-up MEMS fabrication approach that can be used to manufacture the multi-axis force sensor 12 or components thereof is provided, below.

General Description of Pop-Up Laminate Structures

An embodiment of a pop-up laminate structure that can be distorted, flexed or folded (these terms may be used interchangeably herein) can be fabricated by, for example, forming a five -layer composite with the following sequence of layers: rigid layer, adhesive layer, flexible layer, adhesive layer, rigid layer. Alternatively, a thinner composite can be formed from a stacking of just a rigid layer, an adhesive layer, and a flexible layer, though this structure is not symmetric. The rigid layers are machined to have gaps that correspond to fold lines, while the flexible layer is continuous, thereby providing a joint where the flexible layer extends across the gaps machined from the rigid layers. The dimensions and feature sizes of the various apparatus described herein can be, e.g., 0.1 mm to 5 cm or, in more particular embodiments, 0.5 mm to 2 cm. Moreover, the devices described, below, can be mass- fabricated by forming a plurality of the laminate structures across a large-area multi- layer laminate from which the individual devices can then be popped by, e.g., severing sacrificial bridges that joint the devices to the rest of the large-area laminate.

Characterization of the structure as being "super-planar" means taking multiple planar layers and selectively connecting them. An analogy here can be drawn to circuit boards, where electrical vias connect circuits on different layers. Here, in contrast, the superplanar structure is made with "mechanical vias." By stacking multiple planar layers, the range of achievable devices is greatly expanded. The super-planar structure also enables features and components to be packed into the structure that would not fit if the device could only be made out of one planar sheet. Advantageously, super-planar structures with mechanisms that operate normal to the plane can now be made with these techniques. In practice, forming Sarrus linkages between planar layers is an advantageous strategy for designing an assembly mechanism/ scaffold. Other mechanisms can attach to the Sarrus links to effect the intended component rotations.

The multi-layer, super-planar structure can be fabricated via the following sequence of steps, which are further described, below: (1) machining each planar layer, (2) machining or patterning adhesives, (3) stacking and laminating the layers under conditions to effect bonding, (4) post-lamination machining of the multi-layer structure, (5) post-lamination treatment of the multi-layer structure, (6) freeing an assembly degree of freedom in each structure, (7) locking connections between structural members, (8) freeing any non-assembly degrees of freedom, and (9) separating finished parts from a scrap frame.

A schematic representation of a fabrication process is provided in FIG. 24, illustrating how the basic operations of micromachining 101, lamination on dowel pins 50 and pick-and-place 102, folding 103, locking 104, and additional

micromachining 105 can be arranged to manufacture machines 106.

These assembly techniques can include the formation of folding joints 21, as illustrated in steps (a)-(d) of FIG. 25, wherein (a) features are first micro-machined 101 in individual material layers, and the resulting chips 13 are removed; (b) during lamination, dowel pins 50 pass through apertures 54 that align material layers while heat and pressure are applied; here, two rigid carbon-fiber layers 52 bonded to a flexible polyimide-film layer 36 with adhesive 34 form a five-layer laminate 15 referred to as a "linkage sub-laminate"; (c) micro-machining 101 cuts mechanical bridges that constrain individual elements, allowing the creation of articulated structures; and (d) a completed folding joint 21 is formed and removed from the surrounding scaffold 14. The castellated pattern allows this flexure joint 21 to approximate an ideal revo lute joint. All assembly folds in a more-complex assembly can be incorporated into a single "pop-up" degree of freedom, which can be locked in place by a soldering process after pop up and then released by micro-machining.

Polyimide can have a flexural modulus of about 20 GPa; other materials {e.g., polymers) with a flexural modulus with a flexibility within about, e.g., 25%, 50% or 75% of this value (higher or lower) can alternatively be used for the flexible layer. The rigid layers can be, e.g., less than half as flexible {i.e., more than twice as stiff) as the flexible layer.

1) Machining of Layers

In one embodiment, the multi-layer structure is formed from a multitude of thin (1.5-μηι to about 150-μηι thick) layers of various materials.

These layers are laser micro-machined {e.g., by a diode-pumped solid-state pulsed laser) with desired features, usually cutting all the way through the layer to create individual planar structures. Each layer is micro-machined so as to leave a unified (contiguous) part with robust connections to surrounding alignment holes. The micro-machining can produce complex in-plane features with dimensions as small as 10 μηι. In particular embodiments, many copies of the pop-up laminate device are formed on a laminate panel, and the machining process removes sufficient material to form each part and part feature, while leaving thin tabs to connect each device to the surrounding laminate; in this regard, the arrangement of devices in a laminate panel can be similar to that of a batch of circuit boards attached to a surrounding laminate structure by thin, easily breakable tabs.

In this case, the tabs (bridges) 17 (shown in FIG. 25) connecting the devices to the surrounding laminate will be removed after lamination or assembly. Layers of metal, composite, polymer, etc., are machined or formed by virtually any method; and virtually any material may be used. Exemplary machining methods include laser cutting from sheet material, photo-chemical etching, punching, electroforming, electric discharge machining, eic.-basically any method that has appropriate resolution and compatibility with the desired material. Machined layers may then be subjected to additional processes, such as cleaning/etching to remove machining debris, plating {e.g., plating fluxed copper 32 on a layer to facilitate adhesion of solder thereto), preparation for bonding, annealing, etc. The unified nature of each layer makes handling and post-processing easy. Advantageously, each layer can be formed of a different material and can be machined and treated differently from each of the other layers.

Each layer can also advantageously be formed of a material that is sufficiently rigid, strong and tough to allow holes 54 for alignment pins 50 and other features to be machined into the layer to facilitate easy handling and to not distort (a) when placed into the layup and (b) when restrained by alignment pins 50. In other embodiments, layers that do not have the structural stability to support alignment features can nevertheless be used by attaching such layers, in bulk form, to a rigid frame that meets these objectives without introducing enough additional thickness to disturb the other layers or parts in the laminate.

In particular examples, a very thin polymer film {e.g., 2-5 microns thick) is included among the layers. Due to its thinness and insulating qualities, the thin polymer film is prone to wrinkling and electrostatic handling issues. To address this tendency, the thin polymer film can be lightly stretched, in bulk form, to a flat and controlled state and then bonded to a thin frame that is made, for example, of thin metal or fiberglass composite. Next, the thin polymer layer can be machined with the fine part features {e.g., tiny holes in the polymer at precise locations), and the alignment hole features can be machined into the frame material.

In additional embodiments, the device can be designed to mitigate thin-layer handling issues. For example, a part within the device can be designed such that all machining pertinent to a fragile layer is performed post-lamination; and, thus, this layer will not require precision alignment when put into the laminate, though the material is advantageously capable of being placed into the laminate sufficiently flat and extending over a sufficient area to cover the desired parts of the device.

In exemplary embodiments, bulk polymer films (formed, e.g., of polyester, polyimide, etc.); metal sheets and foils [formed, e.g., of stainless steel, spring steel, titanium, copper, invar (FeNi36), nickel-titanium alloy (nitinol), aluminum, etc.]; copper-clad laminates; carbon fiber and glass fiber composites; thermoplastic or thermoset adhesive films; ceramic sheets; etc.; can be laser machined to make the layers that are laminated to form the multi-layer structure. The laser machining can be performed, e.g., with a 355-nm laser (from DPSS Lasers Inc. of Santa Clara, California) with a spot size of about 7 microns on materials with typical thicknesses of l-150-μηι, although thicker layers can be machined with such a laser, well.

Accordingly, this type of laser allows for very high resolution and an ability to machine almost any type of material.

2) Machining or Patterning Adhesives

Adhesion between layers is achieved by patterning adhesive onto one or both sides of a non-adhesive layer or by using free-standing adhesive layers ("bondplies") 34. In the latter case, an intrinsically adhesive layer 34, e.g., in the form of a sheet of thermoplastic or thermoset film adhesive, or an adhesive laminate, such as a structural material layer with adhesive is pre-bonded to one or both sides. The adhesive layer 34 is machined like the other layers. Specific examples of sheets that can be used as the adhesive layer 34 include sheet adhesives used in making flex circuits {e.g., DuPont FR1500 adhesive sheet) or polyimide film 36 coated with FEP thermoplastic adhesive 34 on one or both sides. Free-standing sheet adhesives can be acrylic-based for thermosets; alternatively, the adhesive can be thermoplastic, wherein the thermoplastic film can be formed of polyester, fluorinated ethylene propylene (or other fluoropolymer), polyamide, polyetheretherketone, liquid crystal polymer, thermoplastic polyimide, etc. Any of these adhesives can also be applied on one or both sides to a non-adhesive carrier. In additional embodiments, a layer may serve both as a structural layer and as a thermoset adhesive 34— for example, liquid crystal polymer or thermoplastic polyimide. Furthermore, for special types of structural layers, a variety of wafer bonding techniques that do not require an adhesive may be employed, such as fusion bonding.

In another technique for achieving adhesion between layers, adhesive 34 is applied and patterned directly on a non-adhesive layer. This technique can be used where, for example, the type of adhesive desired may not be amenable to being in a free-standing form. Examples of such an adhesive 34 include solders, which are inherently inclined to form a very thin layer, or adhesives that are applied in liquid form (by spraying, stenciling, dipping, spin coating, etc.) and then b-stage cured and patterned. B-staged epoxy films are commonly available, but they usually cannot support themselves unless they are quite thick or reinforced with scrim.

The resulting bond can be a "tack bond," wherein the adhesive 34 is lightly cross-linked to an adjacent layer before laser micromachining with sufficient tack to hold it in place for subsequent machining and with sufficient strength to allow removal of the adhesive backing layer. The tack bonding allows for creation of an "island" of adhesive 34 in a press layup that is not part of a contiguous piece, which offers a significant increase in capability. Another reason for tacking the adhesive 34 to an adjacent structural layer is to allow for unsupported "islands" of adhesive 34 to be attached to another layer without having to establish a physical link from that desired adhesive patch to the surrounding "frame" of material containing the alignment features. In one embodiment, a photoimagable liquid adhesive, such as benzocyclobutene, can be applied in a thin layer, soft baked, and then patterned using lithography, leaving a selective pattern of adhesive. Other photoimagable adhesives used in wafer bonding can also be used.

The adhesive 34 is patterned while initially tacked to its carrier film, aligned to the structural layer using pins 50, and then tacked to at least one adjoining layer in the layup with heat and pressure {e.g., at 200°C and 340 kPa for one hour).

Alternatively, the adhesive layer can be patterned by micro-machining it as a free sheet. Tack bonding can involve application of heat and pressure at a lower intensity and for less time than is required for a complete bond of the adhesive. In yet another embodiment, the adhesive film 34 can be tack bonded in bulk, and then machined using, for example, laser skiving/ etching. Advantageously, use of this variation can be limited to contexts where the machining process does not damage the host layer. Both of these variations were tried using DuPont FR1500 adhesive sheet and laser skiving.

3) Stacking and Laminating the Layers

To form the multi-layer laminate structure, a multitude of these layers {e.g., up to 15 layers have been demonstrated) are ultrasonically cleaned and exposed to an oxygen plasma to promote bonding and aligned in a stack by passing several vertically oriented precision dowel pins 50 respectively through several alignment apertures 54 in each of the layers and by using a set of flat tooling plates with matching relief holes for the alignment pins 50. In other embodiments, other alignment techniques {e.g., optical alignment) can be used. All layers can be aligned and laminated together.

Linkages in the laminated layers can be planar (where all joint axes are parallel); or the joint axes can be non-parallel, allowing for non-planar linkages, such as spherical joints.

In a fifteen-layer example, the final layup includes the following layers, which formed a pair of linkages {i.e., structures wherein flexible layers 36, formed, e.g. , of polyimide, are bonded to rigid segments 52, formed, e.g., of carbon, and extend across the gaps between the rigid segments 52), thereby enabling flexure of the rigid segments 52 relative to one another at the flexible layer 36 in the gaps between the rigid segments 52, wherein those exposed sections of the flexible layer 36 effectively serve as joints.

The choice of the flexible layers 36, which can be formed of a polymer— polyimide in this example— is based upon compatibility with the matrix resin in the carbon fiber. The cure cycle can reach a maximum temperature of 177°C using a curing profile of four hours. Polyimide film (available, e.g., as KAPTON film from E.I. du Pont de Nemours and Company), for example, has a sufficiently high service temperature (up to 400°C) to survive the curing step. The polyimide film can have a thickness of, e.g., 7.5 μηι.

The rigid layers 52 in this embodiment are standard cured carbon fiber sheets {e.g., with three layers of unidirectional fibers, wherein the fiber layers are oriented at 0°, 90°, and 0° to provide thickness in two orthogonal directions), each sheet having a thickness of, e.g., 100 μηι. Fifteen layers are used because the adhesive sheet 34 {e.g., in the form of a B-staged acrylic sheet adhesive, commercially available, e.g., as DuPont PYRALUX FR1500 acrylic sheets) in this embodiment is separate from each layer of structural material in the layup of this embodiment. Accordingly, the adhesive sheet 34 can be laser machined into a pattern differing from any structural layer, and aligned layups of many layers can be made. This capability enables the fabrication of parts with many linkage layers that are perfectly or near-perfectly aligned.

After the layers are stacked to form the layup, pressure and heat are applied, typically in a heated platen press to cure/crosslink the adhesive layers. Specifically, the layup can be cured in a heated press, autoclave, or other device that provides the atmosphere (or lack thereof), temperature, and pressure to achieve the bonding conditions required by the adhesive. One embodiment of the curing process uses 50- 200 pounds-per-square-inch (psi) clamping pressure, 350°F (177°C) temperature, and two-hours cure time (optionally with temperature ramping control) to cure DuPont PYRALUX FR1500 acrylic sheets in a heated press with temperature, pressure, and atmosphere control.

4) Post-Lamination Machining

The laminate is then machined {e.g., by severing tabs with a laser) to release the device(s) from a surrounding frame structure in the laminate. In some

embodiments, additional machining that is not involved with freeing the device from the external frame (circumscribing the device in the laminate) is reserved for after lamination {e.g., post-lamination machining of a layer that is structurally weak or that, for some other reason, cannot be precisely aligned since the weak layer is better supported after lamination). 5) Post-Lamination Treatment

A post-lamination treatment can include plating or coating on an exposed layer; and/or the post-lamination treatment can include the addition of a material, such as solder paste, by silk screening or some other method, e.g., for the later joint "locking" step, as shown in FIGS. 26-28. Additional components may be attached to the laminate using a pick-and-place methodology. Pick-and-place operations can be used to insert discrete components into layups before press lamination.

For example, a stimulus responsive material, such as an electroactive material, can be inserted among the layers to serve as an actuator. In one embodiment, a lead zirconate titanate piezoelectric plate is mounted on a spring clip in the carbon layer 52 and has been demonstrated to create a functional bimorph cantilever actuator within a device.

Press lamination and laser micro-machining can be conducted multiple times. For example, five layers can be laser micro-machined, then press laminated, then laser micro-machined again. Another three layers can be separately laser micro- machined, then press laminated, then laser micro-machined again. These two partial layups can then be press laminated together with a single adhesive layer between them, for a final layup of nine layers.

6) Freeing the Assembly Degree of Freedom in Each Part The resulting laminate can then be laser micro-machined and/ or scrap materials can be removed from the laminate to "release" functional components in each part. The parts, as laminated, may unfold to have many actuated and passive mechanical degrees of freedom; though, in some embodiments, restraining these non-assembly degrees of freedom during the assembly folding process is

advantageous. For example, elements of a flexural linkage can be held in place (i.e., locked)~to prevent the linkages from flexing— by a rigid bar element alongside the elements or by a fixed tab forming an integral bridge between the elements and the surrounding structure. Using a machining process {e.g., punch die or laser cutting), the tabs or other features that restrain the assembly degree of freedom are severed. 7) Assembly

As fabricated, the pop-up laminate can be a flat multi-layer laminate with limited three dimensional structure. Its components undergo a variety of assembly trajectories to realize the final fully three-dimensional topology. A co-fabricated mechanical transmission called an "assembly scaffold" couples all of these assembly trajectories into a single degree of freedom. The pop-up laminate emerges from the manufacturing process as a three-degree-of-freedom machine, though internal mechanical connections eliminate these active degrees of freedom during assembly.

Assembly of the final device (including unfolding of linkages into multiple planes) can be performed manually by external actuation, or assembly can happen spontaneously. Where assembly is spontaneous, if one or more of the layers is pre- strained, the relaxation of the pre-strained layers can lead to the assembly of the device as soon as the assembly degree of freedom is freed. The layer that is pre- strained can be, for example, a patterned spring formed of spring steel or another spring-capable material, such as a superelastic nickel titanium alloy (nitinol) or an elastomer material that can survive the lamination conditions without annealing or degradation. The dowel pins and the pin alignment holes in the pre-strained layer can be configured to maintain this tension when the pre-strained layer is in the stack through lamination. The pre-strain can be in the form, for example, of tension or compression, though compression may require consideration of tendencies of linkages to buckle out of plane.

In other embodiments, actuators can be built into the laminate to effect assembly. For example, a piezoelectric bending actuator, shape memory layer, or other type of actuator can be laminated into the structure as a pick-and-place component or inserted as an integral part of a layer in the layup; and the actuator can be actuated, e.g., by supplying electrical current or by changing temperature, to assemble the expanded, three-dimensional structure.

Advantageously, in some embodiments, the assembly of all parts is actuated via a single assembly degree of freedom so that assembly proceeds in parallel for an entire panel, rather than part by part. Assembly can be effected in several ways, depending on the design and complexity of the part. For example, a human operator can actuate the assembly degree of freedom manually or semi-automatically. In one embodiment, the assembly degree of freedom is in the form of a plate connected to a Sarrus linkage that is pulled up or pushed down. Spherical joints or four-bar mechanisms can be attached to the Sarrus linkage, raising and folding other components into their three-dimensional position. Note that by having multiple rigid-flex planar layers and selective adhesion, complex mechanisms and collections of mechanisms can be released in the assembly step.

8) Joint Locking of Assembled Part

After assembly into a final three-dimensional structure, structural members can be bonded together in a fixed configuration {i.e., locked, fixed or frozen). In one embodiment, adhesive can be manually applied to structural members and/ or joints, though this approach may not be ideal if many parts are being made.

Alternatively, adjacent members that have come together to form a locked joint can be automatically laser welded. If adjacent members 45 and 46 have metal pads 47 {e.g., formed of brass) on them, then wave or dip soldering can form strong filleted bonds 48 between the members, as shown in FIGS. 26-28. Alternatively, solder paste can be applied, for example, by screen printing before assembly to the laminate; and then, after assembly, a re-flow step in a hot oven creates the bonds. Other variations include the use of two-part adhesives, etc.

In one embodiment, the pop-up laminate device includes brass pads 47 distributed across outer surfaces of its linkage sub-laminates, as shown in FIG. 26. After folding, pads on disparate links align into "bond points," in the form of either two pads 47 meeting at right angles, as shown in FIGS. 27 and 28, or three pads forming the corner of a cube. The structure, held in its folded state, is submerged in a water-soluble flux {e.g., Superior Supersafe No. 30) and then pre-heated in an oven at 100°C for 10 minutes. It is then submerged in 260°C tin-lead eutectic solder for approximately 1 second. Finally, the structure is ultrasonically cleaned in distilled de-ionized water to remove the water-soluble flux residue. The result of this soldering process is the formation of solder fillets 48 at all bond points, as shown in FIG. 28, eliminating the assembly degree of freedom and locking all disparate machine components together.

9) Freeing the Non- Assembly Degrees of Freedom

Any non-assembly degrees of freedom in the part can be unlocked by removing any features {e.g., connected tabs) that restrain them via, e.g., laser machining. 10) Separating Parts from the Scrap Frame

Now that the individual parts are fully assembled and ready for operation, the parts can be separated from the scrap frame {e.g., an outer frame to which the parts are connected by bridges of material) of the scaffold 14 by laser machining, punching, etc. Layer sharing:

In various embodiments, two sub-layers can be thought of as sharing the same layer because they are non-overlapping and both engage with the same adhesive layer, e.g., glue, to bond with another layer.

Two ways to accomplish layer sharing are described, as follows. In the first, multiple layers occupy non-overlapping areas in the

For example, four alignment pins 50 can be used. In one embodiment, a brass layer can cover half of the full area of the device, while a titanium layer can cover the other half. The brass can be used to form solder pads 47, while the titanium can be used to form structural components. Each sub-layer can engage with just two out of the four alignment pins 50 [i.e., two pins can engage with the brass sub-layer, while the two other pins can engage with the titanium sub-layer]. Taken to the extreme, the layer can be split into many sub-layers if each sub-layer is engaged with enough alignment pins 50. For example, a single layer with six sub-layers can look like a map of New England, with each state made out of a different material, and with two alignment pins per state.

A second way of achieving layer sharing is by applying pressure to layers that are unsupported from below to bend the layers into the space below. Basically, if a large hole is cut in a thin layer, the application of pressure to the layer immediately above it (or below it) during lamination can be designed to warp and bend that adjacent layer around the edge of the hole, filling in the hole.

In describing embodiments of the invention, specific terminology is used for the sake of clarity. For the purpose of description, specific terms are intended to at least include technical and functional equivalents that operate in a similar manner to accomplish a similar result. Additionally, in some instances where a particular embodiment of the invention includes a plurality of system elements or method steps, those elements or steps may be replaced with a single element or step; likewise, a single element or step may be replaced with a plurality of elements or steps that serve the same purpose. Further, where parameters for various properties or other values are specified herein for embodiments of the invention, those parameters or values can be adjusted up or down by l/100^th, l/50^th, l/20^th, l/10^th, l/5^th, l/3^rd, 1/2, 2/3^rd, 3/4^th, 4/5^th, 9/10^th, 19/20^th, 49/50^th, 99/100^th, etc. (or up by a factor of 1, 2, 3, 4, 5, 6, 8, 10, 20, 50, 100, etc.), or by rounded-off approximations thereof, unless otherwise specified. Moreover, while this invention has been shown and described with references to particular embodiments thereof, those skilled in the art will understand that various substitutions and alterations in form and details may be made therein without departing from the scope of the invention. Further still, other aspects, functions and advantages are also within the scope of the invention; and all embodiments of the invention need not necessarily achieve all of the advantages or possess all of the characteristics described above. Additionally, steps, elements and features discussed herein in connection with one embodiment can likewise be used in conjunction with other embodiments. The contents of references, including reference texts, journal articles, patents, patent applications, etc., cited throughout the text are hereby incorporated by reference in their entirety; and appropriate components, steps, and characterizations from these references may or may not be included in embodiments of this invention. Still further, the components and steps identified in the Background section are integral to this disclosure and can be used in conjunction with or substituted for components and steps described elsewhere in the disclosure within the scope of the invention. In method claims, where stages are recited in a particular order— with or without sequenced prefacing characters added for ease of reference— the stages are not to be interpreted as being temporally limited to the order in which they are recited unless otherwise specified or implied by the terms and phrasing.

Claims

CLAIMS What is claimed

1. A monolithic, multi-axis force sensor, comprising a laminated structure

including:

a scaffold;

a plurality of arms extending within and across the scaffold at distinct angles, wherein the arms include:

a) a structural support layer;

b) a sensor layer including a strain-gauge alloy; and

c) a flexible and electrically insulating polymer layer sandwiched between the support layer and the sensor layer in a monolithic, multi-layered laminate structure; and

electrically conductive pathways positioned to deliver a voltage through the strain-gauge alloy in the arms.

2. The monolithic, multi-axis force sensor of claim 1, wherein the arms are

joined at a central mount that defines an aperture.

3. The monolithic, multi-axis force sensor of claim 1, further comprising

adhesive layers between the support layer and the polymer layer and between the polymer layer and the sensor layer.

4. The monolithic, multi-axis force sensor of claim 1, wherein the scaffold

includes castellated hinges around the arms, wherein the castellated hinges are configured to allow edges of the scaffold to be folded in a position orthogonal to the remainder of the scaffold.

5. The monolithic, multi-axis force sensor of claim 1, further comprising a

compliant insulating coating on an exterior surface of the scaffold and the arms.

6. The monolithic, multi-axis force sensor of claim 5, wherein the compliant insulating coating comprises parylene.

7. The monolithic, multi-axis force sensor of claim 1, wherein the strain-gauge alloy comprises constantan 40.

8. The monolithic, multi-axis force sensor of claim 1, wherein the polymer layer comprises a polyimide.

9. The monolithic, multi-axis force sensor of claim 1, wherein at least one of the electrically conductive pathways comprises copper.

10. The monolithic, multi-axis force sensor of claim 1, wherein a plurality of the layers define mounting holes via which the layers can be precisely aligned.

11. A method for fabricating a multi-axis force sensor, comprising:

aligning a polymer layer comprising a flexible and electrically insulating polymer on a substrate layer;

aligning a sensor layer comprising a strain-gauge alloy on the polymer layer; and

joining the substrate layer, the polymer layer and the sensor layer to form a monolithic multi-layered laminate structure, wherein the substrate layer forms a scaffold, and wherein the sensor layer is included in a plurality of arms extending within and across the scaffold and is configured to produce a change in electrical resistance through the sensor layer as the arms are displaced with strain.

12. The method of claim 11, wherein the substrate layer comprises a plurality of rigid layers separated by polymer layers.

13. The method of claim 11, further comprising aligning an electrically

conductive pathway on the polymer layer in electrical contact with the strain- gauge alloy.

14. The method of claim 11, wherein the layers are joined by adhesive layers positioned between the substrate layer and the polymer layer and between the polymer layer and the sensor layer.

15. The method of claim 11, further comprising folding edges of the scaffold to form a walled enclosure encircling the arms.

16. The method of claim 11, further comprising folding the monolithic multi- layered laminate structure to produce surfaces with sensors oriented at substantially orthogonal angles with respect to one another.

17. A method for multi-axis force sensing, comprising:

contacting an object with a remote-controlled tool coupled with a monolithic force sensor comprising: a scaffold; and a plurality of arms extending within and across the scaffold at distinct angles, wherein the arms include: (a) a structural support layer; (b) a sensor layer including a strain- gauge alloy; and (c) a flexible and electrically insulating polymer layer sandwiched between the support layer and the sensor layer in a monolithic, multi-layered laminate structure;

passing an electrical current through the strain-gauge alloy as the tool contacts the object;

tracking the resistivity of the strain-gauge alloy as the tool contacts the object;

determining force vectors applied to the tool as it contacts the object as a function of changes in the resistivity of the strain-gauge alloy; and

communicating the force vectors to an operator operating the tool via remote control.

18. The method of claim 17, wherein the force vectors are communicated to the operator by applying a force vector feedback to a control device used by the operator to control the tool, wherein the force vector feedback is a

reproduction of the force vector applied to the tool.

9. The method of claim 17, wherein the object is an internal body part of a human, and wherein the tool is a surgical tool.

The method of claim 17, wherein force vectors along three orthogonal axes are determined and communicated.

The method of claim 20, further comprising determining and communicatin: torque along the three orthogonal axes.