WO2017201028A1

WO2017201028A1 - Robotic surgical module

Info

Publication number: WO2017201028A1
Application number: PCT/US2017/032871
Authority: WO
Inventors: Joshua GAFFORD; Robert Wood; Conor Walsh
Original assignee: President And Fellows Of Harvard College
Priority date: 2016-05-16
Filing date: 2017-05-16
Publication date: 2017-11-23

Abstract

A robotic surgical module can include a distal structure, comprising a robotic backbone configured for mounting on a distal end of an endoscope (e.g., a standard, commercially available endoscope) and an end effector extending from the robotic backbone. The end effector includes (a) an expanded structural framework; (b) a surgical tool extending through the expanded structural framework; and (c) an actuator mounted to the structural framework and configured to flex the expanded laminate structure and the surgical tool when actuated. The structural framework can be in the form of a laminate structure.

Description

ROBOTIC SURGICAL MODULE

GOVERNMENT SUPPORT

This invention was made with government support under Grant No. FA8650- 15-C-7548 awarded by the Defense Advanced Research Projects Agency. The U.S. Government has certain rights in the invention.

BACKGROUND

Burgeoning transendoscopic procedures, such as endoscopic submucosal dissection (ESD), provide promising means of treating early-stage gastric neoplasia in a minimally invasive way. The remote locations of these lesions, however, coupled with their origination in the submucosal layers of the gastrointestinal tract, often lead to extreme technical, cognitive and ergonomic challenges that compromise the widespread applicability and adoption of these techniques. Among these challenges is achieving the in-vivo dexterity required to retract and dissect tissue.

Innovation in robotic surgery is seeing a paradigm shift from rigid,

teleoperative serial manipulators to flexible co-robotic tools capable of accessing remote locations inside the body. This paradigm shift is concurrent with a desire to perform "scarless" surgery through natural orifices to substantially reduce morbidity and recovery. While a number of systems have demonstrated unparalleled dexterity in navigating the body's tortuous anatomy, limited sophistication in end-effector design has severely impeded the therapeutic applications of these systems. To usher in the next generation of therapeutic endoscopic tools, it is of paramount

importance to develop sophisticated end-effector morphologies that can extend the capabilities of these systems.

General trends in surgical endoscopy have pointed towards minimally invasive applications in both diagnostic and therapeutic interventions that were previously impossible or infeasible to perform. A burgeoning transendoscopic technique called endoscopic submucosal dissection (ESD) is being seen as a promising method of removing early-stage gastric neoplasia in a minimally invasive way. The technique involves the insertion of an endoscope or gastroscope through the mouth, navigating to the site of the tumor, and using a combination of forward- cutting and side-cutting bipolar cautery to resect and remove the diseased tissue from the sub-mucosal space. Due to the technical and cognitive complexity involved, coupled with a reliance on surgical devices that are ill-suited for the task at hand, ESD has seen only limited penetration into clinical practice. In addition, the unintuitive mapping between the endoscope controls and the distal dexterity required at the tip make ESD suitable for robotic innovation.

Several groups have explored robotic solutions to the technical challenges imposed by ESD and endoscopic procedures in general. Such implementations typically feature stand-alone robotic systems or roboticized "add-ons" (overtubes or endcaps) to existing endoscopic equipment, where proximal actuator packages control distal end-effectors passed through the endoscope's working ports or around the outside of the endoscope. While these systems demonstrate endoscopic

applications of robotics, proximal actuation schemes can limit modularity, clutter the surgical arena, and preclude the systems from single-use practice and from disposable price points due to the intimate mechanical coupling between expensive proximal actuation systems and distal mechanisms.

SUMMARY

A robotic surgical module and methods for its use are described herein, where various embodiments of the apparatus and methods may include some or all of the elements, features and steps described below.

A robotic surgical module can include a distal structure, comprising a robotic backbone configured for mounting on a distal end of an endoscope {e.g., a standard, commercially available endoscope) and an end effector extending from the robotic backbone. The end effector includes (a) an expanded structural framework {e.g., in the form of a laminate structure); (b) a surgical tool extending through the expanded structural framework; and (c) an actuator mounted to the structural framework and configured to flex the expanded structural framework and the surgical tool when actuated.

The actuator can be a fluid-based actuator or a shape memory alloy and can be entirely on or in the end effector. A control system can also be include in or on the end effector, and a proprioceptive sensor can be included on or in the end effector and in

communication with the control system to provide a distal feedback loop.

The robotic surgical module can be configured for replaceable mounting on and detachment from {e.g., snapping on and off) the endoscope. The surgical tool can be a cautery device. The end effector can include a plurality of spaced-apart rigid plates and an elastic flexure passing through at least one of the rigid plates. A light emitting diode can be mounted to a first member of the rigid plates at a first end of the end effector, and a phototransister can be mounted to a second member of the rigid plates at a second end of the end effector. The light emitting diode and phototransistor can be configured such that the amount of light from the light emitting diode incident upon the phototransistor changes as the end effector flexes.

Herein, a fully distal snap-on robotic module for endoscopic procedures is presented. System design was informed by clinical parameters, which were found experimentally. Integrated sensing and actuation provides fully modular

capabilities, thereby obviating the need for proximal actuator packages and transmission mechanisms. The integrated system demonstrated the ability to provide 51 ± 4:5 degrees of angular dexterity and to generate lateral forces of around 450 mN, which is sufficient for cautery.

By leveraging workspace and force data obtained through clinical studies, a modular, disposable, distally mounted robotic surgical module (an "active endcap"), described herein, can augment an endoscopist's distal dexterity in ways that are not achievable using the endoscope's built-in degrees-of-freedom. The robotic surgical module includes a flexible articulating 'exoskeleton' manufactured via printed- circuit MEMS (PCMEMS) laminate structures that engage and deflect electrosurgical tools that are passed through the endoscopic working channel. Embedded

proprioceptive sensing can be implemented on-board using distributed

LED/phototransistor pairs and the principle of light intensity modulation (LIM). The distal degree-of-freedom can be actuated using shape memory alloy (SMA) technology, and the actuation transmission system can be fully contained within a 1- inch-long robotic backbone (end cap) that can be mounted on the distal end of the endoscope, thereby obviating the need for a mechanical connection to a proximal source. Proof-of-concept tests demonstrate that the actuator adds over 50 degrees of distal articulation to existing tools and can generate 450 mN of lateral force, which has been clinically determined to be sufficient for performing circumferential incisions in ESD.

The robotic surgical module can also include a soft actuator structure {e.g., mounted to the side of the robotic backbone), as described in US provisional application No. 62/336,874, to manipulate tissue.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is an illustration of a distally actuated one-degree-of-freedom (1-DoF) modular wrist, with a magnified inset of the end features. An articulating module interfaces with and deflects existing electrosurgical tools passed through the endoscope working port.

FIG. 2 plots the results of meta-analyses showing dexterity achieved by an endoscope during each subtask (injection 32, axial incision 34, and lateral incision 36) acting on a tumor .

FIG. 3 plots interaction forces between a bipolar cautery system and various layers within a porcine stomach. Circular markers denote cautery pulse events.

FIG. 4 is a conceptual illustration of a flexural backbone operation with integrated optical angle sensing. Passive flexure-based rotary joints (illustrated with circles) facilitate assembly and prevent axial torsion.

FIG. 5 includes (a) a parametric map of angular dexterity 20o as a function of input force, F, and flexure thickness, t, where the red plane shows the design goal, and where the black plane shows the SMA stroke limit and (b) a parametric map of the mechanical factor-of-safety (FOS) as a function of Fand t, where the red plane shows FOS = 1.

FIG. 6 provides a system overview, showing (left) a detail of two serial articulating modules fabricated via PCMEMS, (middle) integrated system with a US penny (with a 19.05 mm diameter) for scale, and (right) exploded detail of the SMA transmission mechanism. FIG. 7 is an exploded view of a 15-layer laminate for forming the expandable actuator structure of the wrist.

FIG. 8 shows laser-machining to release sarrus linkages of the assembly of FIG.

7.

FIG. 9 shows guided assembly of the articulating exoskeleton of the assembly of FIG. 7.

FIG. 10 shows pick-and-place of flexures, phototransistors and LEDs, wherein the assembly scaffold is omitted for clarity.

FIG. 11 shows a thumbscrew-actuated staging and assembly jig for the assembly scaffold, where (left) shows the flat PCMEMS structure and (right) shows the "popped-up" structure.

FIG. 12 plots LIM-based angle sensor calibration results compared to a predictive model.

FIGS. 13 and 14 include images showing how redundant sensors can be used for differential measurements, eliminating common-mode noise.

FIGS. 15-17 plots the results of SMA characterization. FIG. 15 plots a

contraction profile as a function of time for various pre-tension biases. FIG. 16 plots a blocking force profile as a function of time, where the shaded area shows standard deviation for n = 5 runs. FIG. 17 plots blocking force evolution as a function of input current.

FIG. 18 shows a system actuated over a positive angle (inset image) and the resulting on-board sensor measurements (pivot angle of the end effector as a function of time). The inset shows the unactuated system and the angular range of actuation.

FIG. 19 plots a lateral force profile for a step input in current as a function of time.

FIGS. 20-22 relate to a helical SMA design tool for hypothetical analysis, including (clockwise from top left) a plot of valid solution space given design constraints (where the marker denotes the selected configuration) in FIG. 20, a plot of force margin over design stroke in FIG. 21, and a plot of thermodynamic behavior in FIG. 22. 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, an 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 herein defined, used or characterized, 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 be in terms of weight or 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) unless otherwise specified. 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.

I. Introduction:

Described herein is a robotic surgical module 10, including a robotic backbone 14 and end effector 16, that interfaces with commercial endoscopes 12 to provide distal dexterity to existing cautery devices and other surgical tools passed through the endoscopic working channel, as shown in FIG. 1. This embodiment uses a shape memory alloy (SMA) actuator 22 that lengthens or shortens with a change in temperature to displace the structural framework (here, in the form of a laminate structure 18), which includes elastic flexures 24 and to which it is secured. The SMA can be, e.g., a copper-aluminum-nickel or nickel-titanium alloy. The end effector also includes a LIM-based angle sensor 26 mounted on the laminate structure 18. The robotic backbone 14 includes a bearing-base SMA transmission 28 and an SMA tensioner 30.

A composite manufacturing process (PCMEMS), as described in, e.g., US

Patent No. 8,834,666 B2 and WO 2015/020952 Al, can be used to monolithically fabricate the complex articulating end effector structure, greatly reducing assembly requirements.

Using the PCMEMS fabrication process, the three-dimensional structure can be formed by stacking a plurality of patterned layers and bonding the plurality of patterned layers {i.e., layers having a patterned shape/features formed, e.g., by machining) at selected locations to form a laminate structure with inter-layer bonds. The laminate structure can then be expanded into an expanded three-dimensional configuration by selectively distorting at least one of the layers to produce gaps between layers while maintaining at least some of the inter-layer bonds.

The layers in the structure can include at least one rigid layer and at least one flexible layer; the rigid layer includes a plurality of rigid segments, and the flexible layer can extend between the rigid segments to serve as a joint. The flexible layers are substantially less rigid than the rigid layers; e.g., the rigid layers can have a rigidity that is at least an order of magnitude greater than {i.e., greater than lOx or greater than lOOx) the rigidity of the flexible layers; likewise, the flexible layer can have at least 10 times or at least 100 times the flexibility of the rigid layers. The layers can then be stacked and bonded at selected locations to form a laminate structure with inter-layer bonds, and the laminate structure can be distorted or flexed to produce an expanded three-dimensional structure, wherein the layers are joined at the selected bonding locations and separated at other locations.

More specifically, the multi-layer super-planar structure of the end effector can be fabricated via the following sequence of steps: (1) machine each planar layer {e.g., with a laser), (2) machine or pattern adhesives, (3) stack {e.g., using dowel pins and providing the layers with alignment holes) and laminate the layers under conditions to effect bonding, (4) post-lamination machining of the multi-layer structure, (5) post-lamination treatment {e.g., plating or coating exposed layers and/ or further addition of components via a pick-and-place methodology) of the multi-layer structure, (6) freeing an assembly degree of freedom in each structure {e.g., by severing any restraining structural bridges that join layers/components) to release/ expand the laminate structure so as to separate the layers {e.g. via actuation of an actuator, such as a piezoelectric cantilever actuated via application of a voltage or shape memory alloy actuate via a change in temperature, or via release of a loaded spring), wherein the spatially separated layers remain joined by segments selectively spanning the layers [i.e., extending from or bonded to {e.g., via selectively located islands of adhesive) more than one layer], (7) locking connections between structural members, (8) freeing any non-assembly degrees of freedom, and (9) separating finished parts from a scrap frame.

In other embodiments, the end effector can be fabricated from discrete layers joined via conventional fabrication techniques.

Embedded orientation sensing can be implemented on-board using

distributed emitter/detector pairs and the principle of light-intensity modulation. The actuation transmission system leverages shape memory alloy (SMA) technology and is fully contained within a 1 -inch-long end cap that can be mounted on the distal end of the endoscope, making the system fully deployable with no proximal component. The low cost of raw materials coupled with batch manufacturing processes results in a potentially disposable system. Section II presents the results of empirical clinical studies implemented to obtain workspace/ range-of-motion and force data to characterize the procedure and to generate task-specific system functional requirements. These clinical parameters were used to inform the creation of functional requirements for the proposed system, and Section III discusses system design, modeling and optimization based on these requirements. Section IV presents subsystem-level manufacturing processes, as well as integration and validation of the system at a proof-of-concept level.

II. Clinical Parameterization:

When designing robotic systems for task-specific applications, it is important to understand (1) the workspace that the system is required to cover, and (2) the forces that will be encountered and that must be generated by the system. Given the relative sparsity of ESD-specific clinical data in literature, an experimental approach was employed to define these clinical requirements to inform the data-driven design process of a specialized robotic module tailored specifically for ESD. A. Workspace Analysis

An electromagnetic (EM) motion tracking system was used to capture the position and orientation of an endoscope and a simulated tumor during three complete ESDs. Procedures were performed ex vivo on a porcine stomach in a clinical laboratory setting. One EM probe was fastened to the endoscope tip, and another was clipped into the location of the simulated tumor, enabling 6-degree-of- freedom motion data to be captured simultaneously from the tumor and from the endoscope tip at a rate of 1 kHz. Motion and orientation data were parsed based on the subtask that they represent to characterize the following three primary phases of ESD: (1) liquid injection for tumor liftoff, (2) axial (forward-cutting) incision for hole creation, and (3) lateral (side-cutting) incision for tumor resection. Resulting data were filtered using a combined Sauvitsky-Golay/Median filter to remove noise induced by electrosurgical pulses.

A lateral incision process is simulated. Lateral incision utilizes the largest workspace as it is the most difficult maneuver to perform in ESD, often requiring significant dexterity to create and "sweep out" the submucosal space beneath the tumor. Meta-analyses of each subtask are shown in FIG. 2, where, for each subtask (injection 32, axial incision 24, and lateral incision 36), n = 8, 3, and 5, respectively. We see that the average angular dexterity required for lateral incision, in this instance, is around 90.3 ± 50.4 degrees. The large error in the lateral incision subtask is due to a retroflexion event wherein the endoscopist had to deflect the scope nearly 180 degrees to access the distal section of the tumor. If this event is excluded, the average angular dexterity required for the lateral incision is 66:4 ± 20:3 degrees. Therefore, as a first step towards capability augmentation, embodiments of the robotic module 10 were designed to provide 60 additional degrees of angular deflection with this system.

B. Force Analysis

In developing a task-specific robotic system, it is also beneficial to have a thorough understanding of the types and magnitudes of forces that will be encountered. To determine these forces experimentally, a system was built that enabled the measurement of interaction forces between commercial electrosurgical equipment and pre-tensioned tissue specimens during a resection event. A bipolar cautery device (an OLYMPUS DUALKNIFE system from Olympus Corporation of Tokyo, Japan) was clamped onto a load cell (model LCL-005 from Omega

Engineering of Stamford, Connecticut, USA) that was attached to a leadscrew-based linear stage with sub-μηι position encoding. The tissue specimen was clamped and pre-tensioned distal to the surgical tool. The tool was advanced into the specimen and pulsed simultaneously, and data (force and displacement) was captured at a rate of 500Hz until the tool has fully penetrated through the tissue specimen.

The results of tests performed on three different porcine stomach tissue morphologies {i.e., full thickness 38, mucosal layer 40, and muscularis layer 42) are shown in FIG. 3 with circles indicated for each pulse. We see, in each case, that a maximum axial force of around 300-400 mN is required to fully perforate the tissue. Therefore, any system we design for this purpose is targeted to generate at least 400 mN of force for electrocautery. III. System Design

Given the results of the clinical parameterization study, a robotic system was designed to satisfy the aforementioned requirements. Illustrations of the articulating end effector 16 are shown in FIG. 4. Similar to many other tendon-driven continuum systems, various disk-shaped spacers 44 are equally separated axially along an elastic flexure mechanism 24. Tendon actuators 22 pass through intermediate spacers 44, which are joined by guided assembly flexures 50 with constraint hinges 52, and terminate on the distal-most spacer 44, such that when a tensile force is applied to a tendon actuator 22, a moment is created about the elastic flexure mechanism 24, thereby causing the structure 18 to bend.

The system proposed herein differs in a few ways from previous tendon-based continuum systems. The present robotic surgical module 10 can be mounted on an endoscope 12 so the module 10 is provided with an unobstructed bore; and, as a result, the elastic flexure mechanism 24 is not placed in the geometric center of the spacer disks 44, but rather along the outside. Minimizing the system's imposition into the endoscope's vision system introduces a trade-off between footprint minimization and system robustness. In addition, integrated infrared (IR) light emitting diodes (LEDs) 46 and IR phototransistors (PTs) 48, which together form the LIM-based angle sensor 26, can be mounted on subsequent spacer disks 24, enabling distributed angle sensing by light intensity modulation (LIM), as described in J. B. Gafford, S. Member, R. J. Wood, and C.J. Walsh, "Self-Assembling, Low-Cost, and Modular mm-Scale Force Sensor," IEEE Sensors Journal, vol. 16, no. 1, pp. 69-76, 2016.

In order to develop a fully modular, compact, distally mounted robotic surgical module 10 with no proximal component, shape memory alloy (SMA) was pursued as a method of actuation in particular embodiments due to its high energy density, compactness, and relatively simple (on/ off) control. Limitations of SMA actuation were heavily considered and deemed to be insubstantial for this application. For example, as the lateral incision process is relatively slow, actuator bandwidth is not a major concern. The use of very thin SMA wire results in very localized thermal dissipation that can be easily insulated. Embedded orientation sensors that can be used for closed-loop feedback control are described in N. Kha and K. Ahn, "Position Control of Shape Memory Alloy Actuators by Using Self Tuning Fuzzy PID Controller," IEEE Conference on Industrial Electronics and Applications, pp. 1-5, 2006.

The following section presents geometric parameter selection (via brute force optimization methods) as well as a general model (nonlinear and linear) for LIM- based angle sensing for tendon-driven continuum or flexure-based robotic structures.

A. Geometric Optimization of Articulating Module

A brute force optimization approach is employed to determine the flexure stiffness parameter, t, and input force, F, required to (a) generate the required dexterity (deflection angle, φο, according to FIG. 4) and (b) ensure that the flexure remains within its elastic limit over the entire stroke. Other parameters, such as radius, r, and flexure width, w, axe constrained by the functional requirements of the system (i.e., the system is designed to not add more than 2 mm to the outer diameter of the endoscope). To determine 4>o (F; t), we use a large-deflection model [16] for thin-beam bending given a pure moment and a fixed end condition:

where / is the length of the flexure in a single articulating module; iVis the number of articulating modules stacked in series; Eis the Young's Modulus of the flexure material; and I is the second moment of area. In the above Equation, the integral was solved numerically to compute φο (F, t) for some combination of Fand t. Due to symmetry and the use of antagonistic actuation, ±φο can be generated for an overall angular deflection of 2 φο to satisfy the functional requirement.

Parametric surfaces resulting from this optimization are shown in FIG. 5 for N = 2. The plotted contours at left show the (log) angular dexterity of the end effector as a function of input force, F, and flexure thickness, t. The lower plane 54 shows the design requirement, and the upper plane 56 shows the SMA upper-limit (as the actuator is inherently stroke-limited). The design goal is to find a combination that exists on the parametric landscape above the red plane but acknowledging that stroke performance is limited by the black plane. A flexure thickness of 75 μηι and an actuation force of 1 N or above satisfies this requirement. We also want to ensure that the elastic flexure remains within its elastic limits by satisfying the following equation for mechanical factor-of-safety (FOS > 1):

\ΝΙσ

FOS=-

Frt

where o is the yield stress. The plotted contour at the right of FIG. 5 shows the (log) FOS, where the red plane shows an FOS of 1. The chosen combination results in FOS > 1, so both conditions are satisfied.

B. Sensor Modeling

Previous work (J. B. Gafford, S. Member, R. J. Wood, and C.J. Walsh, "Self-

Assembling, Low-Cost, and Modular mm-Scale Force Sensor," IEEE Sensors Journal, vol. 16, no. 1, pp. 69-76, 2016) demonstrated the use of discrete, distally mounted LEDs and phototransistors to sense deflection (or force) in a structure. This work showed that the irradiance striking the phototransistor (PT) from a point-source LED model falls off with the square of the distance separating the LED from the PT. In this implementation, as shown in FIG. 4, misalignment between the two axes as the backbone deforms introduces a cosine term, resulting in the following equation for irradiance Ερχ striking the phototransistor:

where ILED is the radiant intensity from the LED and h = Nl is the spacing between the LED and PT. From this, we can calculate the phototransistor collector current, ipT, as a function of the irradiance:

1 PT ^~ exp (a log (E _r ) + ,#) , where or and β are fit constants.

This function can be linearized about φ = 0 to obtain a simple calibration model relating phototransistor current, ipr, to angular deflection, φ , as follows: φ₀ , where

Numerical analysis shows that there exists less than 6% error between the full model and the linear simplification over the expected range-of-motion.

IV. Manufacturing and Validation

An overview of the fabricated system and its subsystems is shown in FIG. 6. The end effector 16, shown at left, includes the elastic flexure 24, the LIM-based angle sensor 26, assembly holes 58, circuit traces 60, an instrument/ tool through- hole 62, guided assembly flexures 50 , and a tendon-actuator pass-through 64 . The robotic backbone 14, shown in an exploded view at right, includes an endoscope mount 66, flanged bearings 68, polytetrafluoroethylene (PTFE) tubes 70, a polyimide- film -wrapped steel tube 72, tension guide dowels 74, and tensioner blocks 76 and tension screws 78, which together form the actuator tensioner 30.

The following section discusses the manufacturing processes used to fabricate each subsystem and presents preliminary results from both a subsystem and integrated system perspective.

A. Articulating End Effector

The articulating exoskeleton end effector 16 was fabricated using the PC- MEMS technique described in (a) P. S. Sreetharan, J. P. Whitney, M. D. Strauss, and R. J. Wood, "Monolithic fabrication of millimeter-scale machines, "Journal of

Micromechanics and Microengineering, vol. 22, no. 5, p. 055027, 2012, and in (b) J. P. Whitney, P. S. Sreetharan, K. Y. Ma, and R. J. Wood, "Popup book MEMS," Journal of Micromechanics and Microengineering, vol. 21, no. 11, p. 115021, 2011.

In one embodiment, the end effector layup consists of 15 layers of material, including four layers of 75^m-thick 304 stainless steel for the structural layers, two layers of 2^m-thick KAPTON polyimide for the flexure-based assembly layers, seven layers of DuPont FR0100 acrylic sheet adhesive, and two layers of 25 μηι KAPTON polyimide with an 18^m-thick copper cladding layer. Each layer is machined individually using a diode-pumped solid-state (DPSS) laser, and all 15 layers are laminated together in a heat press.

One embodiment of the layup of layers for the end effector 16 is illustrated in FIG. 7, which shows the following layers: 304 stainless steel 80, KAPTON polyimide 82 (from DuPont) coated with traces of copper 84, and adhesive 86 arranged into a top sub-laminate 88 and bottom sub-laminate 90.

Preliminary release cuts are made with a laser 92, as shown in FIG. 8, to free the guided assembly Sarrus linkages, as shown in FIG. 9. The elastic flexure 24 and LED's 46 can then be applied, as shown in FIG. 10. A plurality of the laminate structures 18 (mass produced in large scaffold sheets 94) is then placed in a pin- actuated alignment jig 92, as shown in FIG. 11 (with a US quarter shown for scale), where thumbscrew-driven push pins provide the actuation required for self-folding and assembly. In addition to guiding the assembly, these Sarrus linkages also provide an integrated "mechanical limit" that prevents over-deflection of the structure. Two 75^m-thick tempered spring steel flexures are placed into slots in the structure and epoxied into place, and each laminate 18 is once again machined with the DPSS laser to release the laminate structure 18 from the surrounding hexagonal alignment scaffold 94. LEDs and phototransistors are press-fitted into their respective windows in the structure 19 and reflow soldered to complete assembly.

B. SMA Transmission Module

In order to generate the targeted stroke (3 mm) from the SMA actuator, a 150- mm long actuator wire is required (assuming 4% contraction and accounting for the "doubling back" of the SMA which halves the stroke but doubles the force output). To keep the system compact, a transmission module was designed featuring low- friction sliding surfaces and flanged ball bearings, allowing the wire to be "wound" back and forth several times, thereby significantly driving down the overall footprint of the entire system. An image of this transmission module, including the tension guide dowels 74, tensioner blocks 76 and tension screws 78, in the robotic backbone 14 is shown in FIG. 6 (right). The SMA tensioner 30 is fed into a transmission module from the SMA anchors on the articulating exoskeleton, wound through a series of bearing surfaces 68, and terminated at tensioning blocks 76 that are leadscrew- driven, enabling independent adjustable pre-tensioning of each actuator 22. C Signal Conditioning and Sensor Calibration

Due to the wide spectral bandwidth of the phototransistor's response (400- 1100 nm), special considerations are made in developing signal conditioning infrastructures such that DC-level ambient signals are not amplified. An AC-coupled excitation circuit based around an astable oscillator for the emitter side was implemented, pulsing the IR LEDs 46 at a rate of 8 kHz. The collector-side is filtered and amplified with an active band-pass filter centered at 8 kHz. This scheme is described in more detail in J. B. Gafford, S. Member, R. J. Wood, and C.J. Walsh, "Self-Assembling, Low-Cost, and Modular mm-Scale Force Sensor," IEEE Sensors Journal, vol. 16, no. 1, pp. 69-76, 2016.

Such an implementation renders the sensing system impervious to ambient conditions. It is possible that cross-talk between sensors can affect individual sensor readings as there is no optical isolation between sensor pairs (aside from an axial offset). A robust solution is to pulse the LEDs 46 in each sensor pair at a different frequency and band-passing the respective output of the paired phototransistor 48 centered to the LED frequency, although this would lead to more wires as each LED 46 would have its own excitation lead. For the purpose of this paper, cross-talk is assumed to be suitably accounted for in calibration.

The angle sensor was calibrated statically by hanging weights from the SMA anchors, measuring the angle and recording the sensor outputs. The calibration curve is shown in FIG. 12, where LIM-based angle sensor calibration results compared to a predictive model. Pairs of sensors (SI and S2, represented respectively by circles and squares) oppose each-other, offering the opportunity to perform differential measurements (shown with triangles) to eliminate common-mode sources of noise that were not adequately filtered in signal conditioning (such as local heating due to SMA). We can see that the actual sensor implementation behaves very closely to that predicted by the model, which is plotted as the dashed line. The images of FIGS. 13 and 14 show how redundant sensors can be used for differential measurements, eliminating common-mode noise.

D. SMA Characterization

To better understand the force and stroke capabilities of the SMA wire actuator 22 under ideal conditions, a characterization platform was developed. The platform includes a stationary mount where the SMA transmission module can be dropped in. The SMA is routed through a moving platform which glides smoothly on lubricated steel shafts to constrain motion to the actuation axis. A low-friction slide potentiometer tracks the location of the moving platform. The moving platform is coupled to a stationary load cell (LCL-005, Omega Engineering) via extension springs with a pretension that can be modified by adjusting the distance between the load cell mount and transmission mount.

Using the final transmission design, 0.006-inch (152 μηι) diameter FLEXINOL actuator wire (Dynalloy, Irvine, CA) was tested for its contraction properties under varying force biases. To activate the SMA, 360 mA of current was applied in the form of a 1/10 Hz square wave with 60% duty cycle, resulting in 6 seconds of heat-up time and 4 seconds of cool-down time. Data were collected using a USB-6002 data acquisition (DAQ) device (from National Instruments) with a 1-kHz sample rate. The results are shown in FIG. 15, where current is plotted via a dashed line 94, and contraction is shown for 1 N bias 96, 1.5 N bias 98, and 2 N bias 100. Diminishing returns are observed above a 1.5-N bias force 98; 1 N of bias 96, however, is insufficient to properly pre-tension the wire. Some of the stroke is lost to slack in the system, thereby compromising performance and contraction efficiency. As a result, 1.5 N of pre-tension is sufficient for high-efficiency operation. We also observe that the SMA contracts over 3 mm with adequate pre-tension.

The blocking force capabilities of the SMA (that is, the reaction force that fully prevents the SMA from contracting) are characterized to understand the force output capabilities of the system. By coupling the SMA directly to the load cell {i.e., bypassing the bias springs such that there is no contraction), pre-tensioning to 1.5 N, and providing the same current profile 94 as before, a force profile 102 , as shown in FIG. 16, is achieved. We observe that the SMA is able to provide over 5 N of contraction force in addition to the force required to overcome the bias (that is, the actuator can produce 5 N of usable force). The inherent stroke limitation prevents the SMA actuator 22 from damaging the articulating laminate structure 18, which only requires 1 N to deform, as designed. As such, 4 N is left over for bending the cautery tool 20 to withstand tissue reaction forces. Finally, FIG. 17 shows how the force develops over time as a function of input current given a step input, lending some insight into the system bandwidth.

E. Integrated System Validation

The articulating end effector was attached to a transmission module, and two 140-mm long, 0.006-inch-diameter actuator wires were routed around the bearing transmission and through the articulating module. The individual tensioner blocks were tightened up to about mid-stroke for each SMA (allowing the opposing SMA to "relax" as the actuated SMA contracts). A step current of 360 mA was applied for 10 seconds, and the on-board sensor readings were collected at a rate of 500 Hz. The final deformed shape was measured and compared to the sensor readings. FIG. 18 shows the differential on-board sensor readings 104 over a positive actuation angle given step input currents 94. The system is observed to be able to achieve 25:5 ± 2:25 degrees of motion for positive and negative angles, leading to a total of 51 ± 4:5 degrees; and the sensors were able to resolve this deflection. It was observed that some stroke was lost in deforming the distal-most spacer disk; accordingly, the system can be made more robust by using thicker layers of material.

Additionally, the elasticity of the integrated flexural element was observed to be insufficient to pre-bias the SMAs {i.e., when one SMA had actuated and deformed the structure, in order to straighten the structure out again by returning the actuated wire to its untwinned martensitic state, the antagonistic actuator had to be actuated to provide the necessary bias force). Additional iterations can optimize the trade-off between the flexure's ability to provide both the required dexterity as well as the passive biasing force the SMAs need to relax. From a controllability

perspective, this is beneficial, as the structure would passively straighten.

The system was also tested for its lateral force generation capabilities. A mock instrument [silicone tubing with a 0.020-inch-diameter nickel-titanium-alloy (nitinol) tube inside] was fed through the instrument port, and the system was actuated against a load cell. A pulse of 360 mA was provided to heat the SMA on one side, thereby causing it to actuate against the load cell; and the reaction force was captured at a rate of 500 Hz. The resulting force profile shown in FIG. 19 shows that the system can generate 450 mN of lateral force when actuated and sustain this force for several seconds, thereby satisfying the force requirement.

F. Helical SMA Actuation: Theoretical Analysis

The performance of the current system is limited in large part by the stroke, high stiffness (which makes pre-tensioning a challenge in antagonistic systems) and bandwidth limitations of the SMA wire actuators. In the interest of improving stroke and stiffness tunability characteristics, a preliminary analysis was performed to determine the geometric properties required from a spring-wound helical SMA actuator that would improve the stroke while maintaining similar blocking force characteristics. A design tool was implemented in MATLAB software, using constitutive equations found in B. Holschuh and D. Newman, "Low spring index, large displacement shape memory alloy (SMA) coil actuators for use in macro- and microsystems," Proc. of SPIE, vol. 8975, p. 897505, 2014, to generate a parametric space of the potential net force output as a function of reasonable spring geometrical parameters (wire diameter, d, and coil outer diameter, D) and user-inputted parameters (design force, F, and stroke, S).

An example of a parametric map is shown in FIG. 20, where valid geometrical solutions are those that generate actuation forces that (1) meet or exceed the design force requirements and (2) induce shear stresses in the SMA which are less than the cyclic limit of the material. In this exemplification, wire diameter was -0.5 mm; coil diameter was ~2.5 mm; 14 coils were used; and pitch was -0.725 mm. Meanwhile, this exemplification exhibited the following performance characteristics: static length = -10.15 mm; stretched length = -20.30 mm; blocking force = -9.11 N; stroke margin = 3.5 mm; force margin at design stroke = -1.836 N; cooling time (free) = -4.15 seconds; and cooling time (forced) = -0.830 seconds. Note that more

sophisticated models incorporate martensitic de-twinning phenomena, but these effects are assumed to be negligible for sub-critical shear stress levels. For the configuration of wire diameter (d) = 0.5 mm and coil diameter (D) = 2.5 mm, FIG. 21 demonstrates a usable work range over about 3.5 mm of stroke, thereby exceeding the stroke requirements of the current system and generating 70 degrees of deflection.

The bandwidth is limited by the cooling time of the SMA, which is a thermodynamic process that scales linearly with wire radius. Although it was determined from the application that circumferential incision is a low-bandwidth process, developing faster actuation methodologies will invariably improve system controllability. By implementing forced convection cooling, thermodynamic analyses, provided in FIG. 22 for free convection 106, forced convection (air) 108, and forced convection (fluid) 110 show that cooling time can be sped up significantly, making >Hz bandwidth achievable.

In additional embodiments, the stroke can be increased; and the actuators can be packaged such that they are not exposed to biological tissue. Alternative actuation strategies can also be employed {e.g., electrostatic, hydraulic/pneumatic). Soft materials can also be integrated to encapsulate the articulating structure for added robustness. Further still, alternative materials {e.g., superelastic) can be used for the flexure mechanism, as well as multi-DoF articulation by rotating subsequent articulating modules with respect to each other. The system can be adapted to a commercially available endoscope. Closed-loop control of the system can be employed using on-board sensor data to implement lower-level control to enable subtask automation.

Additional examples consistent with the present teachings are set out in the following numbered clauses:

1. A robotic surgical module, including a distal structure comprising:

a robotic backbone configured for mounting on a distal end of an endoscope; and

an end effector extending from the robotic backbone, wherein the end effector includes (a) an expanded structural framework; (b) a surgical tool extending through the expanded structural framework; and (c) an actuator mounted to the structural framework and configured to flex the expanded structural framework and the surgical tool when actuated.

2. The robotic surgical module of clause 1, wherein the actuator is selected from a fluid-based actuator and a shape memory alloy.

3. The robotic surgical module of clause 1 or 2, wherein the actuator is entirely on or in the end effector.

4. The robotic surgical module of any of clauses 1-3, further comprising a

control system in or on the end effector.

5. The robotic surgical module of clause 4, further comprising a proprioceptive sensor on or in the end effector and in communication with the control system to provide a distal feedback loop.

6. The robotic surgical module of any of clauses 1-5, wherein the robotic surgical module is configured for replaceable mounting on and detachment from the endoscope.

7. The robotic surgical module of any of clauses 1-6, wherein the surgical tool is a cautery device.

8. The robotic surgical module of any of clauses 1-7, wherein the structural

framework is a laminate structure.

9. The robotic surgical module of clause 8, wherein the laminate structure

includes a plurality of spaced-apart rigid plates joined with a more-flexible layer and an elastic flexure passing through at least one of the rigid plates.

10. The robotic surgical module of any of clauses 1-9, further comprising a light emitting diode mounted to a first end of the end effector and a

phototransistor mounted to a second end of the end effector, wherein the light emitting diode and phototransistor are configured such that an amount of light from the light emitting diode incident upon the phototransistor changes as the end effector flexes.

11. A method for surgery using a robotic surgical module, including a distal

structure comprising (i) a robotic backbone configured for mounting on a distal end of an endoscope and (ii) an end effector extending from the robotic backbone, wherein the end effector includes (a) an expanded structural framework; (b) a surgical tool extending through the expanded structural framework; and (c) an actuator mounted to the structural framework, the method comprising:

inserting the distal structure into a body of an organism; and actuating the actuator to flex the expanded structural framework and the surgical tool inside the body.

12. The method of clause 11, further comprising sensing forces with a

proprioceptive sensor on or in the end effector inside the body and governing the actuation of the actuator based on the forces sensed by the proprioceptive sensor.

13. The method of clause 11 or 12, wherein the structural framework is a laminate structure.

14. The method of clause 13, wherein the laminate structure includes a plurality of spaced-apart rigid plates joined with a more-flexible layer and an elastic flexure passing through at least one of the rigid plates.

15. The method of any of clauses 11-14, further comprising:

emitting light from a light emitting diode mounted to a first end of the end effector to a phototransistor mounted to a second end of the end effector; and

determining end effector orientation based on the amount of light from the light emitting diode incident upon the phototransistor changes as the end effector flexes.

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 (or where methods are elsewhere recited), 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 is:

1. A robotic surgical module, including a distal structure comprising:

a robotic backbone configured for mounting on a distal end of an endoscope; and

2. The robotic surgical module of claim 1, wherein the actuator is selected from a fluid-based actuator and a shape memory alloy.

3. The robotic surgical module of claim 1, wherein the actuator is entirely on or in the end effector.

4. The robotic surgical module of claim 1, further comprising a control system in or on the end effector.

5. The robotic surgical module of claim 4, further comprising a proprioceptive sensor on or in the end effector and in communication with the control system to provide a distal feedback loop.

6. The robotic surgical module of claim 1, wherein the robotic surgical module is configured for replaceable mounting on and detachment from the endoscope.

7. The robotic surgical module of claim 1, wherein the surgical tool is a cautery device.

8. The robotic surgical module of claim 1, wherein the structural framework is a laminate structure.

9. The robotic surgical module of claim 8, wherein the laminate structure

10. The robotic surgical module of claim 1, further comprising a light emitting diode mounted to a first end of the end effector and a phototransistor mounted to a second end of the end effector, wherein the light emitting diode and phototransistor are configured such that an amount of light from the light emitting diode incident upon the phototransistor changes as the end effector flexes.

11. A method for surgery using a robotic surgical module, including a distal

inserting the distal structure into a body of an organism; and

actuating the actuator to flex the expanded structural framework and the surgical tool inside the body.

12. The method of claim 11, further comprising sensing forces with a

13. The method of claim 11, wherein the structural framework is a laminate

structure.

14. The method of claim 13, wherein the laminate structure includes a plurality of spaced-apart rigid plates joined with a more-flexible layer and an elastic flexure passing through at least one of the rigid plates.

15. The method of claim 11, further comprising: