FORCE FEEDBACK GRIPPING DEVICE WITH MAGNETORHEOLOGICAL BASED ACTUATOR

[0001] The present invention generally relates to force feedback gripping devices, particularly for medical interventions. The present invention specifically relates to force feedback control of a magnetorheological elastomer (“MRE”) in actuating a mechanical gripper for applying a gripping force upon an object, particularly an anatomical object (e.g., an organ, a blood vessel, a foreign body, etc.).

[0002] Most of the conventional grippers/tweezers used in medicine and industrial manipulation do not provide force-feedback. The lack of force controlled motion can lead to damage of a manipulated object (e.g., tissue tearing in medical applications) or damage to the gripper, which may cause inflammation if part of the gripper remains in the patient's body, or structural defect in manufacturing.

[0003] Force-feedback grippers known in art are using conventional actuation, such as electro-motors and mechanical transmission. Response time of these actuators is typically low since it depends on inertial moment of the transmission unit. Furthermore, these assemblies are difficult to be miniaturized since they are manufactured from a lot of moving parts. The same is true for pneumatic actuators.

[0004] Alternatives, such as electrothermal actuators can be miniaturized, but dissipate heat which may not be acceptable in some applications both in medicine and in manufacturing.

[0005] One form of the present invention is a magnetorheological elastomer (“MRE”) gripping device employing a mechanical gripper and electromagnetic actuator. The mechanical gripper is operable to be actuated to one of a plurality of gripping poses for gripping an object. The electromagnetic actuator includes a MRE, wherein the MRE is operable to be transitioned between a plurality of shapes dependent upon a variable strength of a magnetic field applied to the MRE, and wherein each shape of the MRE actuates the mechanical gripper to one of the gripping poses.

[0006] For purposes of the present invention, a “shape” of the MRE broadly encompasses any general shape (e.g., circles, spheres, polygons, prisms, cylinders, cones, etc.), and a “plurality of shapes” of the MRE broadly encompasses (1) variances in one or more dimensions of the same general shape, and/or (2) a transition between two (2) general shapes. For example, MRE 22a may have a general prism shape that is in cubic form when a minimal strength of the magnetic field is applied to the MRE and that is elongated to a rectangular form when a maximal strength of the magnetic field is applied to the MRE, or vice-versa. By further example, the MRE may have a cylinder shape when a minimal strength of the magnetic field is applied to the MRE that transitions to a cone shape as the maximal strength of the magnetic field is applied to the MRE, or vice-versa.

[0007] A second form of the present invention is a force feedback gripping device employing a mechanical gripper, an electromagnetic actuator and a force feedback controller. The mechanical gripper is operable to be actuated to one of a plurality of gripping poses for gripping an object. The electromagnetic actuator includes a MRE, wherein the MRE is operable to be transitioned between a plurality of shapes dependent upon a variable strength of a magnetic field applied to the MRE, and wherein each shape of the MRE actuates the mechanical gripper to one of the gripping poses. The force feedback controller is operable to control the strength of the magnetic field applied to the MRE based on an estimation of a gripping force of the mechanical gripper and on a sensing of a load force of the object responsive to the gripping force of the mechanical gripper.

[0008] A third form of the present invention is a method of controlling the aforementioned force feedback gripper device. The method involves an application of a magnetic field to the MRE, wherein the MRE is transitioned between the plurality of shapes dependent upon a variable strength of a magnetic field, and wherein each shape of the MRE actuates the mechanical gripper to one of the plurality of gripping poses for gripping the object. The method further involves a control of the variable strength of the magnetic field applied to the MRE based on an estimation of a gripping force of the mechanical gripper and on a sensing of a load force of the object responsive to the gripping force of the mechanical gripper.

[0019] Referring to FIG. 1, a force feedback gripping device 20 of the present invention employs a force feedback controller 21, an electromagnetic actuator 22, a mechanical gripper 23 and a load sensor 24 for gripping an object 10, particularly an anatomical object (e.g., an organ, a blood vessel, a foreign body, etc.). For purposes of the present invention, the term “gripping” is broadly defined herein as any forceful contact of device 20 upon object 10 including, but not limited to, device 20 gripping, grasping, clamping, squeezing, clutching and/or holding object 10.

[0020] In general operation, force feedback controller 21 communicates an actuation signal AS to electromagnetic actuator 22 to control a generation by electromagnetic actuator 22 of a magnetic field H_vfs of variable strength that is applied to a magnetorheological elastomer (“MRE”) 22a. The strength of magnetic field H_vfs determines a two-dimensional (“2D”) or three-dimensional (“3D”) shape of MRE 22a, and any variation in the strength of magnetic field H_vfs alters a shape of MRE 22a between a baseline shape associated with a minimal strength of magnetic field H_vfs (e.g., zero magnetic flux density B of MRE 22a) and a fully actuated shape associated with a maximal strength of magnetic field H_vfs (e.g., saturated magnetic flux density B of MRE 22a).

[0021] In practice, a “shape” of MRE 22a broadly encompasses any general shape (e.g., circles, spheres, polygons, prisms, cylinders, cones, etc.), and a “plurality of shapes” of MRE 22a broadly encompasses (1) variances in one or more dimensions of the same general shape, and/or (2) a transition between two (2) general shapes. For example, MRE 22a may have a general prism shape that is in cubic form when a minimal strength of magnetic field H_vfs is applied to MRE 22a and that is elongated to a rectangular form when a maximal strength of magnetic field H_vfs is applied to MRE 22a, or vice-versa. By further example, MRE 22 may have a cylinder shape when a minimal strength of magnetic field H_vfs is applied to MRE 22a that transitions to a cone shape as the maximal strength of magnetic field H_vfs is applied to MRE 22a, or vice-versa.

[0022] Additionally, the use of the terms “minimal” and “maximum” for the strength magnetic field H_vfs is used for relative value purposes and not absolute minimum and maximum magnetic field strength values that are capable of being generated by electromagnetic actuator 22. Thus, in practice, the minimal strength of magnetic field H_vfs may be a non-zero magnetic field and the maximal strength of magnetic field H_vfs may or may not correspond to a saturated magnetic flux density B of MRE 22a.

[0023] Still referring to FIG. 1, mechanical gripper 23 is coupled to MRE 22a whereby each shape of MRE 22a actuates mechanical gripper 23 to one of a plurality of gripping poses between a minimal gripping pose corresponding to the baseline shape of MRE 22a and a maximum gripping pose corresponding to the fully actuated shape of MRE 22. As mechanical gripper 23 is actuated to grip object 10, object 10 experiences a load force LF dependent upon a gripping force GF of mechanical gripper 23, contact surfaces of mechanical gripper 23 and object 10, material properties of mechanical gripper 23 and object 10 (e.g., yield strength, fatigue strength, tensile strength, compressive strength, toughness, modulus of elasticity, ductility, etc.) and physical properties of mechanical gripper 23 and object 10 (e.g., mass, cross sectional area, etc.). Load sensor 24 communicates a load signal LS to force feedback controller 21 that is indicative of load force LF and in response thereto, force feedback controller 21 utilizes the sensing of load force LF via load signal LS and an estimation of gripping force GF to control the variable strength of magnetic field H_vsf via actuation signal AS with the objective of mechanical gripper 23 securely gripping object 10 without inflicting any damage to object 10.

[0024] The following is a definitional description of force feedback controller 21, electromagnetic actuator 22, mechanical gripper 23 and load sensor 24.

[0025] For purposes of the present invention, force feedback controller 21 is broadly defined herein as any data processing structure configured for controlling the variable strength of magnetic field V_vsf via actuation signal AS based on an estimation of gripping force GF and on a sensing of load force LF via load signal LS. Examples of force feedback controller 21 include, but are not limited to, a microprocessor (e.g., a digital signal processor), a microcontroller, dedicated hardware/firmware, programmable hardware and reprogrammable firmware. In practice, actuation signal AS may have any electrical signal form (e.g., voltage or current, DC or AC, etc.) and may be communicated to electromagnetic actuator 22 by any form of electric signal transmission (e.g., electrical or optical, wired or wireless, etc.). Also in practice, force feedback controller 21 may be integrated with magnetic actuator 22 and/or mechanical gripper 23, or physically remote from magnetic actuator 22 and mechanical gripper 23.

[0026] For purposes of the present invention, electromagnetic actuator 22 is broadly defined herein as any electromagnetic structure employing any type of elastomer having magnetorheological fluid/particles whereby elastic properties of the elastomer are controllable by a locally generated variable strength magnetic field H_vrf. In practice, MRE 22a may have any material composition suitable for desired application(s) of device 20. In one embodiment, MRE 22a is composed of ferromagnetic particles embedded in a polymer matrix. For example, the ferromagnetic particles are 3-5 μm pure iron embedded in a rubber matrix (e.g., silicone rubber).

[0027] Also in practice, MRE 22a may have any alignment relative to the locally generated variable magnetic field H_vfs. Furthermore, MRE 22a may be directly coupled to mechanical gripper 23 or electromagnetic actuator 22 may additionally include component(s) coupling MRE 22a to mechanical gripper 23 (e.g., a spring, a lever, etc.).

[0028] For purposes of the present invention, mechanical gripper 23 is broadly defined herein as any mechanical structure employing one or more components for gripping object(s) of any type in response to physical forces being applied to the component(s) from MRE 22a. In practice, the structural configuration of mechanical gripper 23 is dependent upon particular industry(ies) having use of mechanical gripper 23 (e.g., medical, biomedical, manufacturing, semiconductor, etc.). For the medical industry, exemplary uses of mechanical gripper 23 include, but are not limited to, gripping of tissue in conventional surgery (e.g. laparoscopy) and microsurgery (e.g. eye surgery), gripping of small particles (e.g., blood clots or foreign bodies) or micro-gripping. Also, in practice, mechanical gripper 23 may be an end effector of a robotic system.

[0029] For purposes of the present invention, load sensor 24 is broadly defined herein as any sensor structure configured for converting a load force acting on the structure into any type of electronic signal indicative of a magnitude of the load force. Examples of load sensor 24 include, but are not limited to, load cell sensors, force sensors, torque sensors and pressure sensors. In practice, load signal LS may have any electrical signal form (e.g., voltage or current, DC or AC, etc.) and may be communicated to force feedback controller 21 by any form of electric signal transmission (e.g., electrical or optical, wired or wireless, etc.). Also in practice, one or more load sensors 24 may be employed and each load sensor 24 may be coupled to mechanical gripper 23 adjoined or adjacent to contact surfaces of mechanical gripper 23 in a manner than facilitates a sensing of the load force upon object 10.

[0030] The following is a description of exemplary embodiments of force feedback controller 21, electromagnetic actuator 22 and mechanical gripper 23 as shown in FIGS. 2-9 to facilitate further understanding of the present invention.

[0031] One exemplary embodiment 20a of device 20 is shown in FIG. 2A in an open state and in FIG. 2B in a closed state. Mechanical gripper 23 of device 20a employs a gripper frame having a gripper arm 23a pivotally connected to a lever 23c and a gripper arm 23b fixedly connected to lever 23c. The gripper frame further has a spring 23d for biasing mechanical gripper 23 to the open state as shown in FIG. 2A.

[0032] A load sensor 24a of device 20a is adjoined to a distal contact surface of gripper arm 23a, and a load sensor 24b of device 20a is adjoined to a distal contact surface of gripper arm 23b. Alternatively, load sensors 24a and 24b may be adjacent distal contact surfaces of respective gripper arms 23a and 23b.

[0033] A proximal end of gripper arm 23a and a proximal end of gripper arm 23b are connected to electromagnetic actuator 22. This establishes a parallel alignment of electromagnetic actuator 22 and spring 23d relative to lever 23c. MRE 22a of electromagnetic actuator 22 (symbolically shown by the particle dots) has a shape memory property created by a molding of MRE 22a in a baseline shape shown in FIG. 2A (e.g., a cubic prism shape).

[0034] As shown in FIG. 2A, MRE 22a maintains the baseline shape upon an application a minimal field strength H_min of magnetic field H_vfs (e.g., a zero magnetic field), and gripper arms 23a and 23b are biased by spring 23d to a minimal gripping pose representative of an open state of device 20a. The baseline shape of MRE 22a corresponds to a minimum stiffness of MRE 22a in view of a minimal magnetic flux density B_min and as such a minimal gripping force GF of gripper arms 23a and 23b.

[0035] To close device 20a upon object 10, force feedback controller 21 (not shown) controls an increase in the strength of magnetic field H_vfs toward a maximal strength H_max as shown in FIG. 2b resulting in a parallel expansion of MRE 22a from the baseline shape to a fully actuated shape (e.g., a rectangular prism shape) associated with the maximal strength H_max of magnetic field H_vfs (e.g., a saturated magnetic field). More particularly, as the strength of magnetic field H_vfs is increased, the stiffness of MRE 22a increases responsive to an increasing magnetic flux density B thereby increasing the gripping force GF of gripper arms 23a and 23b to a maximal level corresponding to the fully actuated shape of MRE 22a.

[0036] Load sensors 24a and 24b communicate load signals to force feedback controller 21 (not shown) whereby force feedback controller 21 controls the increase in the strength of magnetic field H_vfs towards the maximal strength H_max with the objective of securely gripping object 10 without inflicting any damage to object 10. As such, force feedback controller 21 may increase the strength of magnetic field H_vfs to a lower strength than maximal strength H_max if the load signals L in conjunction with an estimation of the increasing gripping force GF indicate object 10 may experience damage if the strength of magnetic field H_vfs is further increased toward the maximal strength H_max.

[0037] Another exemplary embodiment 20b of device 20 is shown in FIG. 3A in an open state and in FIG. 3B in a closed state. Mechanical gripper 23 of device 20b employs a gripper frame having a gripper arm 23e and a gripper arm 23f both pivotally connected to a lever 23g, and a spring 23h connected lever 23g and to a distal side of magnetic actuator 22.

[0038] A load sensor 24a of device 20b is adjoined to a distal contact surface of gripper arm 23e, and a load sensor 24b of device 20b is coupled to gripper arm 23f adjoined to a distal contact surface of gripper arm 23f. Alternatively, load sensors 24a and 24b may be adjacent distal contact surfaces of respective gripper arms 23e and 23f A proximal end of gripper arms 23e and 23f are pivotally coupled to a proximal side of electromagnetic actuator 22. This establishes a series alignment of electromagnetic actuator 22 and spring 23g relative to gripper arms 23e and 23f For this embodiment, MRE 22a of electromagnetic actuator 22 (symbolically shown by the particle dots) has a shape memory property created by a molding of MRE 22a in a baseline shape shown in FIG. 3A (e.g., a rectangular prism shape).

[0039] As shown in FIG. 3A, MRE 22a maintains the baseline shape upon an application a minimal field strength H_min of magnetic field H_vfs (e.g., a zero magnetic field), and gripper arms 23e and 23f are biased by spring 23d to a minimal gripping pose representative of an open state of device 20b. The baseline shape of MRE 22a corresponds to a minimum stiffness of MRE 22a in view of a minimal magnetic flux density B_min and as such a minimal gripping force GF of gripper arms 23e and 23f.

[0040] To close device 20b upon object 10, force feedback controller 21 (not shown) controls an increase in the strength of magnetic field H_vfs toward a maximal strength H_max as shown in FIG. 3b resulting in a series compression of MRE 22a from the baseline shape to a fully actuated shape (e.g., a cubic prism shape) associated with the maximal strength H_max of magnetic field H_vfs (e.g., a saturated magnetic field). More particularly, as the strength of magnetic field H_vfs is increased, the stiffness of MRE 22a increases responsive to an increasing magnetic flux density B thereby increasing the gripping force GF of gripper arms 23e and 23f to a maximal level corresponding to the fully actuated shape of MRE 22a.

[0041] Load sensors 24a and 24b communicate load signals to force feedback controller 21 (not shown) whereby force feedback controller 21 controls the increase in the strength of magnetic field H_vfs toward a maximal strength H_max with the objective of securely gripping object 10 without inflicting any damage to object 10. As such, force feedback controller 21 may increase the strength of magnetic field H_vfs to a lower strength than maximal strength H_max if the load signals L in conjunction with an estimation of the increasing gripping force GF indicate object 10 may experience damage if the strength of magnetic field H_vfs is further increased to a maximal strength H_max.

[0042] For device 20a (FIG. 2), device 20b (FIG. 3) and any other embodiment of device 20 of the present invention, FIGS. 4-6 illustrate various embodiments of an electromagnet of electromagnetic actuator 22. Specifically, FIG. 4 shows a coil 22b wrapped around a cylindrical MRE 22a whereby a magnetic field (not shown) generated by coil 22b is aligned along a longitudinal axis of MRE 22a. Coil 22b generates the magnetic field responsive to current flowing through coil 22b whereby MRE 22a may be elongated as shown by the dashed lines along the longitudinal axis of MRE 22a in dependence upon the strength of the magnetic field.

[0043] FIG. 5 shows a serial bipolar arrangement coil 22b and a coil 22c relative to cylindrical MRE 22a whereby magnetic fields (not shown) generated by coils 22b and 22c are parallel to a longitudinal axis of MRE 22a. Coils 22b and 22c generate the magnetic fields when current flows through coils 22b and 22c whereby MRE 22a may be elongated as shown by the dashed lines along the longitudinal axis of MRE 22a in dependence upon the strength of the magnetic field.

[0044] FIG. 6 shows coil 22b wrapped around an iron core 22d providing an air gap around cylindrical MRE 22a whereby a magnetic field (not shown) generated by coil 22b magnetizes iron core 22d, which in turn magnetizes MRE 22a. Coil 22b generates the magnetic field (not shown) responsive to current flowing through coil 22b whereby MRE 22a may be elongated as shown by the dashed lines along the longitudinal axis of MRE 22a in dependence upon the strength of the magnetic field.

[0045] For FIGS. 4-6, the cylindrical shape of MRE 22a as shown may serve as the baseline shape or the fully actuated shape.

[0046] For device 20a (FIG. 2), device 20b (FIG. 3) and any other embodiment of device 20 of the present invention, FIG. 7 illustrates a flowchart representative of gripping force estimation method performed by force feedback controller 21. Specifically, gripping force GF of mechanical gripper 23 is a function of a stiffness of MRE 22a, which is controlled by the magnetic flux density of MRE 22a. In practice, gripping force GF of mechanical gripper 23 may increase as the stiffness of MRE 22a increases as shown in FIGS. 2 and 3, or decrease as the stiffness of MRE 22a increases in dependence upon a particular coupling of electromagnetic actuator 22 to mechanical gripper 23. Force feedback controller 21 therefore utilizes the elastic modulus of MRE 22a as a basis for estimating the gripping force GF of mechanical gripper 23.

[0047] Referring to FIG. 8, a stage S31 of flowchart 30 encompasses an electromagnetic processing 42 of a current signal I(t) for an electromagnet to yield a magnetic flux density B(t). In one embodiment of stage S31, electromagnetic processing 42 includes an execution of the following equation [1]:

B(t)=μ_mnI(t)  [1]

[0048] where μ_m is magnetic permeability of MRE 22a and n is density of wire turns in a coil of the electromagnet.

[0049] A stage S32 of flowchart 30 encompasses a magnetorheological processing 43 of magnetic flux density B(t) to yield an elastic constant km(t) of MRE 22a. In one embodiment of stage S32, magnetorheological processing 43 involves an experimental determination of elastic constant k_m(B) as known in the art. For example, FIG. 9 shows a characteristics curve determined by experiment of a particular designed MRE describing a relationship between elastic modulus (E) and magnetic flux density E=f(B). For this curve, the magnetorheological processing 43 includes an execution of the following equation [2]:

k_m(t)=(EB(t)A)/L  [2]

[0050] where A is cross sectional area of MRE 22a and L is a length of MRE 22a when is not compressed.

[0051] A stage S33 of flowchart 30 encompasses an elastic processing 44 of elastic constant k_m(t) to yield a displacement d(t) of gripper arms from the open state. In one embodiment of stage S33, elastic processing 44 of elastic constant k_m(t) may be based on physical forces on electromagnetic actuator 22, the spring and a geometry of mechanical gripper 24. For simplicity, this embodiment assumes (1) physical forces on mechanical gripper 24 are negligible compared to forces in biasing component and on MRE 22a, (2) MRE 22a and the spring are not changing their shape if fully compressed, (3) a length of the spring in resting position is the exactly a distance between the contact surfaces of the gripper arms when parallel and (4) a length of MRE 22a in resting position is shorter than a distance between the contact surfaces of the gripper arms when parallel. Thus, further assuming that two forces applied on the lever are a force F_S on the spring and a force F_m on MRE 22a, equalizing the moments and using elastic force as F=−k(t)*Δx, where Δx is an elongation of MRE 22a, the forces are in accordance with the following equations [3] and [4]:

F_S·l_S=F_M·l_M  [3]

k_S·l_S·Δx_S=k_M·l_M·Δx_M  [4]

[0052] Further, using trigonometric relations, relationship between vertical displacements from the fulcrum point in accordance with the following equation [5]:

$\begin{array}{cc}\Delta x_s^{\prime }=\frac{l_S}{l_M}·\Delta x_M^{\prime }& \left[5\right]\end{array}$

[0053] Note that displacement from the fulcrum and elastic displacement are not always equal. Under the initial assumptions, Δx′_S=Δx_S and Δx_M=Δx′_M+Δx₀ where Δx₀ is additional extension of the MRE 22a with respect to the fulcrum point. Combining force equation and geometric equation yields the following equation [6]:

$\begin{array}{cc}\Delta x_M^{\prime }=\left(1-\frac{k_S·l_S^2}{k_M·l_M^2}\right)\Delta x_0& \left[6\right]\end{array}$

[0054] Total opening of the gripper jaws is therefore in accordance with the following equation [7]:

$\begin{array}{cc}d=h+\frac{l}{l_M}\left(1-\frac{k_S·l_S^2}{k_M·l_M^2}\right)\Delta x_0& \left[7\right]\end{array}$

[0055] Equation (7) describes the elastic component of the control system. Elastic constant of the spring k_S is constant and depends on the design and material of the spring. Lengths l, l_M, l_S and height h are constant and depend on design of the gripper. Δx₀ is also constant. Thus, the opening of the gripper jaws is proportional to k_m(t) with a transfer function is described by magnetorheological component.

[0056] A stage S34 of flowchart 30 encompasses a force processing 45 of displacement d(t) to yield a gripping force GF(t). In one embodiment of stage S34, gripping force GF(t) may be assumed to be linear: F=−k_oΔx_o, where k_o is an elastic constant of the object and Δx_o is deformation of the object. As deformation of the object is directly proportional to jaw opening d(t), it can be assumed that the gripping force GF(t) is also directly proportional to jaw opening d(t) for any given gripper task.

[0057] Referring to FIG. 8, force feedback control 21 executes software and/or firmware having instructions for implementing a force feedback control scheme incorporating gripper force estimation method 30 shown in FIG. 7. For this control scheme, a current generator 41 processes a differential of a load force LF(t) and gripping force GF(t) to yield current I(t) for the coil of the electromagnet. In one embodiment, current I(t) is a function of the differential of load force LF(t) and gripping force GF(t) as shown in the following Table 1:

	Load Force/Gripping
	Force Differential	Current
	<0	Increase
	=0	Maintain
	>0	Decrease

[0058] In practice, an offset X may be subtracted or added to the Load Force/Gripping Force Differential of Table 1 in dependence upon the particular object 10 being gripped by device 20.

[0059] Also in practice, since the actual force component of the transfer function is not known, an adaptable control 46 as shown in FIG. 8 may be applied. In one embodiment, adaptable control 46 may be realized as adaptive gain component as shown (unknown proportional contest) and force feedback controller 31 may be generalized proportional-integral-derivative (“PID’) controller. This embodiment can be used to provide real-time control of the gripping force GF(t).

[0060] Referring to FIGS. 1-9, those having ordinary skill in the art will appreciate numerous benefits of the present invention including, but not limited to, a force feedback gripping device securely gripping an object via control of a MRE for minimizing, if not preventing, an infliction of damage to the object by the device.

[0061] While various embodiments of the present invention have been illustrated and described, it will be understood by those skilled in the art that the embodiments of the present invention as described herein are illustrative, and various changes and modifications may be made and equivalents may be substituted for elements thereof without departing from the true scope of the present invention. In addition, many modifications may be made to adapt the teachings of the present invention without departing from its central scope. Therefore, it is intended that the present invention not be limited to the particular embodiments disclosed as the best mode contemplated for carrying out the present invention, but that the present invention includes all embodiments falling within the scope of the appended claims.