Force sensor design and readout method

The force sensor design with interdigitated electrodes and a 555 timer readout method addresses the limitations of existing sensors by providing durable, accurate tri-axial force sensing suitable for industrial use.

WO2026239107A1PCT designated stage Publication Date: 2026-11-19TOUCHLAB LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/GB2026/050829
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-11-11
Filing Date
2026-05-13
Publication Date
2026-11-19

AI Technical Summary

Technical Problem

Existing force sensors for robotic limbs are not robust enough to withstand real-world conditions, lack tri-axial force sensing capabilities, and are sensitive to temperature and environmental factors, making them unreliable for applications requiring precise force direction and magnitude measurements.

Method used

A force sensor design featuring interdigitated electrodes with a force-sensitive material and a spacer layer, coupled with a 555 timer readout method, allowing for high-accuracy tri-axial force sensing and durability, enabling reliable operation in harsh environments.

Benefits of technology

The sensor provides accurate tri-axial force measurements, is insensitive to bending and environmental conditions, and can withstand thousands to millions of cycles, making it suitable for industrial applications.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure GB2026050829_19112026_PF_FP_ABST
    Figure GB2026050829_19112026_PF_FP_ABST
Patent Text Reader

Abstract

Force Sensor Design and Readout Method A force sensor design and readout method for a force sensor comprising at least one pair of electrodes, is described. The method comprises: coupling a first node of the pair of electrodes to a threshold pin and a trigger pin of a 555 timer configured to operate in astable mode and coupling a second node of the pair of electrodes to an output pin of the 555 timer. The pair of electrodes is arranged in series with a timing capacitor of the 555 timer. The method further comprises determining a force applied to the force sensor based on a frequency of an output pulse train at the output pin of the 555 timer; wherein the timing capacitor is configured such that when the force sensor is subjected to force a time required to charge the timing capacitor decreases, thereby increasing the frequency of the output pulse train.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] Force Sensor Design and Readout Method

[0002] TECHNICAL FIELD OF THE DISCLOSURE

[0003] The present disclosure relates to a force sensor design and readout method.

[0004] BACKGROUND

[0005] Incorporating sensors into robotic limbs broadens the scope of tasks which can be achieved with robots, providing touch-based feedback for improved perception and control.

[0006] A variety of sensor types have been used for robotic limbs: strain gauges; Halleffect position sensors; motor speed sensors; cameras for vision-based systems; and others. In many cases, the sensor is placed on the ‘fingertip’ of the robotic limb, underneath an artificial skin layer. The artificial skin acts as a protective layer for the underlying sensors, and brings the robotic limb closer to a human hand in terms of look and feel. This has advantages in the fields of human-robot interaction, prosthetic limbs, and bio-inspired design.

[0007] Commercial flexible force sensors on the market only give a scalar value for force, but not its direction or shear forces. They are not capable of withstanding significant flexing, while also changing their value during flexing. Furthermore, array technology, including tri-axial technology is required in order to get the necessary data to enable multiple advanced applications. This is particularly important in e.g. gait analysis of humans and robots which need to measure ground reaction force, robots manipulating objects in a complex manner, robots handling delicate objects without damaging them, monitoring physiological forces such as muscle activity and blood flow on humans with wearable garments, sensing flow phenomena of fluids such as air or water on top of tidal and wind turbines, allowing surgeons to ‘palpate’ tissues with surgical robots, etc.

[0008] Compliant Tri-axial sensors can be created (in research labs) using multiple transduction mechanisms such as: piezoresistive (e.g. using strain gauges or nanocomposites); capacitive; piezoelectric; inductive; optical; or magnetic. Piezoresistive and capacitive sensors dominate due to their excellent performance. However, capacitive sensors can be sensitive to the substrate they are placed on due to capacitive coupling with this surface, and require much more complex readout electronics which is a hindrance due to space requirements and implementation complexity. Drawbacks of the other sensors include: piezoelectric sensors cannot sense

[0009] 55853629-1static contact force since the induced charge in the material dissipates very quickly; most of these sensors are sensitive to temperature, and are generally not robust; there is a lack of dynamic force sensing range; EMC noise-sensitive (especially in magnetic and piezoelectric sensors); some piezoelectric sensors are light-sensitive. In addition, most known processes for making these sensors are inherently low-yield, inconsistent and not scalable. Often, measurements with such sensors are not repeatable making the sensors unreliable.

[0010] Of all the sensors available, none are robust-enough to function in a desirable manner when in real-world settings. There is an urgent need in industry for a robust sensor, with durable materials which can withstand thousands to millions of cycles, is insensitive to temperature and harsh environmental conditions, and which can perform tri-axial force sensing. The sensor must also be extremely thin so that it can be added to existing devices without disturbing their functioning, and the working of the sensor should be insensitive to bending so that it can also function on soft surfaces such as soft robots or human skin.

[0011] It is an aim of at least one aspect of the present disclosure to provide a force sensor design and readout method which address one or more of the problems discussed above.

[0012] SUMMARY

[0013] The inventors have proposed a force sensor design capable of high accuracy triaxial force sensing and which is compact, flexible and durable.

[0014] According to a first aspect of the disclosure there is provided a readout method for a force sensor comprising at least one pair of electrodes, the method comprising:

[0015] coupling a first node of the pair of electrodes to a threshold pin and a trigger pin of a 555 timer configured to operate in astable mode;

[0016] coupling a second node of the pair of electrodes to an output pin of the 555 timer; wherein the pair of electrodes is arranged in series with a timing capacitor of the 555 timer; and

[0017] determining a force applied to the force sensor based on a frequency of an output pulse train at the output pin of the 555 timer;

[0018] 55853629-1wherein the timing capacitor is configured such that when the force is applied to the force sensor, a time required to charge the timing capacitor decreases, thereby increasing the frequency of the output pulse train.

[0019] The readout method exploits a simple adaptation of a known readout method for improved resolution and range in force measurements as well as facilitating a much faster sampling time.

[0020] The readout method may be applied to a force sensor comprising one or more pairs of electrodes (i.e. one or more electrodes plus a common electrode). Thus, two pairs of electrodes may comprise two electrodes and one common electrode.

[0021] The readout method may be applied to a force sensor comprising multiple electrodes (i.e. multiple pairs of electrodes), and the method may further comprise operating a microcontroller and a switch to select one of the pairs of electrodes, at a time, for determining a force applied to the force sensor in a region of said one of the pairs of electrodes. Thus, allowing force determination for each region of the force sensor, thereby enabling a direction of the force to be determined.

[0022] The readout method may be applied to a force sensor comprising: at least one sensing element comprising: an electrode layer on which at least one electrode is interdigitated with a common electrode; a force-dependent active layer which comprises a forcesensitive material configured to change at least one electrical property when subjected to a force; and at least one spacer layer arranged to maintain a separation between the electrode layer and the force-dependent active layer, until a force is applied thereto. As such, the at least one pair of electrodes may comprise at least one pair of interdigitated electrodes.

[0023] According to a second aspect of the disclosure there is provided a readout circuit for a force sensor comprising at least one pair of electrodes, the circuit comprising:

[0024] a 555 timer configured to operate in astable mode;

[0025] a first node for coupling the pair of electrodes to a threshold pin and a trigger pin of the 555 timer;

[0026] a second node for coupling the pair of electrodes to an output pin of the 555 timer;

[0027] 55853629-1wherein the pair of electrodes, when coupled to the first node and the second node, is arranged in series with a timing capacitor of the 555 timer; and

[0028] wherein the timing capacitor is configured such that when a force is applied to the force sensor, a time required to charge the timing capacitor decreases, thereby increasing a frequency of an output pulse train at the output pin of the 555 timer.

[0029] The readout circuit may be configured for a force sensor comprising one or more pairs of electrodes, as described above.

[0030] The readout circuit may further comprise a microcontroller and a switch configured to select one of multiple electrodes (i.e. multiple pairs of electrodes), at a time, for determining a force applied to the force sensor in a region of said one of multiple pairs of electrodes.

[0031] According to a third aspect of the disclosure there is provided a force sensor comprising:

[0032] at least one sensing element comprising:

[0033] an electrode layer on which at least one electrode is interdigitated with a common electrode;

[0034] a force-dependent active layer which comprises a force-sensitive material configured to change at least one electrical property when subjected to a force; and at least one spacer layer arranged to maintain a separation between the electrode layer and the force-dependent active layer, until a force is applied thereto.

[0035] Thus, the proposed force sensor that can be made of thin material layers allowing it to be fully compliant and flexible but which includes a spacer layer to maintain a spacing between certain layers such that the bending of the sensor will not itself cause erroneous force readings and the sensor can therefore be robust, reliable and repeatable. Thus, the sensor can tolerate thousands to millions of cycles without failing, and can therefore effectively operate in an industrial setting. A consistent output also eliminates significant time which may otherwise be required to calibrate the output. In addition, the sensor has a low susceptibility to background noise and can be easily configured for different force sensing ranges depending on the application (i.e. by increasing or decreasing the active layer thickness).

[0036] 55853629-1Having only one electrode interdigitated with the common electrode (i.e. to form a single electrode pair) may only allow for normal (i.e. non-triaxial) force detection (i.e. no directional information can be obtained from the output). In some cases, the common electrode may constitute a signal ground electrode, rather than an electrode interdigitated with multiple electrodes.

[0037] In some cases, more than one electrode (i.e. at least two electrodes; at least three electrodes; or at least four electrodes) may be interdigitated with the common electrode. An advantage of including more than one electrode interdigitated with the common electrode is that direction information may be obtained, allowing for triaxial force measurement. In general, more detailed directional information may be obtained with a greater number of electrodes. However, in some cases, a trade-off needs to be met taking into account factors such as complexity, space and sampling rate, and therefore an optimal number of electrodes (i.e. electrode pairs), in some cases, may be three or four.

[0038] Advantageously, the sensor may be configured to be a thin, stretchable, conformable, nanocomposite-based tactile device which can measure force in three different axes (x, y and z). Embodiments of the invention may be compliant-enough to be wrapped around extremely small bending radii or soft robots, and nonetheless have a minimal effect on signal integrity or response allowing an accurate measurement of force magnitude as well as direction. Accordingly, the sensor performance is insensitive to bending. Flexible connections and electrical wiring may be integrated into the sensor via the layer and electrode structure, and a force transfer ‘bump’ layer may be provided above the forcedependent active layer to translate applied force into torque.

[0039] In addition, the sensor is amenable to mass-production using any of a number of existing facile volume manufacturing processes such as screen printing and roll-to-roll printing using customizable materials whilst maintaining high yield. Thus, it can be deployed in large-scale arrays such as may be required for an electronic skin application. Currently, no general-purpose electronic skin of this type exists. The sensor is highly integratable into different devices, products or garments and can be retro-fittable onto hard, soft, and stretchable surfaces. Moreover, environmentally friendly materials may be employed.

[0040] 55853629-1Each layer may be printed (e.g. by screen printing or inkjet printing) allowing it to be highly miniaturisable (at least to an ideal 1mm by 1mm tri-axis pixel density desired for a robotic end-effector). In some embodiments, the force-sensitive material may be constituted by a conductive ink deposited on force-dependent active layer.

[0041] All of the above advantages enable a commercially viable sensor suitable for ‘real-world’ industrial applications outside of a controlled lab environment or academia.

[0042] The common electrode may comprise one or more branches (e.g. three or four) extending from a central node, each branch comprising a series of comb-like tines which are interdigitated with corresponding comb-like tines of one of the at least one electrodes.

[0043] The one or more branches may be equidistantly spaced around the central node.

[0044] The comb-like tines of the one or more branches may be configured such that they extend from each respective one of the one or more branches in a common direction around the central node (e.g. extending in a clockwise or counterclockwise direction).

[0045] The at least one electrode and the common electrode may form a planar interdigitated structure and the force-sensitive material may have a footprint which substantially overlies the planar interdigitated structure.

[0046] The planar interdigitated structure and / or the footprint may have a substantially circular periphery.

[0047] The planar interdigitated structure and / or the footprint may have a substantially square, rectangular, oval or polygonal periphery.

[0048] The planar interdigitated structure and the footprint may have a substantially similar size and / or shape.

[0049] The force-sensitive material may be provided in discrete sections and each section may be arranged to overlie only one of the at least three electrodes. The sections may be defined by channels in the force-sensitive material. The sections may have any

[0050] 55853629-1appropriate size and / or shape. In some cases, the sections may constitute sectors of a circle.

[0051] The electrode layer may comprise three or four electrodes interdigitated with the common electrode. In which case, the at least one sensing element may effectively comprise three or four segments in which force can be sensed independently.

[0052] The force sensor may comprise a plurality of the at least one sensing elements. Advantageously, the plurality of the at least one sensing elements may form a sensing array.

[0053] The force-sensitive material may comprise at least one of: a quantum tunneling material, a piezoresistive material or a piezoelectric material.

[0054] The at least one spacer layer may comprise one or more of: a planar layer; a three-dimensional structure; and an adhesive.

[0055] According to a fourth aspect of the disclosure there is provided a method of fabricating a force sensor according to the third aspect comprising:

[0056] producing the force-dependent active layer comprising the force-sensitive material;

[0057] producing the electrode layer on which the at least one electrode is interdigitated with the common electrode;

[0058] arranging the force-dependent active layer on a registration jig;

[0059] providing a spacer layer on the force-dependent active layer while on the registration jig; and

[0060] providing the electrode layer on the spacer layer while on the registration jig.

[0061] The method of fabrication is quick and easy, whilst still being highly accurate due to the use of the registration jig and the top-down lamination process. Multiple sensors and / or sensing elements may be fabricated in a straight-forward batch process.

[0062] 55853629-1The method may further comprise producing the spacer layer by forming cut-outs in a planar substrate, the cut-outs being arranged to align with the force-sensitive material on the force-dependent active layer.

[0063] The method may further comprise forming registration features (e.g. slots, holes or projections) in each layer for registration in the registration jig, the registration features being arranged to ensure correct alignment of each layer in the registration jig and therefore ensuring correct alignment of each layer (and the features thereon) relative to each other layer (and the features thereon).

[0064] Each layer may be produced by one or more of: printing, screen-printing, roll-to-roll printing, inkjet printing, 3D printing, electrospinning, depositing, droplet dispensing, casting, coating, moulding, spinning or weaving.

[0065] The force sensor and / or readout method / circuit disclosed here may be employed in any suitable sensor devices such as, for example, those substantially as described in the applicants co-pending Greek application filed concurrently with the present application and entitled “Modular Tactile Sensor” (applicant reference number: PE962788GR), which is hereby incorporated by reference.

[0066] BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Some embodiments of the disclosure will now be described by way of example only and with reference to the accompanying drawings, in which:

[0068] Figure 1A shows a top view of an electrode layer and a spacer layer for a sensing element, in accordance with the present disclosure.

[0069] Figure 1 B shows an underside view of a force-dependent active layer comprising forcesensitive material for a sensing element, in accordance with the present disclosure.

[0070] Figure 2 shows a force sensor comprising a sensing element including the electrode layer and spacer layer of Figure 1A and the force-dependent active layer of Figure 1B.

[0071] 55853629-1Figure 3A shows a top view of a sensing portion of an electrode layer for a sensing element, in accordance with the present disclosure.

[0072] Figure 3B shows a top view of a spacer layer for use with the electrode layer of Figure 3A in a sensing element, in accordance with the present disclosure.

[0073] Figure 3C shows a top view of a force-sensitive area of a force-dependent active layer for use with the electrode layer of Figure 3A and the spacer layer of Figure 3B, in a sensing element, in accordance with the present disclosure.

[0074] Figure 4A shows a top view of a sensing portion of an electrode layer for a sensing element comprising three taxels, in accordance with the present disclosure.

[0075] Figure 4B shows a top view of a sensing element comprising the electrode layer of Figure 4A and a force-sensitive area of a force-dependent active layer having three distinct segments, in accordance with the present disclosure.

[0076] Figure 5 shows a known resistive readout circuit for an electrode of a force sensor.

[0077] Figure 6 shows a capacitive readout circuit for one pair of electrodes (i.e. one taxel) of a force sensor, in accordance with the present disclosure.

[0078] Figure 7 shows a capacitive readout circuit for four pairs of electrodes (i.e. four taxels) of a force sensor, in accordance with the present disclosure.

[0079] Figure 8 shows a signal diagram illustrating a timing voltage capacitor and output signals for the capacitive readout circuit of Figure 7.

[0080] Figure 9 shows a flow diagram of a readout method for a force sensor comprising at least one electrode, in accordance with the present disclosure.

[0081] Figure 10 shows a flow diagram of a readout method for a force sensor comprising multiple electrodes, in accordance with the present disclosure.

[0082] Figure 11A shows a top view of an electrode layer for a force sensor comprising 16 sensing elements, each comprising 1 taxel, in accordance with the present disclosure.

[0083] 55853629-1Figure 11B shows an underside view of the electrode layer of Figure 11A.

[0084] Figure 11C shows a top schematic view of the electrode layer of Figure 11 A, including a spacer layer, in accordance with the present disclosure.

[0085] Figure 11 D shows an underside view of a force-dependent active layer for use with the electrode layer and spacer layer of Figure 11 C, in a force sensor, in accordance with the present disclosure.

[0086] Figure 12A shows a top view of an electrode layer for a force sensor comprising 16 sensing elements, each comprising 4 taxels, in accordance with the present disclosure, showing only the sensing portions of electrode layer in accordance with Figure 3A.

[0087] Figure 12B shows an underside view of a force-dependent active layer for a force sensor comprising 16 sensing elements, in accordance with the present disclosure, showing segmented force-sensitive areas in accordance with Figure 3C.

[0088] Figure 13 shows a top view of a force sensor comprising 16 sensing elements, each comprising 3 taxels, in accordance with the present disclosure, where the sensing elements are in accordance with Figure 4B.

[0089] Figure 14 shows a batch produced set of 6 force sensors fabricated in accordance with Figures 15A to 151.

[0090] Figures 15A-15I show steps in a method of batch producing the set of 6 force sensors in accordance with Figure 14.

[0091] Figure 15J shows a registration jig and compression tool as used in the method of Figures 15A-15I.

[0092] Figure 16 shows a flow diagram of a method of fabricating a force sensor in accordance with the present disclosure.

[0093] 55853629-1Figure 17 shows a more detailed flow diagram of a method of fabricating a force sensor in accordance with the present disclosure.

[0094] Figure 18 shows a single taxel characteristic response under a normal force Fz.

[0095] Figure 19 illustrates mechanical cross-talk insensitivity for adjacent taxels when a single taxel is loaded with a 5N force.

[0096] Figure 20 shows a characteristic response of a triaxial sensor including 4 taxels (4 electrodes plus a common electrode) and decomposition of the Force into Fz, Fx and Fy components.

[0097] DETAILED DESCRIPTION OF THE EMBODIMENTS

[0098] Generally speaking, the disclosure provides a force sensor design and readout methodology capable of high accuracy normal or triaxial force sensing.

[0099] Figure 1A shows a top view of an electrode layer 100 and a spacer layer 110 for a sensing element 202 (shown in Figure 2), in accordance with the present disclosure. The electrode layer 100 comprises four electrodes 102 (also referred to as taxels), interdigitated with a common electrode 104. In other cases, one, two or three electrodes, instead of four, may be provided (along with at least one common electrode). In other cases, more than four electrodes may be provided. Each of the four electrodes 102 and the common electrode 104 are connected to a respective conductive trace 106 for connection to sensing electronics (not shown).

[0100] The common electrode 104 of Figure 1A comprises four branches extending from a central node. Each branch comprises a series of comb-like tines which are interdigitated with corresponding comb-like tines of one of the four electrodes 102. The branches are equidistantly spaced around the central node.

[0101] The electrodes 102 and the common electrode 104 form a planar interdigitated structure, which in this case, has a substantially square periphery. In other cases, the planar interdigitated structure may have a substantially circular, rectangular, oval or polygonal periphery.

[0102] 55853629-1A spacer layer 110 constituted by a circular ring spacer is provided around the planar interdigitated structure.

[0103] As shown in Figure 1B, a force-dependent active layer 120 comprises a force-sensitive material 122 for the sensing element 202, in accordance with the present disclosure. The force-sensitive material 122 in this case has a circular footprint configured to substantially overlie the planar interdigitated structure of Figure 1 A. In other cases, the force-sensitive material 122 may have a different size and / or shape of footprint. For example, the forcesensitive material 122 may have a substantially square, rectangular, oval or polygonal periphery. In some cases, the shape of the footprint may have a peripheral shape that substantially matches the shape of the periphery of the planar interdigitated structure.

[0104] The force-sensitive material 122 comprises at least one of: a quantum tunneling material, a piezoresistive material or a piezoelectric material and is configured to change at least one electrical property when subjected to a force.

[0105] Figure 2 shows a force sensor 200 comprising a sensing element 202 including the electrode layer 100 and spacer layer 110 of Figure 1A and the force-dependent active layer 120 of Figure 1B. The spacer layer 110 is arranged to maintain a separation between the electrode layer 100 and the force-dependent active layer 120, until a force is applied thereto, and to effectively enable the force-dependent active layer 120 to act as a diaphragm structure.

[0106] As shown in Figure 2, the force sensor 200 comprises a polymer encapsulation layer 204, which encapsulates the electrode layer 100, the spacer layer 110 and the forcedependent active layer 120. A connector 206 is provided to facilitate connection of the conductive traces 106 to sensing electronics (not shown).

[0107] As will be explained in more detail below, the sensing element 202 is configured for triaxial force measurement.

[0108] In some cases, a polymer force transfer layer is provided over the force-dependent active layer 120. The force transfer layer may include projecting structures (e.g. ‘bumps’) which may be arranged to project inwardly or outwardly over the force-sensitive material 122. The force transfer layer will be configured to transfer force applied thereto, to the force-

[0109] 55853629-1sensitive material 122, for detection by the electrode layer 100. Notably, the arrangement of the four electrodes 102 allows for independent detection of force in the four regions of the electrodes 102, thus, a force direction can be determined based on the relative sensor readings from each of the four electrodes 102.

[0110] Figure 3A shows a top view of a sensing portion of an electrode layer 100a for a sensing element, in accordance with the present disclosure. Similar to Figure 1A, the electrode layer 100a comprises four electrodes 102a, interdigitated with a common electrode 104a, which comprises four branches equidistantly spaced around a central node, with each branch comprising a series of comb-like tines which are interdigitated with corresponding comb-like tines of one of the four electrodes 102a. However, in this case, the electrodes 102a and the common electrode 104a form a planar interdigitated structure which has a substantially circular periphery. Note, the conductive traces extending to each electrode 102a and the common electrode 104 are omitted from Figure 3A for brevity.

[0111] Figure 3B shows a top view of a spacer layer 110a for use with the electrode layer 110a of Figure 3A in a sensing element, in accordance with the present disclosure. In this case, the spacer layer 110a is substantially planar and is provided with circular cut-outs 112 which are designed to be disposed over the planar interdigitated structure of the electrode layer 100a.

[0112] Figure 3C shows a top view of a force-sensitive area of a force-dependent active layer 120a for use with the electrode layer 100a of Figure 3A and the spacer layer 110a of Figure 3B, in a sensing element, in accordance with the present disclosure. Similar to Figure 1B, the force-dependent active layer 120a comprises a force-sensitive material 122a having a generally circular footprint configured to substantially overlie the planar interdigitated structure of Figure 3A. However, in this case, the force-sensitive material 122a is provided in four discrete sections 124a and each section 124a is arranged to overlie only one of the four electrodes 102a. This arrangement helps to eliminate (or at least reduce) cross-talk by ensuring that only the electrode 102a in the region of the section 124a that is deformed by an external force is electrically affected, thereby providing an accurate sensor readout.

[0113] In Figure 3C, the four sections 124a are formed by two orthogonal grooves 126a through the center of the circular force-sensitive material 122a.

[0114] 55853629-1Figure 4A shows a top view of a sensing portion of an electrode layer 100b for a sensing element 202b (shown in Figure 4B) comprising three electrodes (taxels) 102b, in accordance with the present disclosure. The electrode layer 100b forms substantially circular planar interdigitated structure that is similar to that of Figure 3A but, in the present case, the electrode layer 100b comprises three electrodes 102b, interdigitated with a common electrode 104b, which comprises three branches equidistantly spaced around a central node. In this case, each branch comprises a series of tines extending symmetrically about the length of each branch, with the length of the tines reducing in a distal direction like the veins on a leaf. Each of the three electrodes (taxels) 102b is arranged with a part-circular outer portion and a series of internal tines interdigitating with the tines of the common electrode 104b which extend between two adjacent branches.

[0115] Figure 4B shows a top view of the sensing element 202b comprising the electrode layer 100b of Figure 4A and a force-sensitive area 122b of a force-dependent active layer having three distinct sections 124b, in accordance with the present disclosure. Although not shown, the sensing element 202b will include a spacer layer similar to that of Figure 3B.

[0116] As explained above, the sensing element 202b is configured for triaxial force measurement.

[0117] Typically, the electrodes 102 of a force sensor 200 may serve as Force-Sensing-Resistors (FSRs), which are composed from conductive nanocomposites (e.g. having filler particles dispersed in an insulating elastomeric matrix), and which are interfaced with a circuit to convert resistance or capacitance changes in the FSR into measurable voltages. Such a circuit is generally known as an Analogue Front End (AFE).

[0118] The simplest method to obtain data from the FSR is by measuring the change in its resistance as force is applied.

[0119] Figure 5 shows a known resistive readout circuit 500 for an electrode of a force sensor, and which may be used with any or all of the electrodes 102, 102a, 102b of a force

[0120] 55853629-1sensor 200 similar to that of Figure 2 (although reference 102 will be used hereon in, for brevity).

[0121] In Figure 5, the electrode 102 is represented by a resistor Rs. The circuit 500 comprises a resistive divider and an operational amplifier 502 configured to operate in a noninverting configuration. A source voltage V+ is coupled to a first node of the electrode 102 (resistor Rs). A second node of the electrode 102 (resistor Rs) is coupled to a noninverting input of the operational amplifier 502 and a first node of a first resistor Ra. A second node of the first resistor Ra is coupled to a ground of the resistive readout circuit 500. Resistors Rs and Ra form a resistive divider.

[0122] In the example shown, the electrode 102 of the force sensor 200 is configured to decrease resistance when the force sensor 200 is subjected to a force through the forcedependent active layer 120. A voltage at the non-inverting input of the operational amplifier 502 (Vop+) is governed by equation (1):

[0123] y°p+=^7x y+ Equation (1) Therefore, in the circuit 500, the voltage at the non-inverting input of the operational amplifier 502 increases when the electrode 102 of the force sensor 200 is subjected to a force.

[0124] In the circuit 500, the source voltage V+ is coupled to a first node of a second resistor Rb. A second node of the second resistor Rb is coupled to an inverting input of the operational amplifier 502, a first node of a third resistor Rc, and a first node of a fourth resistor Rd. A second node of the third resistor Rc is coupled to the ground of the circuit 500. Resistors Rb and Rc form a resistive divider. A second node of the fourth resistor Rd is connected to an output of the operational amplifier 502.

[0125] In the example shown, a resistance of resistor Rb is equal to a resistance of resistor Rc. A positive rail of the operational amplifier 502 is connected to the source voltage V+ and a negative rail of the operational amplifier 502 is connected to the ground of the circuit 500.

[0126] The resistive readout circuit 500 is configured to scale an output voltage Vout of the operational amplifier 502 such that the output voltage is 0 V when no force is applied to

[0127] 55853629-1the electrode 102 and the output voltage is 3.3 V when the force applied to the electrode 102 is equal to the maximum force capacity of the force sensor 200.

[0128] In the example shown, the operational amplifier 502 has low output impedance and buffers the output of the operational amplifier 502.

[0129] Capacitance-based measurement offers several advantages in reducing sensitivity degradation, drift and hysteresis, which are key issues in achieving FSRs with high performance and accuracy.

[0130] Figure 6 shows a capacitive readout circuit 600 for one pair of electrodes (i.e. one taxel comprising one electrode 102 and the common electrode 104) of a force sensor 200, in accordance with the present disclosure. The circuit 600 has been simplified to include only the key features, for clarity. In some cases, the circuit 600 may include some or all of the features of the circuit 700 described below (e.g. including the Shotky diodes).

[0131] The circuit 600 comprises a 555 timer 602 configured to operate in astable mode and the electrode pair comprising, in this case, one electrode 102 and the common electrode 104 of the force sensor 200 is represented by a variable capacitor Cs and a variable resistor Rs in parallel.

[0132] The 555 timer 602 is an integrated circuit (IC) that operates based on charging and discharging a capacitor through a timing network to produce a stable square wave output. Its operation revolves around two internal voltage thresholds (1Zs Vcc and % Vcc) monitored by comparators. These are set up internally via 3x 5kQ resistors, hence the term 555. As the capacitor charges and discharges, its voltage increases / decreases until it hits % or1Zs Vcc, respectively, at which point the comparators trigger an internal flip-flop, toggling the output between high and low states. The charge and discharge times of the capacitor, determined by external resistors and capacitors, set the frequency or pulse width of the output signal, therefore a circuit output (Do) is a square signal with frequency f that changes accordingly with the applied force. Larger Rcharge or Recharge resistors or a larger RC timer capacitance will reduce the 555 output frequency.

[0133] A 555 timer 602 can generally work in one of two modes - Monostable and Astable. In monostable mode, the 555 timer 602 generates a single pulse of a defined duration when triggered, while in astable mode, it continuously oscillates to produce a periodic square wave as used herein.

[0134] 55853629-1When integrated with a Force Sensing Resistor (FSR) composite transducer, the 555 timer 602 uses the FSR as a variable capacitor within its timing network. As forces are applied to the FSR, its capacitance increases, directly impacting the charge and discharge times of the capacitor. This leads to variations in the frequency or pulse width of the 555 timer’s output signal (conventionally to a decrease in frequency). By measuring these output changes, the applied force can be quantified, enabling the circuit to function as a force-sensing system.

[0135] In the circuit 600, a source voltage Vcc is coupled to a power supply pin (pin 8) and a reset pin (pin 4) of the 555 timer 602. The source voltage Vcc is also coupled to a first node of a first timing resistor R1. A second node of the first timing resistor R1 and a first node of a second timing resistor R2 are coupled to a discharge pin (pin 7) of the 555 timer 602.

[0136] In the circuit 600, the following nodes and pins are coupled together: a second node of the second timing resistor R2; a trigger pin (pin 2) and a threshold pin (pin 6) of the 555 timer 602; a first node of a timing capacitor Ct; and a first node of the electrode 102 of the force sensor 200. A second node of the timing capacitor Ct is connected to a ground of the capacitive readout circuit 600.

[0137] In the circuit 600, voltage control pin (pin 5) of the 555 timer 602 is connected to the ground of the capacitive readout circuit 600 via a capacitor C2.

[0138] In the circuit 600, the 555 timer 602 generates two reference voltages internally using a resistive divider and two comparators. The timing capacitor Ct charges up to a first reference voltage 2 / 3 Vcc through the timing resistors R1 and R2. When a voltage on the reset pin (pin 4) increases above the first reference voltage, an output pin (pin 3) is configured to transition the output signal Do from a digital high to a digital low voltage.

[0139] In the circuit 600, the timing capacitor Ct discharges through timing resistor R2. When a voltage on the trigger pin (pin 2) drops below a first reference voltage 1 / 3 Vcc, the output pin (pin 3) is configured to transition the output signal Do from a digital low to a digital high voltage.

[0140] 55853629-1The duty cycle of the output signal Do is governed by equation (2) where TON is the time the output signal Do is a digital high and TOFF is the time the output signal Do is a digital low:

[0141] Duty cycle = — — — =fil+fi2% Equation (2)

[0142]

[0143] TON+TOFF R1+2XR2

[0144] The frequency f of the output signal Do is governed by equation (3), where T is the time period of the output signal Do:

[0145] f

[0146] 7= - = - - — Equation (3)

[0147] T In (2)x(R1+2R2)xCtv'

[0148] In the circuit 600, one end of the taxel (electrode pair 102, 104) is connected to the threshold pin (pin 6), and the other end of the taxel (electrode pair 102, 104) is connected to the output pin (pin 3) of the 555 timer 602 forming a feedback loop. The electrode 102 (i.e. taxel) of the force sensor 200 is arranged in series with the timing capacitor Ct. The capacitance Cs of the electrode 102 skews the output signal Do on the output pin (pin 3) of the 555 timer 602.

[0149] In this case the charge / discharge path is now altered, so the timing capacitor Ct now charges not just through R1 and R2 from Vcc, but it also gets “kicked” by transitions at the output pin (pin 3) via the feedback capacitor Cs (i.e. sensing taxel).

[0150] The taxel capacitance acts like a capacitive charge pump into the RC node. Each output transition injects or removes charge to / from the main RC timing capacitor Ct through the feedback capacitor Cs. The amount of charge transferred depends on the capacitance value the feedback capacitor Cs (i.e. force applied to the taxel).

[0151] When the 555 timer output toggles, the feedback capacitor (i.e. sensing taxel) transfers charge to or from the main timing capacitor Ct (depending on output state high / low). This directly modifies the effective charge / discharge slope of the timing capacitor Ct. Since frequency is tied to how quickly the capacitor crosses the threshold voltages (1 / 3 Vcc and 2 / 3 Vcc), small variations in that slope cause noticeable changes in frequency. When the output toggles HIGH, current flows through the feedback capacitor Cs into the timing capacitor Ct, giving it a sudden upward nudge in voltage. When the output goes HIGH, charge is pushed into the fixed timing capacitor Ct, causing a sudden increase in voltage at the Threshold pin (pin 6). When the output goes LOW, charge is pulled out from the timing capacitor Ct.

[0152] 55853629-1As such, increasing the capacitance at the sensor Cs increases the amount of charge injected per transition, speeding up the voltage rise and fall on the timing capacitor Ct. So, the oscillation frequency increases with feedback capacitance Cs. Smaller values of taxel capacitance Cs reduce the “kick”, leading to the charge / discharge of the timing capacitor Ct taking longer, hence reducing the frequency of the output.

[0153] In the circuit 600, the capacitance Cs is configured to increase when the force sensor 200 is subjected to a force. Therefore, as the applied force increases, the capacitance Cs increases and the output signal Do becomes increasingly skewed.

[0154] In the circuit 600, the timing capacitor Ct is arranged in series with the electrode 102, which is coupled to the output pin (pin 3). When the output signal Do transitions from a digital low to a digital high, a voltage across the timing capacitor Ct increases sharply. While the output signal Do is high, the capacitor Ct charges. As the capacitance of the electrode 102 increases, the magnitude of the sharp voltage increase becomes greater. Therefore, the timing capacitor Ct charges up to the first reference voltage 2 / 3 Vcc in a shorter period of time than it would if the timing capacitor Ct was not coupled to the electrode 102 and the output pin (pin 3). As the capacitance of the electrode 102 increases, the time required to charge the timing capacitor Ct decreases, which increases the frequency of the output signal Do pulse train at the output pin (pin 3).

[0155] Figure 7 shows a capacitive readout circuit 700 for four pairs of electrodes 102, 104 (i.e. four taxels formed using four electrodes 102 and one common electrode 104) of a force sensor 200, in accordance with the present disclosure. Figure 7 is similar to Figure 6 but illustrates four pairs of electrodes 102, 104 which are configured to be read individually when selected by a 4-channel analogue multiplexer 702, and an output Doi which is digitised from Do using a forward-biased diode and transistor switch 704. In other cases, the circuit 700 can be configured for switching between any number of pairs of electrodes 102, 104 (i.e. taxels) using an analogue multiplexer 702.

[0156] In the circuit 700, the 555 timer 602 is configured in a modified astable configuration. A positive node of a first diode D1 is coupled to the discharge pin (pin 7), and a negative node of the first diode D1 is coupled to the trigger pin (pin 2) and threshold pin (pin 6). D1 is in series with the first timing resistor R1, ensuring that the timing capacitor Ct

[0157] 55853629-1cannot discharge through R1. A second diode D2 is in series with the second timing resistor R2, ensuring that the timing capacitor Ct cannot charge through R2.

[0158] In the circuit 700, a period in which the output signal is high (TON) is determined by a resistance of timing resistor R1. A period in which the output signal is low (TOFF) is determined by a resistance of timing resistor R2.

[0159] The duty cycle of the output signal Dois governed by equation (4):

[0160] Duty cycle = — — — =R1% Equation (4)

[0161]

[0162] TON+TOFF R±+R2

[0163] In some examples, D1 and D2 may be Schottky diodes with a lower voltage drop than 0.7 V.

[0164] In the circuit 700, the 4-channel analogue multiplexer 702 is coupled to the trigger pin (pin 2), the threshold pin (pin 6), and the timing capacitor Ct. The 4-channel analogue multiplexer 702 is controlled by a microcontroller unit (MCU) 706 which is configured to select one of the four pairs of electrodes at a time. The first node of each electrode pair is coupled to the 4-channel analogue multiplexer 702, and the second node of each electrode pair is coupled together at a common terminal (via the common electrode 104). The common terminal is coupled to the output pin (pin 3) of the 555 timer 602.

[0165] In the circuit 700, the microcontroller unit 706 is supplied by a voltage regulator 708 coupled to the voltage source of the circuit Vcc.

[0166] In other examples, the analogue multiplexer 702 may be used with at least three electrodes or at least three pairs of electrodes. The number of electrodes coupled to the analogue multiplexer 702 may be at least three. In other examples, the number of electrodes may be greater than three.

[0167] In other examples, a single electrode pair may be coupled to the trigger pin (pin 2), the threshold pin (pin 6), and the timing capacitor Ct.

[0168] In the circuit 700, Do is coupled to a gate of the transistor switch 704 via a forward-biased diode D3. A source of the transistor switch 704 is coupled to a digital high voltage source. A drain of the transistor switch 704 is coupled to a microcontroller digital input pin. The

[0169] 55853629-1transistor switch 704 is configured to switch on when a voltage at the gate of the transistor switch 704 is sufficiently high and allow the digital high voltage at the source to pass to the drain. The drain of the transistor switch 704 is coupled to a pull-down resistor Rp. The diode D3 and transistor switch 704 are configured to generate a very low noise digital output Doi with very low pulse error.

[0170] In some examples, the microcontroller digital input pin is a pin on the microcontroller unit 706. In other examples, the microcontroller digital input pin is a pin on a second microcontroller unit (not shown). In other examples, the output pin (pin 3) of the 555 timer 602 is coupled directly to a microcontroller digital input pin.

[0171] In the circuit 700, the microcontroller (or any other logical IC, e.g. Field Programmable Gate Array (FPGA) etc.) is configured to count pulses in the output signal Doi within a fixed gate interval using an interrupt function. In the example shown, the microcontroller is configured to determine the frequency of pulses by dividing the number of pulses by the fixed gate interval.

[0172] In other examples, the output pin (pin 3) of the 555 timer 602 is coupled directly to a microcontroller analogue input pin, wherein the microcontroller is configured to count pulses in the output signal Doi when the analogue input voltage exceeds a threshold. In other examples, the microcontroller is configured to determine the frequency of pulses by dividing the number of pulses by a time period.

[0173] Figure 8 shows a signal diagram illustrating a timing capacitor voltage Vet and output signals Do and Doi for the capacitive readout circuit 700 of Figure 7. The solid lines represent the signals when a greater force is applied to the force sensor and the electrode (taxel) capacitance increases. The broken lines represent the signals when a smaller force is applied to the force sensor and the electrode (taxel) capacitance decreases.

[0174] In the circuits of Figures 6 and 7 the charge / discharge path is altered from a conventional circuit and so the timing capacitor Ct now charges not just through R1 and R2 from Vcc, but it also gets “kicked” by transitions at the output pin (pin 3) via the feedback loop.

[0175] The taxel capacitance acts like a capacitive charge pump into the RC node. Each output transition injects or removes charge to / from the main RC timing capacitor Ct through the

[0176] 55853629-1feedback capacitor (i.e. sensing taxel). The amount of charge transferred depends on the capacitance value (i.e. force applied to the taxel).

[0177] When the 555 timer output toggles, the feedback capacitor (i.e. the sensing taxel) transfers charge to or from the main timing capacitor (depending on output state high / low). This directly modifies the effective charge / discharge slope of the timing cap. Since frequency is tied to how quickly the capacitor crosses the threshold voltages (1 / 3 Vcc and 2 / 3 Vcc), small variations in that slope cause noticeable changes in frequency. When output toggles HIGH, current flows through the feedback capacitor into the timing capacitor, giving it a sudden upward nudge in voltage. When the output goes HIGH, charge is pushed into the timing capacitor, causing a sudden increase in voltage at the Threshold pin. When the output goes LOW, charge is pulled out from the timing capacitor, causing a sudden drop in voltage at the Threshold pin.

[0178] The timing capacitor Ct therefore dictates when the output toggles. It is believed that the magnitude of the “kick” increases as the taxel capacitance increases. When the timing capacitor Ct charges up to 2 / 3 Vcc the output Dofalls low. The timing capacitor Ct charges faster than it would otherwise because of the “kick”. As the magnitude of the “kick” increases, the faster the output Dois toggled and the more pulses are counted. The capacitance of the timing capacitor is not changing, therefore the slope does not change, just the “kick”.

[0179] The result is that a greater force on the taxel, increases the taxel capacitance, causing the timing capacitor Ct to charge faster, and therefore the output Dois toggled faster and the frequency of the output pulse is increased. As such, a force applied to the force sensor can be determined based on a frequency of the output pulse train.

[0180] Increasing the capacitance increases the amount of charge injected per transition, speeding up the voltage rise and fall on the timing capacitor. So, the oscillation frequency increases with feedback capacitance. Smaller values of taxel capacitance reduce the “kick”, leading to the charge / discharge of the timing capacitor taking longer, hence reducing the frequency of the output. This setup behaves like a charge injection mechanism, where the charge AQ transferred on each output transition is in accordance with equation (5):

[0181]

[0182] Equation (5)

[0183] 55853629-1In this case, AVout is approximately a full swing (OV - Vcc), so it's large and constant. Thus, even small changes in Csensor lead to proportionally large changes in injected charge, which means a sensitive mapping between capacitance and frequency. This assumes that the taxel resistance has a negligible effect on the output frequency.

[0184] Running a few different taxel capacitance (C4) values through a model with fixed charge / discharge resistors of 10k yields the following output frequencies:

[0185] C4 = 1pF C4 = 10 pF C4 = 50 pF C4 = 100 pF C4 = 200 pF

[0186] 57.38 kHz 57.89 kHz 60.6 kHz 65.1 kHz 78.9kHz

[0187]

[0188] An advantage of the readout method having increasing frequencies with increasing capacitance under a force, is in that pulse counting in digital controller IC units, such as MCUs (microcontrollers) or FPGAs (Field Programmable Gate Arrays), is performed by counting the number of pulse cycles that occur during a fixed “gate interval” time via an interrupt function. This works well for relatively high frequencies, because many cycles are likely to be counted during a set gate time interval. At lower frequencies, very few cycles are counted, giving limited resolution. In other words, the gate time interval time can be set to a minimal value, meaning also a very fast sampling time, without loss of resolution and with increased range. For example, the gate time interval time can be set such that 1 count of pulses are measured before any force is applied (i.e. F = 0 N) while an increasing number of pulses will be counted as a function of applied force (since the frequency of the pulses increases with applied load in this case).

[0189] In contrast, in a conventional grounded configuration, where one end of the taxel is connected to the threshold pin of the 555 timer, and the other end is grounded, the taxel’s capacitance is effectively in parallel with the fixed capacitor, forming part of the RC network. Changes in pressure on the taxel lead to changes in the lump capacitance forming the 'C part of the RC network and result in a decrease of the pulse frequency. Note that the fixed capacitor value used is typically 1nF. Thin-film (~15micron thick) composite inks (FSR) may be used as the force transducing elements (taxels) and their capacitance is typically in the order of the pF range, so only ~1% of the total capacitance

[0190] 55853629-1in the RC network. This results in very small changes in the RC timing network, which limits the range over which the 555 timer output swings for zero to full pressure applications on the sensing taxel, which fundamentally limits the sensing range significantly.

[0191] Figure 9 shows a flow diagram of a readout method 800 for a force sensor comprising at least one electrode, in accordance with the present disclosure. The method 800 comprises a step 802 of coupling a first node of a taxel to a threshold pin and a trigger pin of an astable mode 555 timer; a step 804 of coupling a second node of the taxel to an output pin of the astable mode 555 timer; a step 806 of determining the frequency of the output pulse train at the output pin of the 555 timer; and a step 808 of determining a force applied to the taxel based on the frequency of the output pulse train.

[0192] Figure 10 shows a flow diagram of a readout method 900 for a force sensor comprising multiple electrodes, in accordance with the present disclosure. The method 900 comprises a step 902 of coupling the output pin of the astable mode 555 timer to a forward-biased diode; a step 904 of coupling a source of an output transistor switch to a digital high voltage; a step 906 of coupling the forward-biased diode to a gate of the output transistor switch; a step 908 of coupling a drain of the output transistor switch to a microcontroller (MCU) and a pull-down resistor; a step 910 of counting the pulses in a fixed gate interval using the microcontroller configured to perform an interrupt function; and a step 912 of determining the frequency of pulses by dividing the number of pulses by the fixed gate interval. Conveniently the microcontroller may be coupled to the microcontroller 706, which switches between the multiple electrodes, so that the frequency of pulses in each gate interval may be used to determine the force applied to each of the electrodes in turn. Thus, the magnitude and direction of a force applied to the force sensor may be determined.

[0193] Figure 11A shows a top view of an electrode layer 100’ for a force sensor 1000 comprising 16 sensing elements 202, each comprising 1 taxel (i.e. one pair of electrodes configured for normal (non-directional) force sensing), in accordance with the present disclosure. In this case, the electrodes 102, 104 for each sensing element 202 may be screen printed silver traces provided on a thin polyimide film or developed directly using thin-film polyimide Flexible Printed Circuit (FPC) technology that allows the inclusion of vias and multiple layers (e.g. 4, 6, 8 etc., for higher levels of footprint integration and

[0194] 55853629-1versatility). Notably, screen printing does not allow vias. In which case only overpasses can be introduced (i.e. with a dielectric layer selectively printed on the traces that need to be overpassed).

[0195] Figure 11 B shows an underside view of the electrode layer 100’ of Figure 11 A, showing traces and vias for the electrodes 102, 104 on the opposite side of the film. The electrodes 102, 104 may be configured in accordance with Figures 3A or 4A for the case where 3 or 4 taxels are included per sensing element 202.

[0196] Figure 11C shows a top schematic view of the electrode layer 100’ of Figure 11 A, including a spacer layer 110’, in accordance with the present disclosure. The spacer layer 110’ includes a circular ring spacer around each planar interdigitated structure of the electrodes 102, 104 and a peripheral wall around the outline of the electrode layer 100’.

[0197] Figure 11 D shows an underside view of a force-dependent active layer 120’ for use with the electrode layer 100’ and spacer layer 110’ of Figure 11 C, in a force sensor 1000, in accordance with the present disclosure. The force-dependent active layer 120’ comprises 16 circular force-sensitive material 122 spots in accordance with Figure 1B. The force-sensitive material 122 spots may comprise floating quantum tunnelling elements which are screen printed on another thin polyimide film.

[0198] In Figures 11A-11 D, the force sensor 1000 is configured with shoulder cut-outs to enable the films to conform into the shape of the face of a fingertip tactile sensor device, without the layers bulging at any point as that would lead to undesired precompression and mechanical instability of the sensing elements 202.

[0199] Notably, the size, shape and arrangement of the sensing elements 202 and / or force sensor 1000 may be adapted to suit different applications. For example, the force sensor 1000 may be formatted for use in the palm of a robot hand or in another robot limb.

[0200] In the examples described herein, the force sensor 1000 comprises an electrode layer 100 provided on a polyimide FPC, a spacer layer 110 (which may comprise an adhesion or lamination substrate), and a force-dependent active layer 120 which comprises a screen-printed floating transducer ink layer (forming the force-sensitive material 122) on

[0201] 55853629-1a polyimide film. In some cases, the polyimide film may be replaced by another substrate comprising a polymer, polyester or elastomer (e.g. polyethylene naphthalate (PEN) or polyethylene terephthalate (PET)). In some cases, the force-dependent active layer may comprise a thin film (e.g. having a thickness of approximately 1 to 30 microns). In some cases, the force-dependent active layer may comprise a force-sensitive material deposited on a thin film.

[0202] Figure 12A shows a top view of an electrode layer 100a’ for a force sensor comprising 16 sensing elements, in accordance with the present disclosure, showing only the 16 sensing portions of electrode layer 100a in accordance with Figure 3A.

[0203] Figure 12B shows an underside view of a force-dependent active layer 120a’ for a force sensor comprising 16 sensing elements, in accordance with the present disclosure, showing the 16 segmented force-sensitive areas 120a in accordance with Figure 3C.

[0204] Figure 13 shows a top view of a force sensor comprising 16 sensing elements, in accordance with the present disclosure, where the sensing elements 202b are in accordance with Figure 4B.

[0205] Figure 14 shows a batch produced set of 6 force sensors 1000 fabricated in accordance with Figures 15A to 151. The 6 force sensors 1000 are fabricated to form a composite thin polymer film panel 1111 having registration features in the form of holes 1112 for alignment of the respective layers forming the force sensors 1000. A variety of materials may be used for each of the layers, the electrodes and the force-sensitive materials in the force sensors 1000, for example, to suit an application, range and sensitivity.

[0206] Figures 15A-15I show steps in a method of batch producing the set of 6 force sensors 1000 in accordance with Figure 14. It should be noted that the panel 1111 size (and hence the number of force sensors 1000 per panel 1111) is only an example, and can be smaller or larger, e.g. fitting the size of A4 paper or a maximum panel size that flexible printed circuit (FPC) manufacturers can allow.

[0207] As shown in Figures 15A-C, a polymer sheet 1120 comprising an arrangement of 16 force-sensitive material 122 spots for each of the 6 force sensors is provided on a registration jig 1002, using holes (not shown) in the polymer sheet 1120 to slot the

[0208] 55853629-1polymer sheet 1120 over a series of six circular pins 1004 to ensure that the forcesensitive material 122 spots are aligned on the registration jig 1002. A planar compression plate 1006, which also has holes (not shown) to allow the planar compression plate 1006 to slot over the circular pins 1004, is then used to compress the polymer sheet 1120 on the registration jig 1002 to ensure alignment.

[0209] As shown in Figures 15D-F, a pre-formed substantially planar adhesive spacer layer 1120, comprising circular cut-outs which are arranged to be disposed over the forcesensitive material 122 spots of the polymer sheet 1120, and which also comprises registration holes, is provided on the registration jig 1002 and compressed using the compression plate 1006 to bond the spacer layer 1120 to the polymer sheet 1120.

[0210] As shown in Figures 15G-I, a pre-formed electrode layer 1100 comprising the electrode layout for the 16 sensing elements in each of the 6 force sensors 1000 is slotted on the registration jig 1002 for alignment with the spacer layer 1120 and polymer sheet 1120 and then compressed using the compression plate 1006 to bond the electrode layer 1100 to the spacer layer 1120 and polymer sheet 1120 below.

[0211] Figure 15J shows the registration jig 1002 and compression tool 1006 as used in the method of Figures 15A-15I.

[0212] Advantageously, the registration jig 1002 ensures alignment and uniformity of the assembly and can be scaled to suit various applications and various numbers and sizes of force sensors.

[0213] Once assembled in Figures 15A-I, each force sensor 1000 may be released from the sheet 1111, for example, by cutting, stamping, laser cutting and / or using a series of breakout tabs or perforations around each force sensor 1000.

[0214] The amount of taxels and their density in the example described is 16, but can be increased to any quantity, as more complex tracing can be achieved with use of vias and a multi-layer FPC stack-up (e.g. 4, 6 or 8 layers may be stacked similarly to a PCB). The readout method and circuit described herein may accommodate as many as 256 elements or more per force sensor 1000 and can be expanded as necessary. In operation, the capacitance and / or resistance of each of the sensing elements in the force

[0215] 55853629-1sensor 1000 may be read in individually or in combination to maximize the accuracy and / or performance of the device to suit different applications.

[0216] Figure 16 shows a flow diagram of a method 2000 of fabricating a force sensor 1000 in accordance with the present disclosure. The method 2000 comprises a step 2002 of producing the force-dependent active layer comprising the force-sensitive material; a step 2004 of producing the electrode layer on which the at least three electrodes are interdigitated with the common electrode; a step 2006 of arranging the force-dependent active layer on a registration jig; a step 2008 of providing a spacer layer on the forcedependent active layer while on the registration jig; and a step 2010 of providing the electrode layer on the spacer layer while on the registration jig. The method may further comprise a step of applying heat and / or pressure to the layers on the registration jig to laminate the sheets together.

[0217] The method may further comprise a step of producing the spacer layer by forming cutouts in a planar substrate, the cut-outs being arranged to align with the force-sensitive material on the force-dependent active layer.

[0218] Figure 17 shows a more detailed flow diagram of a particular method 3000 of fabricating a force sensor 1000 in accordance with the present disclosure. The method 3000 comprises a step 3002 of laser cutting registration holes in polymer sheets and spacer sheet to align with pins on registration jig; a step 3004 of screen printing transducer ink onto one of the polymer sheets; a step 3006 of screen printing electrodes and traces onto another of the polymer sheets; an optional step 3008 of inserting a spacer sheet between the two polymer sheets; a step 3010 of aligning and layering each of the polymer sheets (and spacer sheet) on the registration jig; a step 3012 of applying heat and pressure to laminate the sheets together; and a step 3014 of laser cutting breakout tabs to release the force sensors 1000 from the flexible sheet assembly.

[0219] Thus, the force sensor design enables roll-to-roll assembly during manufacture, meaning sensors based on the proposed design are well-suited for scaled up production. This is due to the fact that all of the driving electrodes / terminals, i.e. the planar interdigitated comb-like terminals (e.g. 1 common terminal and 3 or 4 petals for normal and shear force Fx,Fy,Fz decomposition), as well as the spacer layer and force-sensitive materials can be developed in separate individual films and assembled / laminated bottom-up (or top down) with no folding required.

[0220] 55853629-1The following describes graphs of the taxel characterisation response for both normal and triaxial variants of the force sensor described herein. The sensor measurements were taken with force sensors similar to those shown in Figures 11 A-11 D (for a normal sensor variant) and Figures 12A-B (for a triaxial sensor variant), when integrated into a fingertip sensor device (i.e. having a curved geometrically complex structure).

[0221] Figure 18 shows a single taxel characteristic response under a normal force Fz.

[0222] Figure 19 illustrates mechanical cross-talk insensitivity for adjacent taxels when any single taxel of the array is loaded with a 5N force.

[0223] Figure 20 shows a characteristic response of a triaxial sensor including 4 taxels (4 electrodes plus a common electrode) and decomposition of the Force into Fz, Fx and Fy components. Fz_trx, Fx_trx and Fy_trx represent the raw sensor response in thousands of cycles.

[0224] The graphs shown in Figures 18 to 20 demonstrate an increasing output frequency as force increases; no sensitivity degradation; negation of hysterisis (i.e. loading and deloading convergence); negation of static drift; mechanical insensitivity (no mechanical cross-talk between taxels); and force decomposition into Fx, Fy, Fz components.

[0225] The force sensor may be assembled layer by layer, as described, before being encapsulated in mould with a flexible / conformable rubbery material such as polyurethane (Pll) with a force transfer layer (e.g. having a “bump” structure) optionally provided on the force-dependent active layer. An applied 3D force on the force transfer layer will be translated to the force-sensitive material layer, which then deforms onto (or in the region of) the electrode layer. The decomposed force vector F (Fx, Fy, Fz) may be provided by the 4 interdigitated petal sections individually (+Fx, -Fx, +Fy, -Fy) and in unison (Fz, corresponding to the combined product). The force-dependant capacitive response of the sensing sections of each of the four petals may be read sequentially for each petal via a “shunt” capacitive measurement based on the improved capacitive readout method described above. The readout may be composed collectively for each petal section based on the accumulated parallel lumped capacitances formed between

[0226] 55853629-1the adjacent fingers of the interdigitated electrodes and the force-sensitive material bridging them.

[0227] The interdigitated planar electrodes provide full coverage during contact with the floating force-sensitive material which was found to provide improved translation especially of the shear forces (Fx.Fz) and a smoother response, e.g. very smooth translation of the force vector orientation and amplitude across all angles. In addition, this topology allows tracing all the output lines to terminate on a single output (single layer) which reduces wiring significantly; a key consideration in robotics applications.

[0228] In the present case, counter-intuitively, the sensor, instead of being placed in parallel to the output capacitor of the astable 555 timer, is in turn placed in series to one of the terminals of the timing capacitor and connected directly to the output Do of the 555 timer. This in turns modulates directly the square signal output of the 555 timer, skewing the shape of the square signal output from the 555 timer depending on the capacitance change of the force-sensitive material under force. The now skewed output Do is in turn terminated to an output transistor switch, triggered on-off by the skewed signal waveform, which conditions the output signal to very clean square waves Doi, i.e. very low noise digital output (i.e. +- 1 pulse error), effectively as low as the expected error of the 555 timer output. The full topology for driving a triaxial sensor with three, four or more sensing elements (petals) and a common terminal, as shown in Fig. 7, includes an additional, in this case, 4-channel analogue multiplexer (MUX), which is used to select sequentially between the 4 petals of the sensor, and a Microcontroller Unit (MCU), which controls the selection of the petals, provides the digital high level logic signal to the terminator switch and performs the pulse counting. The latter can be replaced also with other digital controller units such as an FPGA for improved performance (e.g. parallelisation of the readout since the output is effectively a digital signal).

[0229] The improved readout circuit topology was found to result in an increasing frequency between the output pulses with applied force loads, which is advantageous as described above, and enables for both existing “through” configuration triaxial sensors and the new “shunt” triaxial sensors described herein to be driven against a very large dynamic range (0 to +50 N) which was not possible with the previously established readout, as well as leveraging in general the improved performance provided by a capacitive readout as described.

[0230] 55853629-1The pulse frequency increase of the new topology against the frequency decrease of the established topology provides improved resolution and range in force measurement as well as a very fast sampling time.

[0231] Embodiments of the present disclosure can be employed in many different applications including, for example, in assembly and warehouse fulfilment (e.g. to measure grasp stability and detect object slippage to enhance pick rates and boost warehouse or assembly line productivity); in glove box applications (enabling teleoperated robots with touch sensitivity for precise, dexterous tasks in hazardous or remote environments, ensuring safe handling of chemicals, biological, or nuclear materials); in training humanoid robots (e.g. training a humanoid robot to perform dexterous real-world tasks; using the sense of touch to handle scenarios where vision is unreliable or occluded; creating datasets and Al models to advance the next generation of robots for workplaces and households); material evaluation (building a database of materials and object properties, using touch to identify textures, shapes, softness, and more, enabling virtual exploration and allowing customers to feel products before purchase); and robot safety (developing robots that regulate force during object manipulation, enhancing safety with precise touch and compliant control) as well as others.

[0232] The skilled person will understand that in the preceding description and appended claims, positional terms such as ‘above’, ‘along’, ‘side’, etc. are made with reference to conceptual illustrations, such as those shown in the appended drawings. These terms are used for ease of reference but are not intended to be of limiting nature. These terms are therefore to be understood as referring to an object when in an orientation as shown in the accompanying drawings.

[0233] Although the disclosure has been described in terms of preferred embodiments as set forth above, it should be understood that these embodiments are illustrative only and that the claims are not limited to those embodiments. Those skilled in the art will be able to make modifications and alternatives in view of the disclosure which are contemplated as falling within the scope of the appended claims. Each feature disclosed or illustrated in the present specification may be incorporated in any embodiments, whether alone or in any appropriate combination with any other feature disclosed or illustrated herein.

[0234] 55853629-1

Claims

CLAIMS1. A readout method for a force sensor comprising at least one pair of electrodes, the method comprising:coupling a first node of the pair of electrodes to a threshold pin and a trigger pin of a 555 timer configured to operate in astable mode;coupling a second node of the pair of electrodes to an output pin of the 555 timer;wherein the pair of electrodes is arranged in series with a timing capacitor of the 555 timer; anddetermining a force applied to the force sensor based on a frequency of an output pulse train at the output pin of the 555 timer;wherein the timing capacitor is configured such that when the force sensor is subjected to force a time required to charge the timing capacitor decreases, thereby increasing the frequency of the output pulse train.

2. The readout method of claim 1 when applied to a force sensor comprising more than one pair of electrodes.

3. The readout method of claim 2 when applied to a force sensor comprising multiple pairs of electrodes, the method further comprising operating a microcontroller and a switch to select one of the pairs of electrodes at a time, for determining a force applied to the force sensor in a region of said one of the pairs of electrodes.

4. The readout method of any of claims 1 to 3 when applied to a force sensor comprising:at least one sensing element comprising:an electrode layer on which at least one electrode is interdigitated with a common electrode;a force-dependent active layer which comprises a force-sensitive material configured to change at least one electrical property when subjected to a force; and55853631-1at least one spacer layer arranged to maintain a separation between the electrode layer and the force-dependent active layer, until a force is applied thereto.

5. A readout circuit for a force sensor comprising at least one pair of electrodes, the circuit comprising:a 555 timer configured to operate in astable mode;a first node for coupling the pair of electrodes to a threshold pin and a trigger pin of the 555 timer;a second node for coupling the pair of electrodes to an output pin of the 555 timer;wherein the pair of electrodes, when coupled to the first node and the second node, is arranged in series with a timing capacitor of the 555 timer; and wherein the timing capacitor is configured such that when a force is applied to the force sensor, a time required to charge the timing capacitor decreases, thereby increasing a frequency of an output pulse train at the output pin of the 555 timer.

6. The readout circuit of claim 5 comprising a microcontroller and a switch configured to select one of multiple pairs of electrodes, at a time, for determining a force applied to the force sensor in a region of said one of multiple pairs of electrodes.

7. A force sensor comprising:at least one sensing element comprising:an electrode layer on which at least one electrode is interdigitated with a common electrode;a force-dependent active layer which comprises a force-sensitive material configured to change at least one electrical property when subjected to a force; andat least one spacer layer arranged to maintain a separation between the electrode layer and the force-dependent active layer, until a force is applied thereto.55853631-18. The force sensor according to claim 7 wherein more than one electrode is interdigitated with the common electrode.

9. The force sensor according to claim 7 or 8 wherein the common electrode comprises one or more branches extending from a central node, each branch comprising a series of comb-like tines which are interdigitated with corresponding comb-like tines of one of the at least one electrodes.

10. The force sensor according to claim 9 wherein the one or more branches are equidistantly spaced around the central node.

11. The force sensor according to any of claims 7-10 wherein the at least one electrode and the common electrode form a planar interdigitated structure and the force-sensitive material has a footprint which substantially overlies the planar interdigitated structure.

12. The force sensor according to claim 11 wherein the planar interdigitated structure and / or the footprint have a substantially circular periphery.

13. The force sensor according to claim 11 wherein the planar interdigitated structure and / or the footprint have a substantially square, rectangular, oval or polygonal periphery.

14. The force sensor according to any of claims 7-13 wherein the force-sensitive material is provided in discrete sections and each section is arranged to overlie only one of the at least three electrodes.

15. The force sensor according to any of claims 7-14 wherein the electrode layer comprises three or four electrodes interdigitated with the common electrode.

16. The force sensor according to any of claims 7-15 comprising a plurality of the at least one sensing elements.55853631-117. The force sensor according to any of claims 7-16 wherein the force-sensitive material comprises at least one of: a quantum tunneling material, a piezoresistive material or a piezoelectric material.

18. The force sensor according to any of claims 7-17 wherein the at least one spacer layer comprises one or more of: a planar layer; a three-dimensional structure; and an adhesive.

19. A method of fabricating a force sensor according to any of claims 7-18 comprising:producing the force-dependent active layer comprising the force-sensitive material;producing the electrode layer on which the at least one electrode is interdigitated with the common electrode;arranging the force-dependent active layer on a registration jig; providing a spacer layer on the force-dependent active layer while on the registration jig; andproviding the electrode layer on the spacer layer while on the registration jig-20. The method according to claim 19 further comprising producing the spacer layer by forming cut-outs in a planar substrate, the cut-outs being arranged to align with the force-sensitive material on the force-dependent active layer.

21. The method according to claim 19 or 20 wherein each layer is produced by one or more of: printing, screen-printing, roll-to-roll printing, inkjet printing, 3D printing, electrospinning, depositing, droplet dispensing, casting, coating, moulding, spinning or weaving.55853631-1