Calibration of a force sensing device

Abstract

A method of calibrating a force sensing device (201 in figure 2) comprises establishing an optimised or golden force-resistance curve 902(and 302 in figure 3) from the mean of measured calibration data curves for several similar devices (301 in figure 3). When calibrating for a particular device, the difference between measured calibration data 901 of the force sensing device and the optimised curve 902, at a plurality of calibration points (801, 802, 803, 804 in figure 8) is determined from the measuring step. This information is then used to adapt the optimised force-resistance curve to form an adapted force-resistance curve 903 which more closely fits the measured data. The adapted curve may then be used to provide a look-up table which allow a more accurate value of applied force to be determined across the full range of measured resistance values. The method utilises a piecewise cubic Hermitian interpolating polynomial (PCHIP) to provided localised cubic interpolation between data points. The force-resistance sensor may incorporate a quantum tunnelling material and is used on the touch screen of an electronic device.

Description

BACKGROUND OF THE INVENTION The present invention relates to a method of calibrating a force sensing device and apparatus for measuring a mechanical interaction. Force sensing devices are utilised in a number of real-world applications and are often incorporated into electronic devices such as those including touch screens. Force sensing devices are typically used to calculate a magnitude or position of applied force to the touch screen, for example, and, generally, it is important to reduce the margin of error in the applied force calculated in these applications to ensure a response is consistent with a user’s input. In calibrating force sensing devices, in particular those fitted into electronic devices, the force resistance output, represented by force resistance curves, varies due to a number of factors. These factors include the design of the force sensing device itself, the design of the actuator and the integration of the force sensing device into the electronic device. The integration of the force sensing device can vary the force resistance curve depending on the level of preload applied, the force transfer from the electronic device to the force sensing device and the number of materials positioned between the force sensing device and the force applied. Thus, these problems can result in an increase in force error, given that the force error of the force sensing device initially may be reasonable and, for example, around ten percent (10%), the preload, for example, can result in an increase to around fifty percent (50%) error. This can be further compounded if when a force is applied, transfer of the force is lost through the electronic device. Thus, there remains a need to provide an appropriate calibration process which reduces these errors of incorporating force sensing devices into real world applications and corresponding electronic devices. BRIEF SUMMARY OF THE INVENTION According to a first aspect of the present invention, there is provided a method of calibrating a force sensing device, as claimed in claim 1. According to a second aspect of the present invention, there is provided an apparatus for measuring a mechanical interaction, as claimed in claim 9. Embodiments of the invention will be described, by way of example only, with reference to the accompanying drawings. The detailed embodiments show the best mode known to the inventor and provide support for the invention as claimed. However, they are only exemplary and should not be used to interpret or limit the scope of the claims. Their purpose is to provide a teaching to those skilled in the art. Components and processes distinguished by ordinal phrases such as “first” and “second” do not necessarily define an order or ranking of any sort. BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS Figure 1 shows an electronic device in the form of a mobile telephone; Figure 2 shows an example force sensing device which may be utilised in the electronic device of Figure 1; Figure 3 shows an optimised curve in relation to a set of force sensing devices of substantially similar design or arrangement; Figure 4 shows a force-resistance curve derived from measurement of calibration data; Figure 5 shows a single calibration data point used to move adapted optimised force-resistance curve closer to the data measured; Figure 6 shows a force-resistance curve highlighting the effects of a variable pre-load; Figure 7 shows the variation in force transfer in respect of a forceresistance curve; Figure 8 shows a force multiplier curve used to adjust the optimised force-resistance curve; Figure 9 shows a measured force-resistance curve, a corresponding optimised force-resistance curve, and an adapted optimised force-resistance curve in accordance with the present invention; and Figure 10 shows an illustration of the percentage resistance error for a given force. DETAILED DESCRIPTION OF EMBODIMENTS OF THE INVENTION Figure 1 Figure 1 illustrates an electronic device 101 which in this illustrated embodiment is a mobile telephone. In the embodiment, mobile telephone 101 comprises a touch screen 102 which includes a force sensing device which is configured to receive inputs from an operative 103. In use, operative 103 may apply a force to touch screen 102 by means of their finger. Electronic device 101 may comprise any suitable electronic device such as a mobile telephone, tablet, personal computer, or other similar electronic device. Figure 2 An example force sensing device 201 which may be utilised in electronic device 101 is shown in Figure 2. Force sensing device 201 comprises a first conductive layer 202, a second conductive layer 203 and a pressure sensitive layer 204. In the embodiment, pressure sensitive layer 204 between first conductive layer 202 and second conductive layer 203. In the embodiment, force sensing device 201 is a force resistive sensor. Pressure sensitive layer 204 comprises a material which exhibits a change in resistance in response to an applied force. In the embodiment, pressure sensitive layer 204 comprises a quantum tunnelling composite material. In use, a force 205 may be applied to conductive layer 203, and, when connected to an electric circuit, a current may be transmitted from conductive layer 203 to conductive layer 202 via pressure sensitive layer 204 as the layers are brought together through the application of force 205. In this way, the position and magnitude of applied force 205 can be measured from the force sensing device. In calibration, an output from the force sensing device produces a force resistance curve indicating the properties of the force sensing device in combination with the electronic device. It is important that the output is consistent with the application of force 205 made to the force sensing device in use to ensure that any further outputs, such as operations relating to applications running on electronic device 101, operate effectively. Thus, the aim is to provide an optimised curve for a given set of related force resistance curves. This optimised curve is intended to minimise the force error applied and the method described herein looks to provide a force resistance curve as close to a ‘golden curve’ as possible. In this way, the force error in the application of force and corresponding output can be reduced. Figure 3 Figure 3 illustrates an optimised curve in relation to a set of force sensing devices of substantially similar design or arrangement. Figure 3 shows a plurality of force-resistance curves 301 for a set of 20 force resistance devices in an example with zero point five Newtons (0.5N) of preload applied. As can be seen, each force sensing device varies in terms of the output on a force-resistance curve. Each of the force sensing devices for each of the force-resistance curves are substantially similar to the force sensing device described in Figure 2. However, variability in each of the force sensing devices exists in the force resistance output. As further illustrated in Figure 3, an optimised force-resistance curve 302 is shown which corresponds to the plurality of force sensing devices corresponding to the plurality of force-resistance curves 301 shown. The optimised force-resistance curve is calculated by obtaining the mean resistance at each value of force. Thus, an optimised force-resistance curve is one in which the force-resistance curve minimises the force error when obtaining a measured force from a measured resistance for the force sensing device. When calculating a traditional optimised ‘golden’ curve the plurality of force sensing devices typically utilised is between twenty and one hundred force sensing devices which gives a reasonable estimate of a true optimised force-resistance curve. While this optimised curve is a reasonable starting point for suitable calibration, limitations in its accuracy exist. If the optimised force-resistance curve 302 is utilised for calibration with no force sensing device calibration, the input force measured in use deviates from the actual force applied by an operative. In conventional applications of this type, force error values in excess of two hundred and fifty percent (250%) at low forces, and as much as one hundred percent (100%) for other forces when applied to the other force sensing devices in question are obtained. Thus, in order to obtain a force error in a final force sensing device in a preferred region, i.e. under twenty five percent (25%), the invention seeks to modify this modified curve, as will be described. Figure 4 In order to reduce the force error associated with the optimised forceresistance curve 302 of Figure 3, a measurement of calibration data of each force sensing device produces the force-resistance curve shown in Figure 4. In this example embodiment, each force sensing device in the set undertakes a simple test to produce a more accurate force-resistance curve. In this embodiment, a single calibration data point is taken from each force sensing device to produce the results as shown. Figure 4 shows a force-resistance curve 401 which illustrates the data obtained from testing a single force sensing device using one calibration data point at a given force applied. Force-resistance curve 402 represents the original optimised force-resistance curve, which may be substantially similar to optimised force-resistance curve 302. Utilising the calibration data measured in respect of force-resistance curve 401, an adapted force-resistance curve 403 is produced which conforms more accurately to the test output of force-resistance curve 401. Thus, as can be seen, the force error compared to the conventional optimised forceresistance curve 402 has been reduced. Thus, in this way, the force error can be significantly reduced when this same process is utilised for a plurality of force sensing devices such as those utilised in Figure 3. Figure 5 Figure 4 illustrates a single calibration data point being utilised which works to move the adapted optimised force-resistance curve closer to the data measured directly from the force sensing device in question. An increased number of calibration data points can produce even more accurate results. In an embodiment, two calibration data points have been shown to impact on the gradient of a force-resistance curve ensuring that the force-resistance curve in question can be further optimised. In Figure 5, additional calibration data points are utilised in order to produce further adaptation of the original optimised curve. Figure 5 illustrates the data from a force sensing device producing a force-resistance curve 501. Use of three calibration data points results in a force-resistance curve 502. By including a fourth calibration data point, force-resistance curve 503 can be produced which shows increased conformity with the force-resistance curve initially measure. In particular, by utilising four calibration data points, the adapted optimised force-resistance curve adjusts the original optimised curve by means of its curvature, gradient and offset. Thus, four calibration data points is a suitable number for both accuracy and practicality. The process utilised incorporates a method known as Piecewise Cubic Hermite Interpolating Polynomial (PCHIP) which allows for localised cubic interpolations to be undertaken between a pair of data points. This process will be described further with respect to Figure 8. Figure 6 In addition to producing the adapted optimised force-resistance curve by introducing additional calibration data points to adapt the optimised forceresistance curve, further factors including the effect of pre-load are incorporated into the process. In the examples shown in Figures 4 and 5, the examples assumed a similar pre-load of each of the force sensing devices in question. In a realistic scenario, pre-load can differ, and further calibration can be desirable to account for variable pre-loads encountered. In particular, a variable pre-load can lead to an increased force error when utilising a standard optimised forceresistance curve. Figure 6 illustrates a force-resistance curve and the effects of introducing a variable pre-load into the calibration process. In the embodiment, four calibration data points are utilised in the adapted optimised forceresistance curve in a similar manner to that of Figure 5 which accounts for the main differences between each force sensing device. In real-world applications, over the lifetime of the electronic device incorporating the force sensing device, changes occur in the pre-load after the initial calibration in line with that of Figure 5. It is not practical to return the device for repeated calibration, however, the calibration method herein determines a single calibration data point at a zero applied force for an operative. In this way, if the system can determine a situation in which the operative is not applying a force to the electronic device and force sensing device, a measurement of resistance can be calculated from the force sensing device and compared with the original calibration data point at zero applied force. Thus, adjustments to the calibration, such as a fixed resistance offset, a proportional resistance offset, ora graduated resistance offset can be made dependent on the force applied. Figure 6 therefore illustrates the further adaptation of the optimised force-resistance curve in response to a variable pre-load. The original adapted optimised curve 601 is consistent with curve 503, for example, shown in Figure 5. Force-resistance curve 602 shows the effect of applying a fixed offset. Force-resistance curve 603 applies a graduated offset and force-resistance curve 604 applies a proportional offset. Consequently, the force-resistance curves illustrate that a preferred adaptation is a combination of the graduated and proportional offset force-resistance curves 603 and 604. Figure 7 Figure 7 illustrates the variation of force transfer in respect of a forceresistance curve. In Figure 7, force-resistance curve 701 illustrates the effect on the force-resistance curve when force transfer of the applied force to the force sensing device is around ten percent (10%) efficiency. In contrast, forceresistance curve 702 illustrates a force-resistance curve that would be expected if all of the force applied by an operative was transferred fully to the force sensing device. Thus, it can be seen that the force transfer efficiency, or lack thereof, results in the force-resistance curve being translated to the right by a factor of ten. Again, the method further adapts the force-resistance curve to reduce this error. Figure 8 As noted previously, the method utilises Piecewise Cubic Hermite Interpolating Polynomial (PCHIP) which allows for localised cubic interpolations to be undertaken between a pair of data points. In the embodiment, the PCHIP process takes four predetermined forces on the force resistance curve and records the difference between the optimised force-resistance curve and the actual force-resistance curve of the electronic device incorporating the force sensing device. As shown in Figure 8, four multipliers or differences are obtained as illustrated at points 801,802, 803 and 804. The force multiplier curve therefore provides a means to adjust the optimised force-resistance curve to fit the electronic device and force sensing device combination. Thus, the multiplier points 801, 802, 803 and 804 can then be used by means of the PCHIP interpolation. The multiplier value at each value of force can be determined. Multiplier values are determined by measuring the resistance in the force sensing device incorporated into an electronic device build at four predetermined forces and comparing these with the optimised curve resistances. PCHIP interpolation can then predict the value of the multiplier at all forces based around the four sampling point multipliers. In the embodiment, the first calibration data point is always fixed and determined to be at the lowest force. This provides an effective anchor to the adapted optimised force-resistance curve, thereby assisting with the PCHIP interpolation. The remaining three multiplier data points are determined on the basis of the force range of the force sensing device. The PCHIP function provides a prediction which allows a suitable look-up table to be created where a value of the multiplier is obtained for each force value that is to be included in the look up table. Figure 9 Figure 9 shows an actual force-resistance curve 901 measured from the electronic device in combination with the force sensing device, a corresponding optimised force-resistance curve 902, and an adapted optimised force-resistance curve 903 based on the interpolation method described in Figure 8. It can be seen that force-resistance curve 902 and force-resistance curve 903 show close proximity indicating a much closer prediction for real-world results. Figure 10 Using the example of Figures 8 and 9, an illustration of the percentage resistance error for a given force is shown in Figure 10. Force-resistance curve 1001 illustrates the conventional optimised force-resistance curve and the percentage of resistance error associated with utilising this force-resistance curve. In this illustration, at low forces, the resistance error may be as high as fifty percent (50%). In contrast, force-resistance curve 1002 illustrating the resistance error for the adapted force-resistance curve shows a more consistent error of a maximum percentage error of between minus two and three percent (-2 and 3%). Thus, the invention provides a substantial reduction to the force error observed in an electronic device by following the calibration method and, experimental data has indicated that the force error may reduce from a triple digit percentage to a single digit percentage. The invention indicates that the preload variation, preload increases can 5 be calibrated out of the system, and preload decreases can be accounted for through a single automatic zero force calibration outside of factory conditions with a representative force error of no greater than twenty five percent (25%). 10

Claims

The invention claimed is:

1. A method of calibrating a force sensing device, said force sensing device comprising a first conductive layer, a second conductive layer and a pressure sensitive layer between said first conductive layer and said second conductive layer, and said method comprising the steps of:establishing an optimised force-resistance curve by obtaining a mean resistance of a plurality of force-resistance curves for a set of substantially similar force sensing devices;measuring calibration data of said force sensing device;applying a plurality of calibration points defined from said measuring step to said optimised force-resistance curve; andadapting said optimised force-resistance curve to form an adapted force-resistance curve by utilising said plurality of calibration points; whereinsaid step of adapting said optimised force-resistance curve comprises utilising PCHIP to adapt said optimised force-resistance curve into said adapted force-resistance curve.

2. The method of claim 1, wherein said plurality of calibration points is three or more.

3. The method of claim 1 or claim 2, wherein said plurality of calibration points is four.

4. The method of any one of claims 1 to 3, wherein said method provides a maximum force error of twenty-five percent.

5. The method of any one of claims 1 to 4, further comprising the step of:applying a preload to an electronic device comprising said force sensingdevice.

6. The method of claim 1, wherein said pressure sensitive layer comprises a quantum tunnelling material.

7. The method of any preceding claim, wherein said calibration points adapt the optimised force-resistance curve by the gradient, curvature or offset of the optimised curve such that said calibration points are in close proximity to a measured force sensing device data force resistance curve.

8. The method of any preceding claim, further comprising the steps of:measuring a resistance from a force sensing device when no force is applied;comparing said measured resistance with a predetermined zero forceresistance measurement; andproviding further adaptation of said optimised force-resistance curve to account for the difference in the two measurements.

9. Apparatus for measuring a mechanical interaction, comprising:a force sensing device comprising a first conductive layer, a second conductive layer and a pressure sensitive layer between said first conductive layer and said second conductive layer; anda processor configured to:establish an optimised force-resistance curve by obtaining a mean resistance of a plurality of force-resistance curves for a set of substantially similar force sensing devices;measure calibration data of said force sensing device;apply a plurality of calibration points defined from said measuringstep; and10adapt said optimised force-resistance curve to form an adapted force-resistance curve by utilising said plurality of calibration points; wherein said step of adapting said optimised force-resistance curve comprises utilising PCHIP to adapt said optimised force-resistance curve into said adapted force-resistance curve.

10. The apparatus of claim 9, wherein said pressure sensitive layer comprises a quantum tunnelling material.

11. The apparatus of claim 9 or claim 10, wherein said apparatus is incorporated into an electronic device.

12. The apparatus of claim 11, wherein said electronic device comprises a touch screen.15

Images