Tactile sensor, method for manufacturing tactile sensor, method for regenerating tactile sensor, method for detecting wear of tactile sensor, and method for confirming regeneration of tactile sensor

Abstract

To provide a tactile sensor capable of regenerating a worn tactile sensor, a method for manufacturing the tactile sensor, a method for regenerating the tactile sensor, a method for detecting wear of the tactile sensor, and a method for confirming regeneration of the tactile sensor.SOLUTION: A tactile sensor TS for measuring tactile sensation caused by a frictional stimulus, the tactile sensor includes a contact LB that slides at a predetermined speed while being in contact with a predetermined surface of an object to be measured, and a strain gauge embedded in the contact. The contact is a laminate including a skin layer S and a fat layer F. The skin layer has irregularities to be brought into contact with the predetermined surface of the object to be measured, and is formed of a material capable of reproducing the irregularities by heating when the irregularities are worn. The fat layer has a Shore D hardness smaller than that of the skin layer.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates to a tactile sensor, a method for manufacturing a tactile sensor, a method for restoring a tactile sensor, a method for detecting wear in a tactile sensor, and a method for confirming the restoration of a tactile sensor. [Background technology]

[0002] There have been a wide variety of devices available for measuring the tactile sensation of objects with uneven surfaces, such as textiles and car seats. For example, cited document 1 describes a tactile evaluation measuring device as shown in Figure 1, which comprises a measuring device for measuring frictional force (frictional resistance) and acceleration, which are characteristics of a specific surface of an object to be measured, and a control device for performing arithmetic processing to calculate a friction coefficient from the frictional force based on the measurement results obtained by the measuring device. Cited Document 2 describes a tactile sensor, as shown in FIG. 2, that detects the tactile sensation received from an object, and that includes a contact part that forms a closed space between itself and the surface of the object when brought into contact with the object, and a vibration sensor that detects air vibrations within the closed space. Cited Document 3 describes a tactile measurement system as shown in FIG. 3, which includes a mounting base, a tactile sensor, a pressing means, a sliding means, a force sensor, and a control unit. Cited Document 4 describes a tactile sensation extraction device that includes a plate, a holding unit, a moving device, and a tactile sensation extraction means.

[0003] Incidentally, epidermal ridges, one of the structures of human fingers, play an important role in human tactile sensing ability. Epidermal ridges are raised, wavy patterns on the skin surface, especially on the palms of the hands and soles of the feet. Many studies on tactile sensing have attempted to mimic the structure of human fingers to reproduce the sensing function of the human finger pad. Experiments have demonstrated that epidermal ridges significantly enhance tactile signals by 100 times when rubbing a specific surface wavelength of the object being touched. Therefore, without epidermal ridges, the received signal level of the tactile sensor would be reduced. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2016-38317 [Patent Document 2] Japanese Patent Publication No. 2020-112472 [Patent Document 3] Japanese Patent Publication No. 2021-193337 [Patent Document 4] Japanese Patent Application Publication No. 2023-113530 Summary of the Invention [Problem to be solved by the invention]

[0005] Human skin ridges have a healing mechanism, regenerating even when small wounds occur, maintaining the shape of the ridges. In contrast to humans, mechanical touch sensors lack a mechanism to autonomously maintain their structural integrity. Therefore, when the ridges wear out, they cannot accurately measure the sensation of touch. None of the above documents describe how to restore worn parts of mechanical tactile sensors. The parts removed by wear are often very small, making it impractical to restore them by re-adhesion. The present invention has been made to solve the above-mentioned problems, and aims to provide a tactile sensor that can regenerate worn tactile sensors, a method for manufacturing a tactile sensor, a method for regenerating a tactile sensor, a method for detecting wear in a tactile sensor, and a method for confirming the regeneration of a tactile sensor. [Means for solving the problem]

[0006] As a result of extensive research into solving the above problems, the inventors discovered that the above problems could be solved by using a material that can restore its unevenness when heated as the main component of the tactile sensor, and thus completed the present invention. The present invention has the following aspects. [1] A tactile sensor for measuring tactile sensations caused by friction, The tactile sensor has a contactor that slides at a predetermined speed while being brought into contact with a predetermined surface of an object to be measured, and a strain gauge embedded in the contactor; the contact is a laminate having a skin layer and a fat layer, the surface layer has irregularities that come into contact with a predetermined surface of the object to be measured, and is formed of a material that can restore the irregularities by heating when the irregularities are worn; A tactile sensor, wherein the fat layer has a Shore D hardness lower than that of the epidermis layer. [2] The tactile sensor according to [1], wherein the resin composition constituting the skin layer contains a thermoplastic elastomer. [3] The tactile sensor according to [2], wherein the resin composition has a glass transition temperature of 50 to 100°C. [4] The tactile sensor according to [2], wherein the resin composition contains polyvinyl chloride. [5] The tactile sensor according to [1], wherein the longitudinal elastic modulus of the epidermal layer, calculated from the Shore D hardness or Shore E hardness measured in accordance with JIS K6253, is 0.4 to 2.0 MPa, and the longitudinal elastic modulus of the fat layer, calculated from the Shore D hardness or Shore E hardness measured in accordance with JIS K6253, is 0.05 to 0.3 MPa. [6] The tactile sensor according to [1], wherein the difference between the modulus of elasticity calculated from the Shore D hardness or Shore E hardness of the epidermis layer measured in accordance with JIS K6253 and the modulus of elasticity calculated from the Shore D hardness or Shore E hardness of the fat layer measured in accordance with JIS K6253 is 0.1 to 1.95 MPa. [7] The tactile sensor according to [1], wherein the average height of the skin layer is 0.5 to 4.0 mm. [8] A method for manufacturing the tactile sensor according to [1], A method for manufacturing a tactile sensor, comprising a roughness regeneration step of regenerating the roughness of the skin layer by heating the skin layer having worn roughness while contacting it with a mold having a structure corresponding to the roughness. [9] The method for manufacturing a tactile sensor according to [8], wherein the heating temperature in the unevenness regenerating step is 50 to 100°C.

[10] A method for regenerating the tactile sensor according to [1], A method for restoring a tactile sensor, comprising a step of restoring the surface layer having worn irregularities by heating the surface layer while bringing it into contact with a mold having a structure corresponding to the irregularities.

[11] The method for regenerating a tactile sensor according to

[10] , wherein the heating temperature in the unevenness regeneration step is 50 to 100°C.

[12] A method for detecting wear of the tactile sensor according to [1], The strain power V from the strain gauge was measured using a used tactile sensor on an object with a specified surface roughness. signal and the noise power V from the strain gauge noise So, The signal to noise ratio (SNR) is calculated using the following formula (V): A method for detecting wear of a tactile sensor, comprising a wear detection step of determining that the skin layer of the contact of the tactile sensor has worn away if the SNR is less than a predetermined value. SNR(dB)=20log 10 (V signal / V noise ) ···(V) In formula (V), V signal is the power or amplitude of the strain signal when the contact is pressed against a sample with a specified surface roughness and the surface of the sample is repeatedly rubbed; V noise is the power or amplitude of the noise when the contact is pressed against a sample having no surface roughness and the surface of the sample is repeatedly rubbed.

[13] A method for checking the regeneration of the tactile sensor according to [1], The power V of the signal from the strain gauge that measured the tactile sensation of an object with a specified surface roughness using the tactile sensor regenerated by the method described in

[10] . signal and the noise power V from the strain gauge noise So, The signal to noise ratio (SNR) is calculated using the following formula (V): A method for checking the regeneration of a tactile sensor, comprising a regeneration checking step of determining that the skin layer of the contact of the tactile sensor has been regenerated if the SNR is equal to or greater than a predetermined value. SNR(dB)=20log 10 (V signal / V noise ) ···(V) In formula (V), V signal is the power or amplitude of the strain signal when the contact is pressed against a sample with a specified surface roughness and the surface of the sample is repeatedly rubbed; V noise is the power or amplitude of the noise when the contact is pressed against a sample having no surface roughness and the surface of the sample is repeatedly rubbed.

[14] Use of a laminate for manufacturing a tactile sensor, the laminate having a skin layer formed of a material having an irregular surface and capable of restoring the irregularities by heating when the irregularities are worn away, and a fat layer having a Shore D hardness lower than that of the skin layer.

[15] A laminate having a surface with irregularities and a skin layer formed of a material that can regenerate the irregularities by heating when the irregularities wear away, and a fat layer having a Shore D hardness lower than that of the skin layer, is used as a contact in a tactile sensor that slides at a predetermined speed while making contact with a predetermined surface of an object to be measured. [Effects of the Invention]

[0007] The present invention provides a tactile sensor that can regenerate a worn tactile sensor, a method for manufacturing a tactile sensor, a method for regenerating a tactile sensor, a method for detecting wear in a tactile sensor, and a method for confirming the regeneration of a tactile sensor. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a schematic diagram showing a tactile evaluation and measurement device disclosed in Patent Document 1. [Figure 2] FIG. 1 is a schematic diagram showing the tactile sensor of Patent Document 2. [Figure 3] FIG. 1 is a schematic diagram showing a tactile sensation measuring system of Patent Document 3. [Figure 4] (a) is a perspective view of the tactile sensor, (b) is a front view of the contact, and (c) is a side view of the contact. [Figure 5] FIG. 10 is a perspective view showing a contact manufacturing kit. [Figure 6] This is a photo of the contactor being pressed against an object and rubbed by moving it horizontally. [Figure 7] FIG. 2 is a cross-sectional view showing the unevenness of an object. [Figure 8] 1 is a photograph showing a periodic signal output from a strain gauge amplifier. [Figure 9] This is the power spectrum after Fourier transform of a periodic signal. [Figure 10] 10 is a graph showing the relationship between SNR and the number of times of rubbing. [Figure 11] 1A is a perspective view of a contact regeneration kit, FIG. 1B is a perspective view of the regeneration kit after assembly, and FIG. 1C is a photograph of another regeneration kit. [Figure 12] FIG. 1(a) is a front view of the contactor used in the example, and FIG. 1(b) is a side view of the contactor used in the example. [Figure 13] FIG. 2 is a cross-sectional view showing the unevenness of the object used in the examples. [Figure 14] 1 is a photograph of an object used in an example. [Figure 15] 1 is a photograph of a robot arm equipped with a tactile sensor used in the examples. [Figure 16] (a) The raw signal from the strain gauge after rubbing once when the object has a periodic width and height of 1 mm, (b) the spectrum after Fourier transform of (a), (c) The raw signal from the strain gauge after rubbing once when the object has a width and height of 2 mm, (b) the spectrum after Fourier transform of (c). [Figure 17] 10 is a graph showing the relationship between the SNR at 20 Hz after one rub and the number of rubs when the periodic width and height of the object are 2 mm. [Figure 18](a) is the amplitude spectrum of the strain signal recorded when a sample with a surface wavelength of 1 mm was rubbed at a speed of 20 mm / s using a contact before the unevenness on the sensor surface had worn away, and is a graph showing the amplitude spectrum including the average and standard error of five measurements. (b) is the amplitude spectrum of the strain signal recorded when a sample with a surface wavelength of 1 mm was rubbed at a speed of 20 mm / s using a contact after the unevenness on the sensor surface had worn away, and is a graph showing the amplitude spectrum including the average and standard error of five measurements. [Figure 19] (a) is a photograph showing the abraded state of the surface layer after 350 rubs, and (b) is a photograph showing the restored state of the abraded surface layer after it has been regenerated. [Figure 20] This spectrum compares the power spectrum when a new contact is used with the power spectrum when a worn contact is reused. DETAILED DESCRIPTION OF THE INVENTION

[0009] An example of the present invention will be described below with reference to the drawings, but the scope of the present invention is not limited to the embodiment described here, and various modifications can be made without departing from the spirit of the present invention. Furthermore, when multiple upper and lower limit values ​​are specified for a specific parameter, any of these upper and lower limit values ​​can be combined to form a suitable numerical range.

[0010] <<Tactile sensor>> The tactile sensor of the present invention has a contactor that comes into contact with a predetermined surface of an object to be measured, and a strain gauge embedded inside the contactor.

[0011] <Contact> The contacts are laminates made up of two layers: an epidermal layer that mimics human skin and an adipose layer. Figure 4(a) is a perspective view showing an example of a tactile sensor, (b) is a front view showing an example of a contact, and (c) is a side view. The contact LB is a laminate consisting of a skin layer S and a fat layer F. Hereinafter, the left-right direction in Figure 4(b) will be referred to as the width direction, the left-right direction in Figure 4(c) will be referred to as the thickness direction, and the up-down direction in Figures 4(a) to 4(c) will be referred to as the height direction. The width direction, thickness direction, and height direction are assumed to be perpendicular to each other. As shown in Figure 4(a), the tactile sensor TS has a contact LB and a strain gauge embedded inside the contact. The strain gauge is located between the epidermis layer S and the fat layer F and is connected to a strain amplifier (not shown) via a lead L. This allows the power of the strain detected by the strain gauge to be transmitted to the strain amplifier via the lead L. As shown in Figures 4(a) to (c), the fat layer F is a layer in the shape of a substantially rectangular parallelepiped. The upper surface of the fat layer F has a curved surface in which the central portion is curved outward in the height direction. The curved surface has the same height along the thickness direction, and the height of the central portion in the width direction is approximately twice the height of the end portions in the width direction. The skin layer S is laminated so as to cover the curved surface of the fat layer F, and is a layer that is curved outward as a whole. On the outer surface of the skin layer S, multiple convex ridges extending in the thickness direction of the laminate are formed continuously and parallel to each other without any space between them. As shown in Figure 4(c), the height H of the fat layer F of the laminate as seen from the side F and the height H of the epidermis layer S S is almost the same as A cross section of the laminate taken perpendicular to the thickness direction is substantially the same as the front view. In the cross section, the cross section of the multiple convex ridges of the epidermal layer S forms multiple arc-shaped convex portions, and the convex portions are continuously arranged without any space between them along the width direction of the laminate. A line connecting the vertices of the multiple arc-shaped convex portions follows the arc formed by the contact surface of the epidermal layer S with the fat layer F. The multiple convex portions have a structure that resembles a human fingerprint. The multiple arc-shaped convex portions have approximately the same radius, and a line connecting the centers of the multiple circles follows the arc formed by the contact surface of the epidermal layer S with the fat layer F.

[0012] (epidermal layer) The surface layer has irregularities that come into contact with a predetermined surface of the object to be measured, and is made of a material that has a self-repairing function that can repair the irregularities that have worn away. Examples of materials with self-repairing properties include thermoplastic elastomer-containing compositions containing thermoplastic elastomers, such as polyvinyl chloride resins, polyethylene resins, polypropylene resins, polystyrene resins, and polyester resins.

[0013] The polyvinyl chloride resin may be a vinyl chloride homopolymer or a copolymer of vinyl chloride and other monomers copolymerizable therewith. The copolymer may be a graft copolymer, a block copolymer, or a random copolymer. Examples of other monomers include olefins such as ethylene, propylene, and polybutene; vinyl esters of saturated acids such as vinyl acetate and vinyl laurate; alkyl esters of unsaturated acids such as methyl acrylate and methyl methacrylate; alkyl vinyl ethers such as lauryl vinyl ether; maleic acid, acrylonitrile, styrene, methylstyrene, vinylidene chloride, and vinylidene fluoride. When the polyvinyl chloride resin is a copolymer, the content of vinyl chloride units in the copolymer may be 10% by mass or more based on the total amount of monomer units.

[0014] The polyvinyl chloride resin may be, for example, a polymer blend with a three-dimensional polymer such as acrylonitrile-butadiene-styrene, a polyvinyl chloride resin post-treated with alcohol, or a polyvinyl chloride resin post-treated with a chlorine-containing compound. In these cases, the content of vinyl chloride units in the polyvinyl chloride resin may be 10% by mass or more based on the total amount of the resin.

[0015] The polyethylene resin may be a copolymer containing low-density polyethylene, linear low-density polyethylene, linear very low-density polyethylene, medium-density polyethylene, high-density polyethylene, or ethylene as a monomer unit. Specifically, the polyethylene resin may be a copolymer of ethylene and one or more unsaturated compounds selected from α-olefins having 3 to 10 carbon atoms, such as propylene, 1-butene, 1-pentene, 1-hexene, 4-methyl-1-pentene, 1-heptene, and 1-octene; vinyl esters, such as vinyl acetate and vinyl propionate; unsaturated carboxylic acid esters, such as methyl acrylate, ethyl acrylate, methyl methacrylate, and ethyl methacrylate; and unsaturated compounds, such as conjugated dienes and non-conjugated dienes. The content of ethylene units in the polyethylene resin may be 50% by mass or more based on the total amount of monomer units.

[0016] The content of the thermoplastic elastomer may be 70% by mass or more relative to the total mass of the resin composition forming the skin layer.

[0017] In addition to the thermoplastic elastomer, the skin layer may further contain other components such as a plasticizer, an antifogging agent, a heat stabilizer, a light stabilizer, a lubricant, a filler, an antioxidant, a release agent, a viscosity reducer, a surfactant, a colorant, a fluorescent agent, a surface treatment agent, a crosslinking agent, a processing aid, an adhesive, an antistatic agent, and an ultraviolet absorber.

[0018] Examples of plasticizers include epoxidized oils such as epoxidized linseed oil, epoxidized soybean oil, and epoxidized fatty acid alkyl esters; adipates having a linear or branched alkyl group with 6 to 10 carbon atoms; hydroxy polycarboxylic acid esters such as acetyl tributyl citrate and acetyl triethyl; fatty acid dibasic acid esters such as di-normal butyl sebacate; glycol esters such as pentaerythritol esters and diethylene glycol benzoate; phosphate esters such as triphenyl phosphate and tricresyl phosphate; glycerin diacetomonolaurate; chlorinated paraffin; and polyester plasticizers. These plasticizers may be used alone or in combination of two or more. The content of the plasticizer is appropriately adjusted, for example, within a range of 20 to 50 parts by mass per 100 parts by mass of the polyvinyl chloride resin.

[0019] The resin composition forming the skin layer is preferably one that can be gelled and molded at a predetermined temperature. The predetermined temperature is, for example, preferably 50 to 100°C, more preferably 60 to 90°C. If the resin composition gels within the above range, when the irregularities of the skin layer are worn away, the irregularities of the skin layer can be easily restored by contacting the skin layer with a mold having a structure corresponding to the irregularities and heating the surface. The glass transition point of the resin composition forming the skin layer is, for example, preferably 50 to 100°C, and more preferably 60 to 90°C.

[0020] The melting point of the resin composition forming the surface layer is preferably 150 to 200° C., more preferably 160 to 190° C. When the melting point is within the above range, the surface layer does not become too liquid when being regenerated, and the irregularities of the surface layer can be more easily regenerated while maintaining a shape that is easy to handle.

[0021] Among other things, polyvinyl chloride plastisol (Plastic Worm, Two-L Co., Ltd., Japan) can be used as the thermoplastic elastomer, whose hardness can be adjusted by adjusting the mixing ratio of the softener and plasticizer.

[0022] (fat layer) The fat layer has a lower Shore D hardness than the epidermis layer. The fat layer is made of different materials than the epidermis layer. The fat layer may be made of the same materials as those listed above for the skin layer, or may be made of a thermoplastic resin or a thermosetting resin. The fat layer is preferably made of a material that does not deform when heated, and is preferably made of a thermosetting resin, more preferably rubber. Examples of rubber include acrylic rubber, nitrile rubber, silicone rubber, fluororubber, and vulcanized rubber.

[0023] When the resin composition constituting the fat layer contains a thermoplastic resin, the melting point of the fat layer is preferably higher than that of the epidermis layer. The melting point of the fat layer is, for example, preferably 200°C or higher, more preferably 250 to 500°C. When the melting point of the fat layer is within the above range, the structure of the fat layer is less likely to deform when the epidermis layer is brought into contact with a mold and heated to recreate the irregularities, making it easier to recreate the irregularities of the epidermis layer while maintaining a shape that is easy to handle.

[0024] The longitudinal elastic modulus of the surface layer, calculated from the Shore D hardness or Shore E hardness determined in accordance with JIS K6253, is preferably 0.4 to 2.0 MPa, more preferably 0.4 to 1.0 MPa, and even more preferably 0.5 to 0.9 MPa. The longitudinal elastic modulus of the fat layer, calculated from the Shore D hardness or Shore E hardness determined in accordance with JIS K6253, is preferably 0.05 to 0.3 MPa, more preferably 0.05 to 0.2 MPa, and even more preferably 0.1 to 0.2 MPa. The difference expressed as [modulus of elasticity converted from Shore D hardness or Shore E hardness of the epidermal layer] - [modulus of elasticity converted from Shore D hardness or Shore E hardness of the fatty layer] is preferably 0.1 to 1.95 MPa, more preferably 0.3 to 1.0 MPa, and even more preferably 0.5 to 1.0 MPa. If the hardness of the epidermis layer and the Shore D hardness of the fat layer are within the above ranges, the touch can be made to be closer to that of a human touch, and the accuracy of the touch sensor can be improved.

[0025] The height of the skin layer is preferably 0.5 to 4.0 mm, more preferably 0.5 to 3.0 mm, and even more preferably 0.5 to 2.0 mm. The average height of the skin layer is preferably 0.5 to 3.0 mm, more preferably 0.5 to 2.0 mm, and even more preferably 0.5 to 1.0 mm. The height of the fat layer is preferably 5.0 to 30.0 mm, more preferably 5.0 to 20.0 mm, and even more preferably 5.0 to 15.0 mm. The average height of the fat layer is preferably 5.0 to 25.0 mm, more preferably 5.0 to 15.0 mm, and even more preferably 5.0 to 10.0 mm. The height ratio represented by [height of the epidermis layer] / [height of the fat layer] is preferably 0.01 to 0.1, more preferably 0.02 to 0.1, and even more preferably 0.03 to 0.1. When the height or average height of the epidermis layer and the height or average height of the fat layer are within the above ranges, the accuracy of the tactile sensor can be improved, and the unevenness of the epidermis layer can be more easily reproduced while maintaining a shape that is easy to handle. In this specification, the height can be determined by measuring the height at any five points using a microgauge. In this specification, the average height can be determined by measuring the height at any five points using a microgauge and calculating the average value.

[0026] The epidermis layer covers the upper surface of the fat layer, and has irregularities on the surface opposite to the surface in contact with the fat layer. The unevenness of the surface layer may be a structure in which a plurality of convex portions having an arc-shaped cross section are continuously arranged without any space therebetween. The distance (pitch) between the centers of adjacent arc-shaped convex portions is preferably 0.2 to 2 mm, more preferably 0.3 to 1.0 mm, and even more preferably 0.3 to 0.5 mm. The maximum height of the projections in the skin layer is preferably 0.1 to 1.0 mm, and more preferably 0.1 to 0.5 mm.

[0027] The fat layer is shaped like a plate with an outward curved upper surface. The width of the fat layer is preferably 30 to 60 mm, more preferably 40 to 60 mm, and even more preferably 40 to 50 mm. The height of the fat layer at the widthwise ends is preferably 5 to 30 mm, more preferably 5 to 20 mm, and even more preferably 5 to 15 mm, and the height at the center is preferably 5 to 40 mm, more preferably 5 to 30 mm, and even more preferably 5 to 20 mm. The thickness of the fat layer is preferably 10 to 40 mm, more preferably 10 to 30 mm, and even more preferably 10 to 25 mm.

[0028] The width of the laminate is preferably 30 to 60 mm, more preferably 40 to 60 mm, and even more preferably 40 to 50 mm. The height of the laminate at the widthwise end is preferably 10 to 40 mm, more preferably 10 to 30 mm, and even more preferably 15 to 30 mm, and the height at the center is preferably 20 to 40 mm, more preferably 20 to 35 mm, and even more preferably 20 to 30 mm. The thickness of the laminate is preferably 10 to 40 mm, more preferably 10 to 30 mm, and even more preferably 10 to 25 mm.

[0029] To determine the hardness of each layer, the following formula (1) can be used to determine the relationship between Shore D hardness and Young's modulus.

[0030]

number

[0031] In formula (1), E represents Young's modulus (MPa), and s represents Shore D hardness. The Shore hardness of a material can be measured using a durometer in accordance with JIS K7215. More specifically, it can be measured by the method described in Dental Materials, Volume 25, Issue 8, August 2009, Pages 956-959.

[0032] The contact can be manufactured using a mold having a structure corresponding to the desired uneven structure. For example, as shown in FIG. 5, the contact may be manufactured using a contact manufacturing kit k1 including a skin layer front surface mold m1 having a shape corresponding to the unevenness of the surface of the skin layer, a skin layer rear surface mold m2 having a structure corresponding to the rear surface of the skin layer, a fat layer strain gauge mold m3 for forming holes in the fat layer for embedding strain gauges, and a contact outer surface mold m4 corresponding to the overall shape of the contact. The skin layer front surface mold m1 and the contact outer surface mold m4, when combined, form the outer wall of a skin layer molding chamber for forming the skin layer. The skin layer rear surface mold m2 has an injection hole for injecting resin into the skin layer molding chamber. The skin layer rear surface mold m2, the fat layer strain gauge mold m3, and the contact outer surface mold m4, when combined, form the outer wall of a fat layer molding chamber for forming the fat layer.

[0033] The procedure for manufacturing a contact using a contact manufacturing kit is described below. First, the contact outer surface mold m4, the skin layer back surface mold m2, and the fat layer strain gauge mold m3 of the contact manufacturing kit shown in Figure 5 are combined. At this time, the tip of the strain gauge is inserted into the hole in the fat layer strain gauge mold m3. The resin composition for forming the skin layer, heated to a temperature above its melting point, is poured into the recess formed by the contact outer surface mold m4, the skin layer back surface mold m2, and the fat layer strain gauge mold m3. This fills the skin layer molding chamber formed by the contact outer surface mold m4, the skin layer back surface mold m2, and the fat layer strain gauge mold m3 with the resin composition. The skin layer molding chamber is cooled to solidify the resin composition and form the fat layer. Next, the mold m2 for the back surface of the skin layer is removed, and the resin composition for forming the skin layer, heated to a temperature above its melting point, is poured into the recess formed by the mold m4 for the outer surface of the contact, the mold m1 for the front surface of the skin layer, and the mold m3 for the strain gauge. The skin layer molding chamber is cooled to harden the resin composition and form the skin layer. The epidermal layer may be formed before the fat layer is formed, or the fat layer may be formed before the epidermal layer is formed.

[0034] <Strain gauge> The strain gauge is installed to detect the strain caused by bending when the contact is pressed against the surface of the object, and to measure the amount of strain (power, unit: volts). The strain gauge is embedded so as to straddle the fat layer and the epidermis layer of the contact, and detects the strain caused by bending when the epidermis layer is pressed against the surface of the object. The amount of strain detected by the strain gauge is amplified by a strain gauge amplifier and output as a voltage signal.

[0035] For example, as shown in Figure 6, if a contactor is pressed against a sample with a known surface roughness (spatial frequency) (for example, as shown in Figure 7, the width of the convex portion is 1 mm, the width of the concave portion is 1 mm, and the height of the convex portion from the bottom of the concave portion is 1 mm), and the surface of the sample is repeatedly rubbed, a voltage signal is output by the strain gauge amplifier, as shown in Figure 8. Separately, when the contacts are pressed against a sample with a smooth surface and the surface of the sample is rubbed repeatedly, noise is output as a voltage signal to the strain gauge amplifier, which allows the measurement of noise originating from vibrations of the device or amplifier. The obtained distortion spectrum is subjected to a Fourier transform to obtain an amplitude spectrum or power spectrum as shown in Fig. 9. In Fig. 9, the vertical axis is decibels (unit: dB) and is calculated using the following formula (PS). dB value = 20log 10 (V / 1) (PS) In equation (PS), V is the signal (unit: V) after being amplified by the gauge amplifier, and the dB value represents the ratio of amplitude or power to 1 V (volt). In FIG. 9, the horizontal axis represents the frequency (unit: Hz) obtained by Fourier transforming the periodic signal, and is calculated using the following formula (f).

[0036]

number

[0037] In equation (f), j is the imaginary unit, π is the ratio of the circumference of a circle to its diameter, f is frequency, t is time, and x(t) represents a time signal. 9 shows frequencies from 0 to 200 Hz, but since the frequencies that a human finger can detect are in the range of 0 to 1000 Hz, the object may be designed so that a power spectrum in the frequency range of 0 to 1000 Hz can be obtained with a single object. For example, the object may be designed so that the frequency changes from a region where the spacing between the projections and recesses on the object is large and the frequency is small to a region where the spacing between the projections and recesses is small and the frequency is large. The noise spectrum is similarly obtained by Fourier transform.

[0038] The signal-to-noise ratio (SNR) is calculated from the power spectrum and the noise spectrum using the following formula (V). SNR(dB)=20log 10 (V signal / V noise ) ···(V) In formula (V), V signal is the amplitude or power of strain when the contact is pressed against a sample with a specified surface roughness and the surface of the sample is repeatedly rubbed; V noise is the amplitude or power of the noise when the contact is pressed against a sample having no surface roughness and the surface of the sample is repeatedly rubbed. In other words, the SNR can also be calculated by the following formula (V'). SNR(dB)=20log 10 (V signal )-20log 10 (V noise ) (V') In formula (V'), V signal , and V noise are the same as above.

[0039] When the unevenness of the contact's surface layer wears away due to wear, the SNR decreases, which means a decline in the performance of the tactile sensor. For example, an SNR of 10 dB or higher is preferable, 15 dB or higher is more preferable, and 20 dB or higher is even more preferable. Although it is preferable that the frequency is equal to or higher than the lower limit at any frequency, it is sufficient that the frequency is equal to or higher than the lower limit in the predetermined frequency range to be detected, and if the irregularities in the skin layer are worn out outside the frequency range to be detected, the contacts may or may not be regenerated. The frequency range to be detected varies depending on the type of object, but when measuring the feel of, for example, woven fabric or a vehicle seat, it is preferably 0 to 500 Hz, more preferably 20 to 500 Hz, and even more preferably 20 to 300 Hz.

[0040] Figure 10 is an example of a spectrum showing the relationship between SNR and the number of times the sample surface is rubbed. In the example shown in Figure 10, it can be seen that the SNR begins to decrease after the contactor is pressed against the sample and the surface of the sample is rubbed about 700 times. In this way, the SNR of the contactor is calculated using a sample with a known surface frequency, and if the SNR is below a predetermined value (for example, the above-mentioned lower limit), the contactor of the tactile sensor is regenerated.

[0041] <<How to Recycle a Tactile Sensor>> The method for regenerating a tactile sensor of the present invention includes an unevenness regeneration step of regenerating the unevenness of the skin layer by heating the skin layer having worn unevenness while contacting it with a regeneration mold having a structure corresponding to the unevenness.

[0042] The heating temperature in the unevenness regeneration step is preferably 50 to 100°C, more preferably 60 to 90°C. When the heating temperature is within the above range, only the epidermis layer can be softened without deforming the fat layer, resulting in excellent handleability. Furthermore, since the epidermis layer can be regenerated without being completely liquefied, the tactile sensor can be regenerated at low cost.

[0043] The mold only needs to be in contact with the surface of the skin layer, and may be a regeneration kit K having, for example, as shown in Figure 11(a), a mold M having a shape corresponding to the unevenness of the surface of the skin layer, a heater H placed on the opposite side of the contacts to heat the mold, a housing Ho that stores the entire contacts and has multiple holes through which bolts can pass to press the contacts from the fat layer side, a side mold SM that presses the contacts placed on the mold M, and a side cover SL that forms the side of the housing H and has multiple holes through which bolts can pass to press the contacts from the side.

[0044] For example, when using the regeneration kit K shown in FIG. 11(a) to regenerate the skin layer, the skin layer side of the contact is brought into contact with the irregularities of the mold M, a heater H is installed under the mold M, the side mold SM and the side cover SL are arranged in that order and stored in the housing Ho, and the bolts B are tightened as shown in FIG. 11(b) to heat the heater H to a predetermined heating temperature. This softens the skin layer and regenerates the irregularities. The regeneration kit only needs a mold and a heater, and may consist of only a mold and a heater as shown in FIG. 11(c). The regeneration kit in FIG. 11(c) has a heating cable C for heating and a temperature sensor T for measuring the heating temperature.

[0045] <<Method for detecting wear on tactile sensors>> The method for detecting wear of the tactile sensor of the present invention is to use the tactile sensor of the present invention after use to measure the tactile feel of an object having a predetermined surface roughness, and measure the power V of the signal from the strain gauge. signal and the noise power V from the strain gauge noise The method includes a wear detection step of calculating the signal-to-noise ratio SNR using the following formula (V) and determining that the skin layer of the contact of the tactile sensor has worn away if the SNR is less than a predetermined value. SNR(dB)=20log 10 (V signal / V noise ) ···(V) In formula (V), V signal is the amplitude or power of the strain signal when the contact is pressed against a sample with a specified surface roughness and the surface of the sample is repeatedly rubbed; V noise is the amplitude or power of the noise when the contact is pressed against a sample having no surface roughness and the surface of the sample is repeatedly rubbed.

[0046] The predetermined value may be any value greater than or equal to the SNR in the predetermined frequency range to be detected. For example, an SNR of less than 20 dB may be determined to be worn, or less than 15 dB, or less than 10 dB. The frequency range to be detected is preferably 0 to 500 Hz, more preferably 20 to 500 Hz, and even more preferably 20 to 300 Hz.

[0047] <<How to check if the tactile sensor is working properly>> A method for confirming the restoration of a tactile sensor of the present invention, comprising: measuring the strain power V of a tactile sensor restored by the restoration method of the present invention to measure the tactile sensation of an object having a predetermined surface roughness; signal and noise V noise The method includes a regeneration confirmation step of calculating the signal to noise ratio SNR using the following formula (V) and determining that the skin layer of the contact of the tactile sensor has been regenerated if the SNR is equal to or greater than a predetermined value. SNR(dB)=20log 10 (V signal / V noise ) ···(V) In formula (V), V signal is the strain power when the contact is pressed against a sample with a specified surface roughness and the surface of the sample is rubbed repeatedly; V noise is the power of the noise when the contact is pressed against a sample having no surface roughness and the surface of the sample is repeatedly rubbed.

[0048] The predetermined value may be any value greater than or equal to the SNR in the predetermined frequency range to be detected. For example, an SNR of less than 20 dB may be determined to be worn, or less than 15 dB, or less than 10 dB. The frequency range to be detected is preferably 0 to 500 Hz, more preferably 20 to 500 Hz, and even more preferably 20 to 300 Hz.

[0049] <<Tactile Sensor Manufacturing Method>> The method for manufacturing a tactile sensor of the present invention includes a roughness regeneration step in which the surface layer having worn roughness is heated while being in contact with a mold having a structure corresponding to the roughness, thereby regenerating the roughness of the surface layer. The unevenness regeneration process is the same as the method for regenerating a tactile sensor described above.

[0050] <<How to use>> A laminate having a surface irregularity, a skin layer formed of a material that can regenerate the irregularity by heating when the irregularity is worn, and a fat layer having a Shore D hardness lower than that of the skin layer, can be used to manufacture a tactile sensor.

[0051] A laminate having a surface with irregularities and a skin layer formed of a material that can regenerate the irregularities by heating when the irregularities wear away, and a fat layer with a lower Shore D hardness than the skin layer, can be used as a contact in a tactile sensor that slides at a predetermined speed while making contact with a predetermined surface of an object to be measured. [Example]

[0052] The present invention will be specifically explained below with reference to examples, but the present invention is not limited to these examples.

[0053] The width (W), thickness (T), and height (H) of the designed sensor were 50 mm, 30 mm, and 30 mm, respectively, as shown in Figure 12. The sensor's plastic substrate measured 60 mm (W), 40 mm (T), and 5 mm (H). The top of the sensor was curved outward and designed to mimic the shape of a human finger. On the epidermis layer, multiple convex ridges, which mimicked the ridges of the epidermis of a human finger, were formed continuously and parallel to each other along the thickness direction without any gaps. These convex ridges had a cross-section in the shape of a circular arc with a diameter of 2 mm. The height of the epidermis layer was also the same, 2 mm. The uneven epidermis layer covered the upper half of the contact. The fat layer of the designed contact exhibited a modulus of elasticity of approximately 0.17 MPa, while the epidermal layer exhibited a modulus of elasticity close to 0.82 MPa. These values ​​are close to the hardness of a human finger determined by anatomical studies. The modulus of elasticity was calculated from the Shore hardness measured in accordance with JIS K7215. The material used to make the epidermal and fat layers was polyvinyl chloride plastisol (Plastic Worm, Two-El Co., Ltd., Japan), whose hardness can be adjusted by adjusting the mixing ratio of the softener and hardener. The material was prepared so that it begins to gel at temperatures above 70°C and becomes completely liquefied at approximately 180°C. The strain gauge used was a uniaxial type "KFGS-1-120-C1-11 L3M2R" manufactured by Kyowa Electronics Industries Co., Ltd., to mimic the sensory function of human tactile receptors. The strain gauge specifications were 1 mm in length, a gauge factor of 2.10, and a gauge resistance of 119.6 Ω. During the experiment, a strain amplifier (DPM-913B, Kyowa Electronics Industries Co., Ltd., Japan, nominal operating frequency: 10 kHz) was used to condition the signal from the strain gauge. The strain setting was set to 1500 με / V. A data acquisition device (USB-6002, National Instruments Co.) was also used for data collection.

[0054] To fabricate the contacts, we used a plastic mold made from stereolithography resin. This mold was created using a 3D printer (Form 3+, Formlabs, USA) and consisted of four main parts, as shown in Figure 5. First, a fat layer was formed, followed by a skin layer, and then the contacts were manufactured.

[0055] As shown in Figure 13, two types of samples were used, each with alternating concave and convex portions. The ratio of RW (protruding width) to GW (depressed width) was 1:1, and the surface wavelength was defined as RW + GW. As shown in Figure 14, the total length of the sample was 13 cm. One sample had RW = 1 mm and GW = 1 mm, and the other had RW = 0.5 mm and GW = 0.5 mm.

[0056] The contactor was attached to a six-degree-of-freedom articulated robot arm (MyCobot, Elephant Robotics, China) as shown in Fig. 15. The robot arm was designed to move a 13 cm specimen under controlled motion at a constant speed of 30 mm / s. The contact pressure of the contactor against the specimen was approximately 1 N.

[0057] In this experiment, the scanning procedure was stopped when the SNR level fell below 20 dB.

[0058] Although the contact asperities have dimensions larger than a human fingerprint, the contact sensor demonstrates its ability to sense the surface features of the sample, as evidenced by signal data that follows the fundamental equation (2) describing the spatiotemporal propagation of waves:

[0059]

number

[0060] where v is the sliding velocity (mm / s), f is the frequency (Hz), and λ is the surface wavelength of the test piece (mm).

[0061] Using this contactor and two types of sample, tests were performed with surface wavelengths (λ) of 1 mm and 2 mm and a speed of approximately 30 mm / s. Figure 16 shows an example of the raw signal (unit: volts) from the strain gauge and its corresponding amplitude spectrum (unit: dB). The results were as expected, with the frequencies at which the amplitudes were maximum being 30 Hz and 15 Hz, respectively.

[0062] Considering the signal amplification role of the surface roughness, it is thought that the detection of wear on the roughness can be estimated from the signal level. To identify the deterioration of signal detection performance due to wear, the SNR of the voltage signal was used. The SNR was calculated using the following equation (3).

number

[0063] where V signal represents the amplitude or power of the distortion signal, and V noise represents the amplitude or power of the noise.

[0064] The experimental protocol consisted of two distinct steps: the first step was to identify wear damage within the sensor by repeatedly scanning the sample until the SNR level fell below 10 dB, and the second step was to regenerate the asperity ridges in the surface layer, after which the scanning test was repeated and the SNR levels before and after regeneration were compared.

[0065] The experiment was set up to collect data before and after regeneration. Figure 17 shows the SNR at 20 Hz after one rub vs. the number of rubs for a surface wavelength of 2 mm and a speed of 40 mm / s. Figure 17 shows that the SNR of the data collected before regeneration steadily decreases as the number of tactile sensor uses increases. This tendency occurs because the unevenness of the surface of the skin gradually wears away as the number of times the tactile sensor is used increases.

[0066] Signal degradation can be identified by comparative analysis of the power spectrum of the sensor's signal wave, as shown in Figure 18, by comparing the spectrum of a new sensor with the last spectrum obtained before reconditioning. Figure 18(a) shows the amplitude spectrum of the strain signal recorded when a sample with a surface wavelength of 1 mm was rubbed at a speed of 20 mm / s using a contact tip before the sensor's surface roughness had worn away. This amplitude spectrum includes the average and standard error of five measurements. Figure 18(b) shows the amplitude spectrum of the strain signal recorded when a sample with a surface wavelength of 1 mm was rubbed at a speed of 20 mm / s using a contact tip after the sensor's surface roughness had worn away. This amplitude spectrum includes the average and standard error of five measurements. In the scan of a new sensor, the peak frequency of 20 Hz is prominent, and the other frequencies are clearly distinguishable. In the scan of a worn tactile sensor, the peak frequency is barely visible.

[0067] Wear of the surface layer irregularities can be visually identified by direct inspection of the tactile sensor. As shown in Figure 19(a), in contrast to a tactile sensor without wear, the surface irregularities of the worn tactile sensor are significantly flattened. After visually confirming the wear, the surface layer was regenerated. This was performed using the regeneration kit shown in Figure 11. The heating temperature was 80°C for 1 minute. The surface layer was pressed against a mold, and the mold was heated to regenerate the surface layer irregularities. This process was performed without removing the strain gauge from the laminate. Based on evaluation of the sensor's physical condition before and after the regeneration process, it was confirmed that the worn surface irregularities had been successfully restored, as shown in Figure 19(b). For example, comparing the power spectrum of a new sensor with that of a regenerated sensor, as shown in Figure 20, revealed that the sensor had recovered to the same extent as a new sensor. [Industrial Applicability]

[0068] The present invention provides a tactile sensor that can regenerate a worn tactile sensor, a method for manufacturing a tactile sensor, a method for regenerating a tactile sensor, a method for detecting wear in a tactile sensor, and a method for confirming the regeneration of a tactile sensor. [Explanation of symbols]

[0069] TS Tactile Sensor LB contact S epidermal layer F fat layer

Claims

1. A tactile sensor for measuring a tactile sensation caused by a friction stimulus, The tactile sensor has a contactor that slides at a predetermined speed while being brought into contact with a predetermined surface of an object to be measured, and a strain gauge embedded in the contactor; the contact is a laminate having a skin layer and a fat layer, the surface layer has irregularities that come into contact with a predetermined surface of the object to be measured, and is formed of a material that can restore the irregularities by heating when the irregularities are worn; The fat layer has a Shore D hardness lower than that of the epidermis layer.

2. 2. The tactile sensor according to claim 1, wherein the resin composition constituting the skin layer includes a thermoplastic elastomer.

3. 3. The tactile sensor according to claim 2, wherein the resin composition has a glass transition temperature of 50 to 100°C.

4. The tactile sensor according to claim 2 , wherein the resin composition contains polyvinyl chloride.

5. 2. The tactile sensor according to claim 1, wherein the epidermal layer has a modulus of longitudinal elasticity of 0.4 to 2.0 MPa converted from the Shore D hardness or Shore E hardness measured in accordance with JIS K6253, and the fat layer has a modulus of longitudinal elasticity of 0.05 to 0.3 MPa converted from the Shore D hardness or Shore E hardness measured in accordance with JIS K6253.

6. 2. The tactile sensor according to claim 1, wherein the difference expressed as [modulus of longitudinal elasticity converted from the Shore D hardness or Shore E hardness of the epidermis layer measured in accordance with JIS K6253] - [modulus of longitudinal elasticity converted from the Shore D hardness or Shore E hardness of the fat layer measured in accordance with JIS K6253] is 0.1 to 1.95 MPa.

7. 2. The tactile sensor according to claim 1, wherein the average height of the skin layer is 0.5 to 4.0 mm.

8. A method for manufacturing the tactile sensor according to claim 1, A method for manufacturing a tactile sensor, comprising a roughness regeneration step of regenerating the roughness of the skin layer by heating the skin layer having worn roughness while contacting it with a mold having a structure corresponding to the roughness.

9. 9. The method for manufacturing a tactile sensor according to claim 8, wherein the heating temperature in the unevenness regenerating step is 50 to 100°C.

10. A method for regenerating a tactile sensor according to claim 1, comprising: A method for restoring a tactile sensor, comprising a step of restoring the surface layer having worn irregularities by heating the surface layer while bringing it into contact with a mold having a structure corresponding to the irregularities.

11. 11. The method for regenerating a tactile sensor according to claim 10, wherein the heating temperature in the unevenness regeneration step is 50 to 100°C.

12. 2. The method for detecting wear of a tactile sensor according to claim 1, The strain power V from the strain gauge that measured the tactile feel of an object having a specified surface roughness using a used tactile sensor ｓｉｇｎａｌ and the noise power V from the strain gauge ｎｏｉｓｅ So, The signal to noise ratio SNR is calculated using the following formula (V): A method for detecting wear of a tactile sensor, comprising a wear detection step of determining that the skin layer of the contact of the tactile sensor has worn away if the SNR is less than a predetermined value. SNR(dB)=20log １０ (V ｓｉｇｎａｌ / V ｎｏｉｓｅ ) ・・・(V) In formula (V), V ｓｉｇｎａｌ is the power or amplitude of the strain signal when the contactor is pressed against a sample having a predetermined surface roughness and the surface of the sample is repeatedly rubbed; V ｎｏｉｓｅ is the power or amplitude of the noise when the contact is pressed against a sample having no surface roughness and the surface of the sample is repeatedly rubbed.

13. 2. A method for checking regeneration of a tactile sensor according to claim 1, comprising: The power V of the signal from the strain gauge that measures the tactile feel of an object having a predetermined surface roughness using the tactile sensor regenerated by the method of claim 10 is ｓｉｇｎａｌ and the noise power V from the strain gauge ｎｏｉｓｅ So, The signal to noise ratio SNR is calculated using the following formula (V): The method for checking the regeneration of a tactile sensor includes a regeneration checking step of determining that the skin layer of the contact of the tactile sensor has been regenerated if the SNR is equal to or greater than a predetermined value. SNR(dB)=20log １０ (V ｓｉｇｎａｌ / V ｎｏｉｓｅ ) ・・・(V) In formula (V), V ｓｉｇｎａｌ is the power or amplitude of the strain signal when the contactor is pressed against a sample having a predetermined surface roughness and the surface of the sample is repeatedly rubbed; V ｎｏｉｓｅ is the power or amplitude of the noise when the contact is pressed against a sample having no surface roughness and the surface of the sample is repeatedly rubbed.

14. Use of a laminate for manufacturing a tactile sensor, the laminate having a surface irregularity, a skin layer formed of a material that can regenerate the irregularity by heating when the irregularity is worn away, and a fat layer having a Shore D hardness lower than that of the skin layer.

15. A laminate having a surface with irregularities and formed of a material that can regenerate the irregularities by heating when the irregularities wear away, and a fat layer having a Shore D hardness lower than that of the skin layer, is used as a contact in a tactile sensor that slides at a predetermined speed while making contact with a predetermined surface of an object to be measured.

Images