WO2023182012A1 - Estimation device, estimation method, estimation program, and learning model generation device - Google Patents

Abstract

This estimation device includes: a detection unit for detecting electrical characteristics between a plurality of detection points defined in advance on a flexible material that is electrically conductive and of which the electrical characteristics change in response to deformation; and an estimation unit for inputting, into a learning model trained using time-series electrical characteristics that change with deformation of the flexible material and deformation state information indicating the deformation state pertaining to the deformation of the flexible material as training data to use time-series electrical characteristics as input and to output deformation state information, time-series electrical characteristics, detected by the detection unit, of an object subject to estimation that includes the flexible material, and estimating deformation state information indicating the deformation state corresponding to the inputted time-series electrical characteristics.

Description

Estimation device, estimation method, estimation program, and learning model generation device

The present disclosure relates to an estimation device, an estimation method, an estimation program, and a learning model generation device.

Conventionally, changes in shape that occur in objects have been detected and the detection results have been used to estimate the state of the person or object that deforms the object. In terms of detecting changes in shape that occur in objects, it is difficult to detect deformation without inhibiting the deformation of the object. Furthermore, since strain sensors used to detect rigid bodies such as metal deformation are difficult to use for articles, a special detection device is required to detect deformation of an object. For example, a technique is known in which a camera measures the displacement and vibration of an object, obtains a deformation image, and extracts the amount of deformation (for example, see International Publication No. 2017/029905). Furthermore, a technique related to a flexible tactile sensor that estimates the amount of deformation from the amount of light transmitted is also known (for example, see Japanese Patent Application Laid-Open No. 2013-101096).

However, in terms of detecting changes in shape that occur in an object, when detecting the amount of deformation such as displacement of an object using a camera and image analysis method, the system including the camera and image analysis must be large-scale, and the equipment This is not preferable because it leads to an increase in size. On the other hand, when a flexible material is deformable as an object, the amount of deformation changes depending on the magnitude of external stimulation. Furthermore, the amount of restoration may change as the amount of deformation changes. The amount of deformation is thought to depend on the flexibility, which indicates the softness of the flexible material, but evaluation of flexibility depends on the user's subjective sense of softness and hardness when the flexible material is deformed. This is not sufficient to determine the amount of deformation of a flexible material. Therefore, there is room for improvement in identifying the deformation state of flexible materials.

The present disclosure estimates the deformation state of a flexible material by using the electrical properties of the conductive flexible material without using a special detection device.

One aspect of the present disclosure includes:
a detection unit that detects electrical characteristics between a plurality of predetermined detection points on a flexible material that has conductivity and whose electrical properties change according to deformation;
Using the time-series electrical characteristics that change according to the deformation of the flexible material and deformation state information indicating the deformation state related to the deformation of the flexible material as learning data, the time-series electrical characteristics are input, and the The time-series electrical characteristics detected by the detection unit of the estimation target including the flexible material are input to the learning model that has been trained to output deformation state information, and the time-series electrical characteristics are matched to the input time-series electrical characteristics. an estimating unit that estimates deformation state information indicating a deformation state;
This is an estimation device including:

FIG. 1 is a diagram showing the configuration of an estimation device according to an embodiment. FIG. 3 is a diagram showing the arrangement of conductive urethane. 1 is a diagram showing a conceptual configuration of a learning model generation device. It is a figure showing a measuring device. FIG. 3 is a diagram showing an example of electrical characteristics of conductive urethane. FIG. 3 is a diagram showing an example of electrical characteristics of conductive urethane. FIG. 2 is a conceptual diagram of electrical properties of conductive urethane from the viewpoint of value fluctuations. FIG. 3 is a diagram showing the functional configuration of a learning processing section. It is a figure showing an example of the flow of learning processing. FIG. 3 is a diagram showing another functional configuration of the learning processing section. FIG. 3 is a diagram showing an electrical configuration of an estimation device. It is a flowchart which shows the flow of estimation processing. FIG. 2 is a conceptual diagram showing an example of electrical characteristics of conductive urethane. It is a figure which shows the modification of an estimation part.

Hereinafter, embodiments that implement the technology of the present disclosure will be described in detail with reference to the drawings. Note that components and processes that have the same actions and functions are given the same reference numerals throughout the drawings, and overlapping explanations may be omitted as appropriate. Further, the present disclosure is not limited to the following embodiments, and can be implemented with appropriate changes within the scope of the purpose of the present disclosure.

Note that in the present disclosure, a person is a concept that includes at least one of a human body and an object that can provide stimulation to a target object using a physical quantity. In the following description, the term "person" will be used generically as a concept including humans and things, without distinguishing between at least one of a human body and an object. That is, a single human body and an object, as well as a combination of a human body and an object, are collectively referred to as a person.

First, a state estimation process is performed to estimate the deformation state of the flexible material that deforms depending on the state of the imparting side to the flexible material, using a flexible material to which conductivity is applied and the flexible material to which the technology of the present disclosure is applied. explain. It is not impossible to estimate the state of the application side of the flexible material using the estimation result of the deformation state of the flexible material.
The application side state for the flexible material is an example of the state in which the flexible material of the present disclosure is stressed.

<Flexible material>
In the present disclosure, "flexible material" is a concept that includes materials at least partially deformable such as bending, and includes a soft elastic body such as a rubber material, a structure having a skeleton of at least one of a fibrous and a mesh-like structure. It includes a body and a structure in which a plurality of micro air bubbles are scattered inside. An example of these structures includes polymeric materials such as urethane materials. Further, in the present disclosure, a flexible material imparted with conductivity is used. "Flexible material with electrical conductivity" is a concept that includes materials that have electrical conductivity, such as materials in which a conductive material is added to a flexible material to impart electrical conductivity, and materials in which the flexible material has electrical conductivity. including. The flexible material imparting conductivity is preferably a polymeric material such as urethane material. In the following description, as an example of a flexible material imparted with conductivity, a member formed by blending (e.g., internal addition) and infiltration (e.g., impregnation) with a conductive material in all or part of a urethane material will be referred to as " It will be described as "conductive urethane". In the figure, conductive urethane 22 is drawn. The conductive urethane 22 can be formed by mixing a conductive material and infiltrating it, can be formed by combining or infiltrating a conductive material, or can be formed by combining a conductive material and infiltrating it. . For example, if the conductive urethane 22 formed by infiltration (impregnation) has higher conductivity than the conductive urethane 22 formed by blending, it is preferable to form the conductive urethane 22 by infiltration.
Information indicating internal addition and impregnation is an example of information indicating physical properties of the flexible material of the present disclosure.

The conductive urethane 22 has the function of changing its electrical properties depending on the physical quantity given by the state of the supply side. An example of the condition on the application side is a condition indicating a physical action by a person such as compression and expansion to expand and contract the conductive urethane 22, and corresponding examples are conditions such as pressing and pulling. Furthermore, an example of a physical quantity that causes a function that changes electrical characteristics is a stimulus value based on a pressure value indicating a pressure stimulus (hereinafter referred to as pressure stimulus) that deforms a structure such as bending. Pressure stimulation includes application of pressure to a predetermined region and pressure distribution in a predetermined range. In addition, other examples of the physical quantity include stimulation values such as moisture content, which indicates a stimulus (hereinafter referred to as material stimulus) that changes (degrades) the properties of a material due to moisture content, moisture addition, and the like. The conductive urethane 22 changes its deformation state and electrical properties depending on the given physical quantity. An example of a physical quantity representing electrical characteristics is an electrical resistance value. Further, other examples include a voltage value or a current value.

An example of the deformed state is a stretched state in which the conductive urethane 22 expands and contracts. The expanded/contracted state indicates a compressed state on the side where the conductive urethane 22 is compressed by applying a positive pressure stimulus, and an example of a physical quantity indicating the compressed state includes at least one of the amount of compression and the compression ratio. The amount of compression may be the total amount of conductive urethane 22 to be compressed, or may be the maximum value of the amount of sinking when compressed in a predetermined direction. The compression rate may be the ratio of the amount of compression to the total amount of conductive urethane 22, or the ratio of the amount of depression to the thickness of conductive urethane 22 when compressed in a predetermined direction. Further, the side surface where the conductive urethane is stretched by applying a negative pressure stimulus shows a compressed state, and an example of the physical quantity thereof includes at least one of the amount of stretching and the stretching rate. An example of the amount of elongation and the elongation rate is the length of the conductive urethane 22 that is elongated in a predetermined direction. The elongation rate is the ratio of the amount of elongation to the total length of the conductive urethane 22 when it is elongated in a predetermined direction.
The deformed state according to the present embodiment is an example of a stretched state in which the conductive urethane 22 is stretched and contracted by applying stress. The information indicating the compressed state (expansion/contraction state) of the conductive urethane 22 is an example of deformation state information indicating the deformation state of the present disclosure.

By imparting conductivity to a flexible material having a predetermined volume, the conductive urethane 22 exhibits electrical properties (i.e., changes in electrical resistance) when deformed (expanded and contracted) according to a given physical quantity, and its electrical The resistance value can be regarded as a value based on the volume resistance (volume resistance value) of conductive urethane. In the conductive urethane 22, electric paths are interconnected in a complicated manner, and for example, the electric paths expand/contract or expand/contract in response to deformation. In addition, the electrical path may be temporarily disconnected, or a connection may be different from the previous one. Therefore, between positions separated by a predetermined distance (for example, the position of a detection point where an electrode is placed), the conductive urethane 22 can respond to the size and distribution of the applied physical quantity stimulation (pressure stimulation and material stimulation). It exhibits behavior with different electrical characteristics due to deformation and alteration. Therefore, the electrical characteristics change depending on the magnitude and distribution of the stimulus caused by the physical quantity applied to the conductive urethane 22.

Hereinafter, a case will be described as an example in which a pressure stimulus is applied to the conductive urethane 22 and the conductive urethane 22 is compressed as a stimulus that causes a function to change the electrical properties of the flexible material.

Note that by using conductive urethane, there is no need to provide detection points such as electrodes at specific locations. Detection points such as electrodes may be provided at at least two arbitrary locations sandwiching the location where the conductive urethane 22 is compressed (for example, FIG. 1).

Furthermore, in order to improve the detection accuracy of the electrical properties of conductive urethane, more detection points than two detection points may be used. Further, the conductive urethane of the present disclosure may be formed of a conductive urethane group formed by arranging a plurality of conductive urethane pieces, with the conductive urethane 22 shown in FIG. 1 being one conductive urethane piece. In this case, the electrical characteristics may be detected for each of the plurality of conductive urethane pieces, or the electrical properties of the plurality of conductive urethane pieces may be combined and detected. When detecting electrical properties for each of a plurality of conductive urethane pieces, electrical properties such as electrical resistance can be detected for each location. Further, as another example, the detection range on the conductive urethane 22 may be divided, a detection point may be provided for each divided detection range, and the electrical characteristics may be detected for each detection range.

<Estimation device>
Next, an example of an estimation device that estimates the compression state (deformed state) of the conductive urethane 22, that is, the compressed state (deformed state) of the conductive urethane that is compressed (deformed) depending on the state of the application side, will be described.

FIG. 1 shows an example of the configuration of an estimating device 1 that can perform an estimating process for estimating the compressed state of conductive urethane. The estimating device 1 includes an estimating section 5 and is connected to the object 2 so that the electrical characteristics of the conductive urethane 22 are input. The estimation device 1 estimates the compressed state of the conductive urethane 22 contained in the object 2. The estimation device 1 can be realized by a computer equipped with a CPU as an execution device, which will be described later.

The conductive urethane 22 deforms in response to an applied physical quantity (here, pressure stimulation). The pressure stimulus is applied in time series, and shows the compressed state of the conductive urethane 22 that deforms depending on the pressure stimulus (state on the application side). Therefore, the electrical characteristics of the conductive urethane 22 indicate the compressed state of the conductive urethane 22 corresponding to the state of the pressure stimulus (physical quantity) applied to the conductive urethane 22 . Therefore, it is possible to estimate the compressed state of the conductive urethane 22 from the electrical characteristics of the conductive urethane that change over time.

The estimation device 1 estimates and outputs the compression state of the unknown conductive urethane 22 using the learned learning model 51 through an estimation process that will be described later. This makes it possible to estimate the compressed state of the conductive urethane 22 in the object 2 without using special or large equipment or directly measuring the deformation of the conductive urethane 22 contained in the object 2. Become. The learning model 51 is learned by inputting the compressed state of the conductive urethane 22 and the electrical characteristics of the object 2 (that is, the electrical characteristics such as the electrical resistance value of the conductive urethane 22 placed on the object 2). . Learning of the learning model 51 will be described later.

Note that the conductive urethane 22 can be placed on the flexible member 21 to form the object 2 (FIG. 2). The object 2 constituted by the member 21 on which the conductive urethane 22 is disposed includes an electrical property detection section 76 . The conductive urethane 22 may be placed on at least a portion of the member 21, and may be placed inside or outside. Further, the conductive urethane 22 may be arranged so that its compressed state can be estimated, and for example, it may be arranged so that it can be contacted directly or indirectly with a person, or both.
The object 2 including the electrical property detection section 76 is an example of the detection section of the present disclosure. Therefore, although the details will be described later, the object 2 including the electrical property detecting section 76 has an electrical property that indicates a change in the volume resistance of the conductive urethane 22, which changes from the first tendency to the first tendency according to the compressed state (stretching state). It is possible to detect electrical characteristics that change in two trends.

FIG. 2 shows an example of the arrangement of the conductive urethane 22 on the object 2. The conductive urethane 22 may be formed so as to completely fill the inside of the member 21, as shown in the AA cross section of the object 2 as the object cross section 2-1. Further, as shown in object cross section 2-2, the conductive urethane 22 may be formed on one side (surface side) inside the member 21, and as shown in object cross section 2-3, the conductive urethane 22 may be formed on one side (surface side) inside the member 21. The conductive urethane 22 may be formed on the other side (back side) inside. Furthermore, as shown in the object cross section 2-4, a conductive urethane 22 may be formed in a part of the inside of the member 21. Further, as shown in the object cross section 2-5, the conductive urethane 22 may be placed separately on the outside of the surface side of the member 21, and as shown in the object cross section 2-6, the conductive urethane 22 may be placed on the other side ( It may be placed outside the back side). When the conductive urethane 22 is disposed outside the member 21, the conductive urethane 22 and the member 21 may be simply laminated, or the conductive urethane 22 and the member 21 may be integrated by bonding or the like. Note that even when the conductive urethane 22 is disposed outside the member 21, the flexibility of the member 21 is not inhibited because the conductive urethane 22 is a urethane member having conductivity.

As shown in FIG. 1, the electrical properties of the electrically conductive urethane 22 (i.e., the volume resistivity value, which is an electrical resistance value) are detected by signals from at least two detection points 75 placed apart from each other. ) is detected. In the example of FIG. 1, detection set #1 is shown, which detects electrical characteristics (time-series electrical resistance values) using signals from two detection points 75 placed diagonally on the conductive urethane 22. There is. Note that the number and arrangement of the detection points 75 are not limited to the positions shown in FIG. good. Note that the electrical properties of the conductive urethane 22 can be determined by connecting an electrical property detection section 76 that detects electrical properties (for example, a volume resistivity value, which is an electrical resistance value) to a detection point 75, and using the output thereof.

In this embodiment, since the conductive urethane 22 is used as a sensor, for example, when a person is present, the sense of discomfort given to the person is extremely small compared to conventional sensors. Therefore, it is possible to perform measurement and estimate the compressed state of the conductive urethane 22 at the same time without damaging the condition of the person on the application side (for example, contact with the person) during measurement. This is an advantage compared to conventional sensors that perform measurement and estimation separately, and is particularly advantageous in estimation based on long-term measurement and evaluation that follows time-series changes.

In order to simplify the explanation, hereinafter, the conductive urethane 22 formed so that the inside of the member 21 is completely filled with conductive urethane will be applied to the object 2 (object cross section 2 in FIG. 2). -1). Therefore, although the conductive urethane 22 included in the object 2 will be described below, the object 2 and the conductive urethane 22 can be treated as the same thing.

The estimation unit 5 is connected to the conductive urethane 22 included in the object 2, and uses a learning model 51 to estimate the compressed state of the conductive urethane 22 based on the electrical characteristics that change according to the deformation of the conductive urethane 22. This is a functional unit that estimates. Specifically, time-series input data 4 representing the magnitude of electrical resistance (electrical resistance value, etc.) in the conductive urethane 22 is input to the estimation unit 5 . Input data 4 corresponds to state data 3 indicating the compressed state of conductive urethane 22 . For example, when a person contacts the object 2, the person contacts the object 2 in a predetermined state such as a posture, and corresponding to the state, a pressure stimulus is applied as a physical quantity to the conductive urethane 22 constituting the object 2, and the conductive urethane 22 is applied as a physical quantity. The electrical properties of the polyurethane 22 change. Therefore, the electrical characteristics of the conductive urethane 22 that change over time as indicated by the input data 4 correspond to the compressed state of the conductive urethane 22 that changes in accordance with the state of the application side to the conductive urethane 22. Furthermore, the estimation unit 5 outputs output data 6 representing the compressed state of the conductive urethane 22 corresponding to the electrical characteristics of the conductive urethane 22 that change over time as an estimation result using the trained learning model 51. .

The learning model 51 is a model that has undergone learning to derive output data 6 representing the compressed state of the conductive urethane 22 from the electrical resistance (input data 4) of the conductive urethane 22 that changes due to applied pressure stimulation. The learning model 51 is, for example, a model that defines a trained neural network, and is expressed as a set of information on the weights (strengths) of connections between nodes (neurons) forming the neural network.

Here, as described above, the conductive urethane 22 exhibits behavior in response to changes (deformations) such as expansion/contraction, expansion/contraction, temporary disconnection, new connections, etc. of the electrical paths, which are interconnected in a complex manner. . As a result, the conductive urethane 22 behaves with electrical characteristics depending on the compressed state corresponding to the applied pressure stimulus. However, deformations due to changes such as expansion/contraction, expansion/contraction, temporary disconnection, and new connections (recombinations) of electrical paths do not occur regularly. Therefore, in this embodiment, as data regarding the deformation (compression) of the conductive urethane 22, the electrical properties (electrical resistance) of the conductive urethane 22 that change in response to pressure stimulation (here, compression of the conductive urethane 22) are used. A learning model 51 capable of estimating the compressed state of the conductive urethane 22 is generated by a learning process.

The estimation process in the estimation device 1 uses the trained learning model 51 to estimate and output the compression state of the unknown article (estimated object). The learning model 51 is trained using learning data including the electrical properties of the conductive urethane 22 and the compressed state of the conductive urethane 22 contained in the object 2 .

<Learning process>
Next, a learning process for generating the learning model 51 will be explained. In the learning process, learning is performed using, as learning data, electrical characteristics of the conductive urethane 22 (input data 4) labeled with compressed state information (state data 3) indicating the compressed state of the conductive urethane 22. In the learning process, a learning data collection process and a learning model generation process are executed.

FIG. 3 shows the conceptual configuration of a learning model generation device that generates the learning model 51. The learning model generation device includes a learning processing section 52. The learning model generation device can be configured to include a computer equipped with a CPU (not shown), and is executed as a learning processing unit 52 by learning data collection processing and learning model generation processing executed by the CPU to generate the learning model 51. .

<Learning data collection processing>
Next, a learning data collection process for collecting learning data used when generating the learning model 51 will be described. In the learning data collection process, the learning processing unit 52 measured the electrical characteristics (for example, electrical resistance value) of the conductive urethane 22 in time series using the state data 3 representing the compressed state (compression degree) of the conductive urethane 22 as a label. A large amount of input data 4 is collected as learning data. Therefore, the learning data includes a large set of input data 4 indicating electrical characteristics and state data 3 indicating the compressed state of the conductive urethane 22 corresponding to the input data 4.

In the learning data collection process, electrical characteristics (for example, electrical resistance value) that change depending on the compressed state of the conductive urethane 22 are acquired in time series. Next, state data 3 is assigned as a label to the acquired time-series electrical characteristics (input data 4), and the set of state data 3 and input data 4 reaches a predetermined number or a predetermined time. Repeat the process until. A set of these state data 3 and the electrical characteristics of the conductive urethane 22 in time series (input data 4) becomes learning data. The state data 3 in the learning data is stored in a memory (not shown) so as to be treated as output data 6 indicating the compressed state of the conductive urethane 22 for which the estimation result is correct in the learning process described later.

Note that the learning data may be associated with time series information by adding information indicating the measurement time to each of the electrical resistance values (input data 4) of the conductive urethane 22. In this case, for the period determined as the compressed state of the conductive urethane 22, information indicating measurement time may be added to a set of time-series electrical resistance values in the conductive urethane 22 to associate the time-series information.

Note that the electrical characteristics detected in the conductive urethane 22 (temporal characteristics based on time-series electrical resistance value data) can be regarded as a characteristic pattern related to the compressed state of the conductive urethane 22 (details will be described later).

FIG. 4 shows an example of a measuring device 8 that measures electrical characteristics used as learning data. The measuring device 8 is an example of a measuring device that measures electrical characteristics that change depending on the deformation of the conductive urethane 22. Note that the measuring device 8 is capable of repeatedly applying pressure stimulation.

In the measuring device 8, a pressure applying part 83 for applying pressure stimulation to the conductive urethane 22 is attached to a fixed part 82 fixed to a base 81. The pressure application unit 83 includes a pressure application main body 83A, an arm 83B that is extendable and retractable from the pressure application main body 83A, and a distal end portion 83C attached to the distal end of the arm 83B. The pressure application main body 83A is fixed to the fixed part 82, and the arm 83B is expanded and contracted in response to an input signal, and the tip 83C is moved in a predetermined direction (arrow Z direction and the opposite direction). This allows the pressing member 84 to come into contact with the conductive urethane 22 installed on the base 81, press it after contact, or separate from the conductive urethane 22.

Conductive urethane 22 is placed on base 81 . Note that a flexible member 21 such as urethane may be placed on the conductive urethane 22. A pressing member 84 having a predetermined shape is attached to the tip portion 83C of the pressure applying portion 83. The conductive urethane 22 is arranged so that the pressing member 84 attached to the tip 83C of the pressure applying part 83 can at least come into contact with the conductive urethane 22. In this embodiment, as an example of the pressing member 84 having a predetermined shape, a pressing member 84 having a curved tip (for example, a part of a sphere) is used. The pressing member 84 is a member that applies pressure stimulation to the conductive urethane 22 with a predetermined pressure. Note that the shape of the pressing member 84 may be any of a rectangular, trapezoidal, circular, elliptical, or polygonal cross-sectional shape, or may be any other shape.

The pressure applying section 83 operates to press (compress) the conductive urethane 22 with the pressing member 84 when the arm 83B extends.

The pressure applying main body 83A includes a force sensor 85 having a function of detecting force in six axial directions, for example. The force sensor 85 has a function of detecting the pressing state of the pressing member 84 against the conductive urethane 22 and a function of detecting the pressure applied to the conductive urethane 22 from the detected force. This force sensor 85 can detect the force (physical quantity) of the pressing member 84 against the conductive urethane 22 in a time series, and can detect the pressure applied to the conductive urethane 22 in a time series. Note that when testing only the compressed state (deformed state), the force sensor 85 can be omitted.

The measuring device 8 includes a pressure applying section 83 and a controller 80 connected to a force sensor 85. The controller 80 includes a CPU (not shown), which controls the pressure applying unit 83, applies pressure stimulation to the object 2, and acquires time-series electrical characteristics due to the pressure stimulation to the conductive urethane 22. and remember.

Specifically, the controller 80 controls the pressure application unit 83 to apply and release pressure stimulation to the conductive urethane 22 by reciprocating the arm 83B to extend and contract it. Further, the controller 80 acquires the electrical characteristics of the conductive urethane 22 in synchronization with the application and release of pressure stimulation to the conductive urethane 22. Therefore, the measuring device 8 can acquire, as one of the learning data, electrical characteristics related to the deformation of the conductive urethane 22 that is intermittently (for example, periodically) deformed in a time series.

Additionally, the controller 80 can acquire data indicating the deformed state, that is, the compressed state, of the conductive urethane 22 in synchronization with the application and release of pressure stimulation to the conductive urethane 22. An example of data indicating the compressed state is the distance that the pressing member 84 sinks after contacting the conductive urethane 22. Specifically, the controller 80 acquires data indicating the compressed state by measuring the distance that the pressing member 84 sinks into the conductive urethane 22 in response to pressure stimulation. Therefore, the measuring device 8 can acquire data indicating the compression state as other learning data. The distance that the conductive urethane 22 sinks in response to pressure stimulation, which is data indicating the compressed state, can be used as data indicating the degree of compression. As described above, the data indicating the degree of compression includes at least one of the amount of compression and the compression ratio.

Here, the electrical characteristics of the conductive urethane 22, which exhibits complicated behavior in a compressed state, will be explained.
5A and 5B show an example of electrical characteristics when the conductive urethane 22 is compressed (deformed). 5A and 5B show electrical characteristics of the conductive urethane 22 corresponding to compressed states compressed by different pressure stimuli. The conductive urethane 22 shown in FIGS. 5A and 5B is a flexible material in which at least a portion of the urethane material is infiltrated (for example, impregnated) with a conductive material. It has been confirmed that the following explanation is also applicable to a flexible material in which a conductive material is blended (for example, internally added) into at least a portion of a urethane material.

In FIG. 5A, the electrical characteristics when the conductive urethane 22 is compressed with the first amount of sinking are shown as electrical characteristics 41, and in FIG. The electrical properties when the polyurethane 22 is compressed are shown as electrical properties 44. Furthermore, in FIGS. 5A and 5B, the time when the pressure stimulation is applied is shown as P1, and the time when the applied pressure stimulation is canceled is shown as P2. As shown in FIGS. 5A and 5B, the electrical characteristics of the conductive urethane 22 vary depending on the magnitude of the applied pressure stimulus. These electrical characteristics include characteristics related to the compressed state indicated by the degree of compression (in this case, the amount of subsidence).

As shown in FIGS. 5A and 5B, the electrical characteristics when the conductive urethane 22 is compressed change from an upward trend where the electrical resistance value gradually increases during the compression process to a downward trend where the electrical resistance value gradually decreases. It has the characteristic that the electrical resistance value converges to the value corresponding to the size of .

Specifically, the first feature is that it has an electrical characteristic portion 42 that switches from an upward trend in which the electrical resistance value gradually increases to a downward trend in which the electrical resistance value gradually decreases when compressed. The second feature is that there is a difference 43 in the electrical resistance value that is on a downward trend compared to the electrical resistance value at the beginning of compression. The third feature is that the difference 43 in electrical resistance values increases as the degree of compression (in this case, the amount of sinking) increases.

In other words, the electrical characteristics when the conductive urethane 22 is compressed tend to have the first to third characteristics. These characteristics are due to the tendency of the electrical resistance value to increase due to the expansion and contraction of at least a part of the electrical path that forms conductivity due to compression, and the tendency for electrical resistance to increase due to the cutting and recombination of the electrical path in a certain compression state (compression degree). It is presumed that this is caused by a phenomenon in which the electrical resistance value switches to a downward trend. However, deformations of electrical paths (expansion/contraction, expansion/contraction, temporary disconnection, new connections (recombination), etc.) do not occur regularly. Therefore, in this embodiment, the degree of compression, which is the compressed state, is estimated from the first to third characteristics described above. Specifically, with the first and second characteristics as conditions, the physical quantity based on the third characteristic is quantified as a tendency for the difference in electrical resistance values to increase, and is defined as the degree of compression. That is, the degree of compression corresponds to the degree of tendency for the difference in electrical resistance values to increase.

Note that the timing of the electrical characteristics when deriving the difference in electrical resistance values can be set to predetermined timings before and after the electrical characteristics portion 42. For example, on the one hand, the timing immediately before shifting to an upward trend, such as at the beginning of compression, or when the electrical characteristics are stable before compression (for example, the rate of change in electrical resistance value is less than a predetermined value for an upward trend). The following timings are applicable. On the other hand, the electrical characteristics become stable (for example, the rate of change of the electrical resistance value is less than a predetermined value for a downward trend) after a predetermined time has elapsed from the electrical characteristic portion 42, or after switching to a downward trend. applicable timing.

The upward trend and downward trend described above are examples of the first trend and second trend of the present disclosure. The electrical characteristic portion 42 is an example of a changed portion of the present disclosure. The difference 43 in electrical resistance values is an example of the difference in electrical property values of the present disclosure. Further, the electrical properties of the conductive urethane 22 are an example of the changing properties of the present disclosure.

FIG. 6 is a conceptual diagram showing an example of electrical characteristics that exhibit complicated behavior depending on the compressed state of the conductive urethane 22. The example shown in FIG. 6 shows characteristics that appear when applying and releasing pressure stimulation to the conductive urethane 22 in a compressed state is repeated while changing the degree of compression. In FIG. 6, a curve Erx indicates the correspondence between the degree of compression in the compressed state and the value of electrical characteristics (electrical characteristic difference) indicating the difference in electrical resistance value at the degree of compression. For example, the electrical property difference (difference in electrical resistance value) Er1 and the degree of compression Pw1 have a corresponding relationship. Note that the degree of compression indicates the distance of pressure stimulation (the distance that the conductive urethane 22 sinks due to pressure) applied to the conductive urethane 22 in a predetermined direction (for example, the normal direction to the contact surface of the conductive urethane 22). The difference in the value of electrical properties (electrical resistance value) is the difference between the electrical resistance value at the beginning of compression and the electrical resistance value at a predetermined degree of compression, specifically, the difference between the electrical resistance value at the beginning of compression and the tendency of the electrical resistance value to increase. It shows the difference from the electrical resistance value when it switches to a downward trend and a predetermined period of time has elapsed.

As shown in FIG. 6, the electrical properties of the conductive urethane 22 are such that the difference in electrical properties (difference in electrical resistance) of the conductive urethane 22 tends to increase as the degree of compression increases. . Therefore, using the above-mentioned first to third characteristics such as the difference in the electrical property values of the conductive urethane 22, the difference in the electrical property values (difference in the electrical resistance value) in time series can be used to determine the difference in the electrical property values of the conductive urethane 22. It is possible to estimate the degree of compression.

The compression state information (state data 3) indicating the compression state of the conductive urethane 22 can be set by the compression degree described above. Therefore, by deriving the above-mentioned characteristics shown in the electrical characteristics of the object 2, we can obtain the conductivity for the time-series electrical characteristics regarding the deformation (compression) of the conductive urethane 22 included in the object 2 as another learning data. It becomes possible to correlate the compressed states of the polyurethane 22.

Therefore, a set of information indicating each of the time-series electrical characteristics of the object 2 (that is, the conductive urethane 22) and the compressed state of the conductive urethane 22 with respect to the electrical characteristics becomes the learning data. An example of the learning data is shown in Table 1 below. Table 1 is a data set in which time-series electrical resistance value data (r) is associated with degree of compression (Pw) indicating state data (R) indicating the compressed state, as learning data regarding the compressed state of the conductive urethane 22. This is an example.

Table 1 shows the correspondence between the state data (R) and the degree of compression indicating the specific state of compression. That is, the state data (R) may be associated with the degree of compression (Pw), which is data indicating the characteristics appearing in the electrical characteristics described above. Therefore, since the compressed state of the conductive urethane 22 characteristically appears in the time-series electrical characteristics of the object 2 (conductive urethane 22), it functions effectively in the learning process.

<Learning model generation process>
Next, learning model generation processing will be explained. The learning model generation device shown in FIG. 3 generates a learning model 51 using the above-mentioned learning data through learning model generation processing in the learning processing unit 52.

FIG. 7 is a diagram showing the functional configuration of the learning processing unit 52, that is, the functional configuration of the CPU (not shown) regarding the learning model generation process executed by the learning processing unit 52. A CPU (not shown) of the learning processing unit 52 operates as a functional unit of the generator 54 and the arithmetic unit 56. The generator 54 has a function of generating an output in consideration of the context of the input electric resistance values acquired in time series.

The learning processing unit 52 uses, as learning data, the above-mentioned input data 4 (for example, electrical resistance value) and state data indicating the compressed state of the conductive urethane 22 when stimulation (compression) is applied to the conductive urethane 22. A large number of sets of output data 6, which is 3, are held in a memory (not shown).

The generator 54 includes an input layer 540, an intermediate layer 542, and an output layer 544, and constitutes a known neural network (NN). Since the neural network itself is a well-known technology, a detailed explanation will be omitted, but the intermediate layer 542 includes a large number of node groups (neuron groups) having inter-node connections and feedback connections. The data from the input layer 540 is input to the intermediate layer 542, and the data of the calculation result of the intermediate layer 542 is output to the output layer 544.

The generator 54 is a neural network that generates output data 6A as data representing the compressed state or data close to the compressed state from the input data 4 (for example, electrical resistance value). The generated output data 6A is data obtained by estimating the compressed state in which the conductive urethane 22 is stimulated based on the input data 4. The generator 54 generates output data representing a state close to a compressed state from the input data 4 inputted in time series. By learning using a large number of input data 4, the generator 54 can generate output data 6A that is close to the compressed state when the object 2, that is, the conductive urethane 22 is stimulated. In another aspect, the electric characteristics that are the input data 4 inputted in time series are captured as a pattern, and by learning this pattern, when the object 2, that is, the conductive urethane 22 is stimulated and compressed, It becomes possible to generate output data 6A that is close to a compressed state.

The computing unit 56 is a computing unit that compares the generated output data 6A and the output data 6 of the learning data and computes an error in the comparison result. The learning processing unit 52 inputs the generated output data 6A and the output data 6 of the learning data to the arithmetic unit 56. In response to this, the calculator 56 calculates the error between the generated output data 6A and the output data 6 of the learning data, and outputs a signal indicating the result of the calculation.

The learning processing unit 52 performs learning of the generator 54, which tunes the weight parameter of the connection between nodes, based on the error calculated by the calculator 56. Specifically, the weight parameter of the connection between the nodes of the input layer 540 and the hidden layer 542 in the generator 54, the weight parameter of the connection between the nodes in the hidden layer 542, and the node of the hidden layer 542 and the output layer 544. Each of the weight parameters of the connections between the two is fed back to the generator 54 using a method such as gradient descent or backpropagation. That is, with the output data 6 of the learning data as a target, the connections between all nodes are optimized so as to minimize the error between the generated output data 6A and the output data 6 of the learning data.

Note that the generator 54 may use a recurrent neural network that has a function of generating an output by considering the context of the time-series input, or may use other methods.

The learning processing unit 52 generates the learning model 51 using the above-mentioned learning data through learning model generation processing. The learning model 51 is expressed as a collection of information on weight parameters (weights or strengths) of connections between nodes as a learning result, and is stored in a memory (not shown).

FIG. 8 shows an example of the flow of the learning process executed in the learning processing section 52. The learning process is performed by the CPU (not shown) in the learning processing section 52 described above.

In step S110, the electrical characteristics (input data 4) of the conductive urethane 22 are acquired. In the next step S111, first, the electrical characteristics (input data 4) of the conductive urethane 22 are analyzed and the degree of compression representing the above-mentioned characteristics is derived, thereby obtaining state data 3 indicating the compressed state of the conductive urethane 22. get. In this step S111, input data 4 representing the electrical characteristics of the conductive urethane 22 is associated with state data 3 indicating the compressed state of the conductive urethane 22 as a result of the analysis, and the state data 3 (degree of compression) is labeled. A set of input data 4 (electrical resistance) is acquired as learning data. Next, in step S112, a learning model 51 is generated using the acquired learning data. That is, a set of information on weight parameters (weights or strengths) of connections between nodes of the learning results learned using a large amount of learning data as described above is obtained. Then, in step S114, data expressed as a set of information on weight parameters (weights or strengths) of connections between nodes as a learning result is stored as a learning model 51.

The estimation device 1 uses a trained generator 54 (that is, data expressed as a set of information on weight parameters of connections between nodes as a result of learning) as a learning model 51. By using the sufficiently learned learning model 51, the compressed state of the conductive urethane 22 (which may be the object 2) can be determined from the time-series electrical characteristics of the conductive urethane 22 (for example, the characteristics of the electrical resistance value that changes over time). It is not impossible to estimate.
The process in step S110 is an example of a process executed by the acquisition unit of the present disclosure. Step S112 is an example of a process executed by the learning model generation unit of the present disclosure.

<PRC>
By the way, as described above, the conductive urethane 22 has electrical paths that are interconnected in a complicated manner, and changes (deformation) such as expansion and contraction of the electrical paths, expansion and contraction, temporary disconnection, and new connections, as well as changes in the properties of the material. (alteration). As a result, the conductive urethane 22 behaves with different electrical properties depending on the applied stimulus (for example, pressure stimulus). This means that the conductive urethane 22 can be treated as a reservoir that stores data regarding the deformation of the conductive urethane 22. That is, the estimation device 1 can apply the conductive urethane 22 to a network model (hereinafter referred to as PRCN) called physical reservoir computing (PRC). That is, the above-mentioned learning model 51 can be generated by learning using a network based on reservoir computing using the conductive urethane 22 as a reservoir. Since PRC and PRCN themselves are known techniques, a detailed explanation will be omitted, but PRC and PRCN are suitable for estimating information regarding deformation and deterioration of the conductive urethane 22.

FIG. 9 shows an example of the functional configuration of the learning processing unit 52 to which PRCN is applied. The learning processing unit 52 to which PRCN is applied includes an input reservoir layer 541 and an estimation layer 545. The input reservoir layer 541 corresponds to the conductive urethane 22 included in the object 2 . That is, the learning processing unit 52 applying PRCN handles the object 2 including the conductive urethane 22 as a reservoir for storing data regarding the deformation and alteration of the object 2 including the conductive urethane 22 for learning. The conductive urethane 22 has electrical properties (electrical resistance values) that correspond to each of various stimuli, functions as an input layer for inputting electrical resistance values, and also provides data regarding the deformation (and alteration) of the conductive urethane 22. It functions as a reservoir layer. Since the conductive urethane 22 is deformed (compressed) and outputs different electrical characteristics (input data 4) according to the applied pressure stimulus depending on the state of the application side, the estimation layer 545 calculates the applied conductive urethane 22 It is possible to estimate the unknown compression state from the electrical resistance value. Therefore, in the learning process in the learning processing unit 52 to which PRCN is applied, the estimation layer 545 may be learned.

As described above, in the estimation device 1, by using the trained learning model 51 generated by the method exemplified above, if the sufficiently learned learning model 51 is used, the conductive It is not impossible to estimate the compressed state of the polyurethane 22.
Note that the estimation device 1 is an example of an estimation unit and an estimation device of the present disclosure. The target object 2 and the conductive urethane 22 are examples of the detection unit of the present disclosure.

<Configuration of estimation device>
Next, an example of a specific configuration of the estimation device 1 described above will be further explained.
FIG. 10 shows an example of the electrical configuration of the estimation device 1. The estimation device 1 shown in FIG. 10 is configured to include a computer as an execution device that executes processes to realize the various functions described above. The estimation device 1 described above can be realized by causing a computer to execute a program representing each of the functions described above.

A computer functioning as the estimation device 1 includes a computer main body 100. The computer main body 100 includes a CPU 102, a RAM 104 such as a volatile memory, a ROM 106, an auxiliary storage device 108 such as a hard disk drive (HDD), and an input/output interface (I/O) 110. These CPU 102, RAM 104, ROM 106, auxiliary storage device 108, and input/output I/O 110 are connected via a bus 112 so as to be able to exchange data and commands with each other. Further, the input/output I/O 110 is connected to a communication section 114 for communicating with an external device, an operation display section 116 such as a display or a keyboard, and a detection section 118. The detection unit 118 functions to acquire input data 4 (electrical characteristics such as electrical resistance values in time series) from the object 2 including the conductive urethane 22 . That is, the detection unit 118 can acquire the input data 4 from the electrical property detection unit 76 connected to the detection point 75 on the conductive urethane 22 . Note that the detection unit 118 may be connected via the communication unit 114.
The operation display section 116 is an example of an output section of the present disclosure.

A control program 108P for causing the computer main body 100 to function as the estimation device 1 as an example of the estimation device of the present disclosure is stored in the auxiliary storage device 108. The CPU 102 reads the control program 108P from the auxiliary storage device 108, expands it to the RAM 104, and executes the process. Thereby, the computer main body 100 that has executed the control program 108P operates as the estimation device 1.

Note that the auxiliary storage device 108 stores a learning model 108M including the learning model 51 and data 108D including various data. The control program 108P may be provided on a recording medium such as a CD-ROM.

<Estimation processing>
Next, the estimation processing in the estimation device 1 implemented by a computer will be further explained. In the estimation process, the compressed state of the conductive urethane 22 is estimated using the learning model 51 described above.

FIG. 11 shows an example of the flow of estimation processing by the control program 108P executed by the computer main body 100. The estimation process shown in FIG. 11 is executed by the CPU 102 when the computer main body 100 is powered on. The CPU 102 reads the control program 108P from the auxiliary storage device 108, expands it to the RAM 104, and executes processing.
The control program 108P is an example of the estimation program of the present disclosure. Further, the processing by the control program 108P executed by the CPU 102 is an example of the processing by the estimation method of the present disclosure.

First, the CPU 102 reads the learning model 51 from the learning model 108M in the auxiliary storage device 108, and develops it in the RAM 104, thereby acquiring the learning model 51 (step S200). Specifically, by deploying a network model (see FIGS. 7 and 9) representing connections between nodes using weight parameters expressed as a learning model 51 in the RAM 104, connections between nodes based on weight parameters are realized. A learning model 51 is constructed.

Next, the CPU 102 acquires unknown input data 4 (electrical characteristics), which is a target for estimating the compressed state of the conductive urethane 22, in time series via the detection unit 118 (step S202). Next, the CPU 102 uses the learning model 51 to estimate the output data 6 (compression degree that is an unknown compression state) corresponding to the acquired input data 4 (step S204). Then, the CPU 102 outputs the estimation result output data 6 (compression degree, which is the compression state) via the communication unit 114 (step S206), and ends this processing routine.

In this way, according to the estimation device 1, the compressed state can be estimated from the electrical resistance value of the conductive urethane 22. Specifically, the estimation device 1 is capable of estimating the compressed state of the conductive urethane 22 from the input data 4 (electrical characteristics) that changes according to the pressure stimulation applied to the conductive urethane 22. That is, it becomes possible to estimate the compressed state of the conductive urethane 22 without using a special device or a large device or directly measuring the deformation of the flexible member.

Note that in step S206, it is also possible to output caution information indicating caution regarding the conductive urethane 22. For example, a message that is intuitive to the user may be output as a compressed state (degree of compression) to the communication unit 114 or the operation display unit 116. Specifically, if the compression state (degree of compression) is less than or equal to a predetermined threshold value for determining the compression state (degree of compression), a message indicating that the compression state of the conductive urethane 22 has a margin is output as caution information. On the other hand, if the threshold value is exceeded, a message indicating that there is no margin in the compression state and that it is preferable to return to the compression state below the threshold value is output as caution information. By outputting the caution information in this way, the user can check the compressed state with intuitive caution information without compressing while checking the flexibility of the conductive urethane 22 (object 2). becomes. This caution information may be obtained by measuring in advance the compression state (degree of compression) that indicates a destructive state that may lead to cutting of the electrical path in the conductive urethane 22, and using the measured value as a threshold value. Further, the compression state (degree of compression) in which the electrical resistance value changes from the upward trend to the downward trend described above may be used as the threshold value.
The caution information indicating the caution to the conductive urethane 22 can also be considered as information indicating the caution state of the conductive urethane 22, and the caution information functions as an example of the caution information of the present disclosure.

The estimation process shown in FIG. 11 described above is an example of the process executed by the estimation method of the present disclosure.

As explained above, according to the present disclosure, it is possible to estimate the compressed state of the conductive urethane 22 from the electrical characteristics of the conductive urethane 22 without using a special detection device. Furthermore, by outputting the estimation result of the compression state of the estimated object, that is, the conductive urethane 22, it is possible to estimate the compression state including the flexibility of the conductive urethane 22 that functions as a sensor included in the object 2. It becomes possible.

<Deformed state>
In the embodiment described above, the compressed state, that is, the deformed state in which the conductive urethane 22 is compressed by applying a positive pressure stimulus to the conductive urethane 22 has been described, but the technology of the present disclosure is limited to the compressed state. It's not something you can do. The technique of the present disclosure can also be applied to an elongated state in which the conductive urethane 22 is stimulated with negative pressure, for example, the conductive urethane 22 is stimulated to stretch by pulling or the like.

FIG. 12 shows an example of electrical characteristics when the conductive urethane 22 is expanded (deformed).
In FIG. 12, the electrical characteristics of the conductive urethane 22 in a stretched state in which the conductive urethane 22 is stretched by separating both ends of the conductive urethane 22 in opposite directions with a predetermined force are shown as electrical characteristics 45. Note that the example shown in FIG. 12 shows the relationship between the electrical characteristic value (electrical resistance value) and the time when both ends of the conductive urethane 22 are continuously separated by a predetermined force. Further, FIG. 12 shows an example in which, as the conductive urethane 22, a flexible material in which at least a portion of the urethane material is infiltrated (for example, impregnated) with a conductive material is used. It has been confirmed that the following explanation is also applicable to a flexible material in which a conductive material is blended (for example, internally added) into at least a portion of a urethane material.

As shown in FIG. 12, the electrical property 45 is such that the electrical property (electrical resistance value) increases in accordance with the magnitude of the applied negative pressure stimulus (ie, extension). Therefore, the electrical characteristics include a characteristic 46 related to the state of elongation, which is indicated by the degree of elongation (in this case, the amount of elongation).

Furthermore, in the example shown in FIG. 12, the electrical property 45 has a characteristic that the electrical resistance value switches from a first upward trend in which the electrical resistance value gradually increases during the elongation process to a second upward trend. This characteristic is presumed to be due to the phenomenon that the degree of the tendency for the electrical resistance value to rise changes due to the cutting of at least a portion of the electrical path due to elongation. Such cutting of the electrical path may cause the conductive urethane 22 to be disconnected. However, changes in the degree of increasing tendency of the electrical resistance value do not occur regularly. Therefore, as described above, the degree of elongation, which is the elongated state, can be estimated from the features included in the electrical characteristics. Specifically, the physical quantity at the time of elongation due to the feature is quantified and defined as the degree of elongation. That is, the degree of elongation corresponds to a degree that tends to increase as the amount or ratio of elongation of the conductive urethane 22 increases.

It is also possible to derive the degree of elongation in advance through experiments or the like, which indicates a state of elongation that may cause disconnection of the conductive urethane 22. By estimating the degree of elongation indicating a state of elongation that may cause such discontinuity and outputting it as caution information if applicable, it is also possible to avoid discontinuity of the conductive urethane 22.

In addition, although the case where the characteristic at the time of elongation is switched from the first upward trend to the second upward trend has been described above, it is not limited to having a portion where the electrical characteristics change significantly. For example, the degree of elongation indicating a state of elongation that may cause disconnection of the conductive urethane 22 is derived in advance through experiments (elongation degree Pu1 shown in FIG. 12), and the degree of elongation is set as a boundary from the first upward trend. , the electrical characteristics may switch to a second upward trend.

<Modified example>
The estimation unit 1 described above can be configured according to functions corresponding to the features included in the electrical characteristics.

FIG. 13 shows a modification of the configuration of the estimation unit 1. The estimation unit 1 includes a compression state analysis unit, a comparison determination unit, and a threshold storage unit.

The compression state analysis unit is a functional unit that analyzes the compression state of the object 2 including the conductive urethane 22 using time-series electrical characteristics (input data 4) that change according to the compression of the conductive urethane 22. . The compression state analysis unit derives the degree of compression indicated by the characteristics included in the electrical characteristics according to the third characteristics from the first characteristics described above. The comparison/judgment unit is connected to a threshold storage unit that stores a threshold for determining the compression state, and is a functional unit that compares the compression state of the analysis result with the compression state determination threshold, and outputs the comparison result as a degree of compression. It is. The threshold value storage unit may store the above-mentioned threshold values in the ROM 106 or the auxiliary storage device 108. Further, the comparison/determination unit may determine the compressed state using, for example, a threshold value for determining the compressed state. In this way, by configuring the estimator 1 according to functions in accordance with the features included in the electrical characteristics, it can be easily applied to independent device configurations and hardware configurations such as electric circuit configurations.

In the present disclosure, a case has been described in which conductive urethane is used as an example of the flexible member, but the flexible member only needs to have flexibility, and is of course not limited to the above-mentioned conductive urethane.

Furthermore, the technical scope of the present disclosure is not limited to the scope described in the above embodiments. Various changes or improvements can be made to the embodiments described above without departing from the spirit thereof, and forms with such changes or improvements are also included within the technical scope of the present disclosure.

Furthermore, in the above embodiment, the estimation process and the learning process are realized by a software configuration using flowcharts, but the invention is not limited to this. For example, each process is realized by a hardware configuration. It may also be in the form of

Also, a part of the estimation device, for example, a neural network such as a learning model, may be configured as a hardware circuit.

The first aspect of the technology of the present disclosure described above is
a detection unit that detects electrical characteristics between a plurality of predetermined detection points on a flexible material that has conductivity and whose electrical properties change according to deformation;
Using the time-series electrical characteristics that change according to the deformation of the flexible material and deformation state information indicating the deformation state related to the deformation of the flexible material as learning data, the time-series electrical characteristics are input, and the The time-series electrical characteristics detected by the detection unit of the estimation target including the flexible material are input to the learning model that has been trained to output deformation state information, and the time-series electrical characteristics are matched to the input time-series electrical characteristics. an estimating unit that estimates deformation state information indicating a deformation state;
This is an estimation device including:

In a second aspect, in the estimation device of the first aspect,
The deformed state is a stretched state in which the flexible material is stretched and contracted by applying stress,
The detection unit detects the electrical property indicating a change in volume resistance of the flexible material, which changes from a first tendency to a second tendency depending on the expansion/contraction state,
The learning model is trained to output information indicating the expansion/contraction state as the deformation state information.

In a third aspect, in the estimation device of the second aspect,
The elastic state is a compressed state indicated by a degree of compression that increases as the amount of compression increases when the flexible material is compressed,
The estimation unit estimates the degree of compression of the flexible material.

A fourth aspect is the estimation device of the third aspect,
The detection unit is configured to detect a change from one of an upward trend and a downward trend to the other when the flexible material is compressed as a change from the first tendency to a second tendency, and as the degree of compression increases, the first tendency changes from one to the other. detecting the electrical property in which the difference between the value of the electrical property in one trend and the value of the electrical property in a second trend that has changed from the first trend is large;
The estimating unit estimates the degree of compression based on a difference between the values of the electrical characteristics.

A fifth aspect is the estimation device according to the third aspect or the fourth aspect,
The flexible material has a structure having at least one of a fibrous and mesh-like skeleton, or a urethane material having a structure in which a plurality of micro air bubbles are scattered inside, and at least a part of the urethane material has a conductive material inside. A member coated or impregnated with
The learning model is learned using a physical quantity indicating a change point where the first trend changes to the second trend determined in response to internal addition or impregnation,
The estimation unit further estimates information indicating internal addition or impregnation as information indicating physical properties of the flexible material.

A sixth aspect is the estimation device of the second aspect,
The stretched state is a stretched state indicated by a degree of stretch that increases as the amount of stretch when the flexible material is stretched,
The detection unit detects the electrical property that changes depending on the elongated state,
The learning model is trained to output information indicating the caution state as the deformation state information using electrical characteristics based on a predetermined threshold value indicating the caution state with respect to the flexible material.

A seventh aspect is the estimation device according to any one of the first to sixth aspects,
The learning model includes a model generated by learning using a network based on reservoir computing using the flexible material as a reservoir.

An eighth aspect is the estimation device according to any one of the first to seventh aspects,
The apparatus further includes an output section that outputs the estimation result of the estimation section.

The ninth aspect is
the computer acquires the electrical properties from a detection unit that detects electrical properties between a plurality of predetermined detection points on a flexible material that is conductive and whose electrical properties change according to deformation;
Using the time-series electrical characteristics that change according to the deformation of the flexible material and deformation state information indicating the deformation state related to the deformation of the flexible material as learning data, the time-series electrical characteristics are input, and the The time-series electrical characteristics detected by the detection unit of the estimation target including the flexible material are input to the learning model that has been trained to output deformation state information, and the time-series electrical characteristics are matched to the input time-series electrical characteristics. This estimation method estimates deformation state information indicating the deformation state.

The tenth aspect is
acquiring the electrical properties from a detection unit that detects the electrical properties between a plurality of predetermined detection points on a flexible material that is conductive and whose electrical properties change according to deformation;
Using the time-series electrical characteristics that change according to the deformation of the flexible material and deformation state information indicating the deformation state related to the deformation of the flexible material as learning data, the time-series electrical characteristics are input, and the The time-series electrical characteristics detected by the detection unit of the estimation target including the flexible material are input to the learning model that has been trained to output deformation state information, and the time-series electrical characteristics are matched to the input time-series electrical characteristics. This is an estimation program for executing a process of estimating deformation state information indicating a deformation state.

The eleventh aspect is
Time-series electrical properties detected between a plurality of predetermined detection points on a flexible material that has conductivity and whose electrical properties change according to deformation, and a deformation state regarding the deformation of the flexible material. an acquisition unit that acquires deformation state information;
A learning model trained to use the time-series electrical characteristics acquired by the acquisition unit and the deformation state information as learning data, take the time-series electrical characteristics as input, and output the deformation state information. a learning model generation unit that generates;
This is a learning model generation device that includes:

According to the present disclosure, it is possible to estimate the deformation state of an object by using the electrical characteristics of the object including a conductive flexible material without using a special detection device. .

All documents, patent applications, and technical standards mentioned herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard was specifically and individually indicated to be incorporated by reference. Incorporated herein by reference. Further, the disclosure of Japanese Application No. 2022-045902 filed on March 22, 2022 is incorporated herein by reference in its entirety.

Claims (11)

1. a detection unit that detects electrical characteristics between a plurality of predetermined detection points on a flexible material that has conductivity and whose electrical properties change according to deformation;
   Using the time-series electrical characteristics that change according to the deformation of the flexible material and deformation state information indicating the deformation state related to the deformation of the flexible material as learning data, the time-series electrical characteristics are input, and the The time-series electrical characteristics detected by the detection unit of the estimation target including the flexible material are input to the learning model that has been trained to output deformation state information, and the time-series electrical characteristics are matched to the input time-series electrical characteristics. an estimating unit that estimates deformation state information indicating a deformation state;
   Estimation device including.
2. The deformed state is a stretched state in which the flexible material is stretched and contracted by applying stress,
   The detection unit detects the electrical property indicating a change in volume resistance of the flexible material, which changes from a first tendency to a second tendency depending on the expansion/contraction state,
   The estimation device according to claim 1, wherein the learning model is trained to output information indicating the expansion/contraction state as the deformation state information.
3. The elastic state is a compressed state indicated by a degree of compression that increases as the amount of compression increases when the flexible material is compressed,
   The estimating device according to claim 2, wherein the estimating unit estimates the degree of compression of the flexible material.
4. The detection unit is configured to detect a change from one of an upward trend and a downward trend to the other when the flexible material is compressed as a change from the first tendency to a second tendency, and as the degree of compression increases, the first tendency changes from one to the other. detecting the electrical property in which the difference between the value of the electrical property in one trend and the value of the electrical property in a second trend that has changed from the first trend is large;
   The estimating device according to claim 3, wherein the estimating unit estimates the degree of compression based on a difference between the values of the electrical characteristics.
5. The flexible material has a structure having at least one of a fibrous and mesh-like skeleton, or a urethane material having a structure in which a plurality of micro air bubbles are scattered inside, and at least a part of the urethane material has a conductive material inside. A member coated or impregnated with
   The learning model is learned using a physical quantity indicating a change point where the first trend changes to the second trend determined in response to internal addition or impregnation,
   The estimating device according to claim 3 or 4, wherein the estimating unit further estimates information indicating internal addition or impregnation as information indicating physical properties of the flexible material.
6. The stretched state is a stretched state indicated by a degree of stretch that increases as the amount of stretch when the flexible material is stretched,
   The detection unit detects the electrical property that changes depending on the elongated state,
   The estimation device according to claim 2, wherein the learning model is trained to output information indicating the caution state as the deformation state information using electrical characteristics based on a predetermined threshold value indicating the caution state with respect to the flexible material. .
7. The learning model includes a model generated by learning using a network using reservoir computing using the flexible material as a reservoir. Estimation device.
8. The estimation device according to any one of claims 1 to 7, further comprising an output unit that outputs an estimation result of the estimation unit.
9. the computer acquires the electrical properties from a detection unit that detects electrical properties between a plurality of predetermined detection points on a flexible material that is conductive and whose electrical properties change according to deformation;
   Using the time-series electrical characteristics that change according to the deformation of the flexible material and deformation state information indicating the deformation state related to the deformation of the flexible material as learning data, the time-series electrical characteristics are input, and the The time-series electrical characteristics detected by the detection unit of the estimation target including the flexible material are input to the learning model that has been trained to output deformation state information, and the time-series electrical characteristics are matched to the input time-series electrical characteristics. An estimation method for estimating deformation state information indicating a deformation state.
10. acquiring the electrical properties from a detection unit that detects the electrical properties between a plurality of predetermined detection points on a flexible material that is conductive and whose electrical properties change according to deformation;
    Using the time-series electrical characteristics that change according to the deformation of the flexible material and deformation state information indicating the deformation state related to the deformation of the flexible material as learning data, the time-series electrical characteristics are input, and the The time-series electrical characteristics detected by the detection unit of the estimation target including the flexible material are input to the learning model that has been trained to output deformation state information, and the time-series electrical characteristics are matched to the input time-series electrical characteristics. An estimation program for executing a process of estimating deformation state information indicating a deformation state.
11. Time-series electrical properties detected between a plurality of predetermined detection points on a flexible material that has conductivity and whose electrical properties change according to deformation, and a deformation state regarding the deformation of the flexible material. an acquisition unit that acquires deformation state information;
    A learning model trained to use the time-series electrical characteristics acquired by the acquisition unit and the deformation state information as learning data, take the time-series electrical characteristics as input, and output the deformation state information. a learning model generation unit that generates;
    A learning model generation device including: