State estimation system and state estimation method

Abstract

According to one embodiment of the present invention, in estimation of a state of contact between a sensing unit and an object or of the state of the object when the sensing unit is brought into contact with the object, a sensor device in which a first piezoelectric element (11) and a second piezoelectric element (12) are integrally configured in a state of overlapping one another is provided to the sensing unit. An acoustic interface unit drives the first piezoelectric element (11) to transmit a first sound wave, and detects a reception signal corresponding to a second sound wave received by the second piezoelectric element (12) in accordance with the transmission of the first sound wave. An estimation device extracts, from the reception signal, a frequency feature quantity pertaining thereto, and estimates the state of contact of the sensing unit with the object or the state of the object on the basis of the extracted frequency feature quantity.

Description

[Technical Field]

[0001] One aspect of the present invention relates to a state estimation system and a state estimation method used to estimate, for example, a contact state when an object is contacted or a state of the object. [Background technology]

[0002] In recent years, various sensing technologies have been proposed in line with the development of robotics technology and the like. One of these is a technology for estimating the contact state and state of an object when the sensor comes into contact with the object. For example, Non-Patent Document 1 discloses a technology called active acoustic sensing that uses a sensor having a pair of piezoelectric elements. This technology uses one of the piezoelectric elements as a speaker and the other as a microphone, transmits ultrasonic waves from the speaker, receives the ultrasonic waves with the microphone, and performs frequency analysis of the received signal to estimate the contact state and state of the object when the sensor comes into contact with the object. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] Ono, Makoto, Buntarou Shizuki, and Jiro Tanaka. "Touch & activate: adding interactivity to existing objects using active acoustic sensing." Proceedings of the 26th annual ACM symposium on User interface software and technology. 2013. Summary of the Invention [Problem to be solved by the invention]

[0004] However, the technology described in Non-Patent Document 1 involves arranging a pair of piezoelectric elements spaced apart on a plane, which results in a large sensor size, making it difficult to install in sensing units with limited installation space, such as a human finger or a robot probe.

[0005] This invention was made with the above circumstances in mind, and aims to provide a technology that can be installed on a small sensing unit with limited installation space, and that can estimate with high accuracy the state of an object that the sensing unit comes into contact with. [Means for solving the problem]

[0006] In order to solve the above problem, one aspect of the state estimation system or state estimation method according to the present invention is to provide a sensing unit. The device is attached to a muscle area whose hardness changes depending on the strength of the person's grip. Object Grip strength when holding When estimating, a sensor device is provided in the sensing unit, which is configured by integrating a first piezoelectric element and a second piezoelectric element stacked on top of each other. Then, an acoustic interface unit drives the first piezoelectric element to transmit a first sound wave, and detects a received signal corresponding to a second sound wave received by the second piezoelectric element in response to the transmission of the first sound wave. An estimation device extracts a frequency feature from the received signal, and calculates an estimate based on the extracted frequency feature. The strength of the person's grip on the object The method is designed to estimate the following.

[0007] According to one aspect of the present invention, the sensor device is configured by stacking the first and second piezoelectric elements together, which allows the sensor device to be smaller in size in the planar direction than when the piezoelectric elements are arranged side by side on a plane, making it possible to install the sensor device on a sensing unit with limited installation area, such as a human finger or a robot probe.

[0008] Furthermore, by integrating the first piezoelectric element and the second piezoelectric element, the first sound wave transmitted from the first piezoelectric element is By human hands Object Grip strength when holdingThis allows the second sound wave, which is reflected by the By human hands Object Grip strength when holding It is possible to estimate with high accuracy. [Effects of the Invention]

[0009] In other words, according to one aspect of the present invention, it is possible to provide a technology that can be installed on a small sensing unit with limited installation space and that can estimate with high accuracy the state of an object that the sensing unit comes into contact with. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram showing the overall configuration of a state estimation system according to a first embodiment of the present invention. [Figure 2] FIG. 2 is a side view showing the configuration of a sensor device used in the state estimation system shown in FIG. [Figure 3] FIG. 3 is a block diagram showing the functional configuration of an acoustic interface unit and an estimation device used in the state estimation system shown in FIG. [Figure 4] FIG. 4 is a flowchart showing an example of the procedure and content of the model learning process executed by the control unit of the estimation device shown in FIG. [Figure 5] FIG. 5 is a flowchart showing an example of the processing procedure and processing content of the state estimation processing executed by the control unit of the estimation device shown in FIG. [Figure 6] FIG. 6 is a diagram showing an example of the result of estimation of the state of the contacted object by the state estimation system shown in FIG. [Figure 7] FIG. 7 is a diagram showing an example of an estimation result of a contact state with respect to an object obtained by the system according to the second embodiment of the present invention. [Figure 8A] FIG. 8A is a diagram showing an example of a power spectrum in the entire frequency band of a received sound wave measured in a state where a sensor device placed on a "hard floor" is enclosed in a housing. [Figure 8B]FIG. 8B is an enlarged view of the power spectrum of the frequency band from which feature quantities are extracted, among the characteristics shown in FIG. 8A. [Figure 9A] FIG. 9A is a diagram showing an example of the power spectrum in the entire frequency band of the received sound wave measured in the same manner when the sensor device is placed on a "soft floor" and enclosed in a housing. [Figure 9B] FIG. 9B is an enlarged view of the power spectrum in the frequency band from which feature amounts are extracted, among the characteristics shown in FIG. 9A. [Figure 10A] FIG. 10A shows an example of the power spectrum in the entire frequency band of the received sound wave measured under the same conditions when the sensor device was placed on a "hard floor" and not enclosed in a housing. [Figure 10B] FIG. 10B is an enlarged view of the power spectrum in the frequency band from which feature amounts are extracted, among the characteristics shown in FIG. 10A. [Figure 11A] FIG. 11A shows an example of the power spectrum in the entire frequency band of the received sound wave measured under the same conditions when the sensor device was placed on a "soft floor" and not enclosed in a housing. [Figure 11B] FIG. 11B is an enlarged view of the power spectrum in the frequency band from which feature amounts are extracted, among the characteristics shown in FIG. 11A. DETAILED DESCRIPTION OF THE INVENTION

[0011] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0012] [First embodiment] (Configuration example) (1) System FIG. 1 is a diagram showing an example of the overall configuration of a state estimation system according to a first embodiment of the present invention.

[0013] The state estimation system according to the first embodiment includes a sensor device 1, an acoustic interface unit 2, an estimation apparatus 3, and an input / output device 4. The sensor device 1 is connected to the acoustic interface unit 2 via, for example, a signal cable, and the acoustic interface unit 2 is connected to the estimation apparatus 3. The input / output device 4 is connected to the estimation apparatus 3. The sensor device 1 is placed in contact with an object BX. In this example, the sensor device 1 is placed on, for example, the floor.

[0014] (2) Equipment (2-1) Sensor Device 1 The sensor device 1 functions as an active acoustic sensor, and is configured, for example, as follows: FIG.

[0015] That is, the sensor device 1 is composed of a sensor 10 and a housing 14 that is arranged to surround the sensor 10 on all sides. The sensor 10 is composed of a first piezoelectric element 11 for transmitting sound waves and a second piezoelectric element 12 for receiving sound waves, which are integrally formed back-to-back with an insulating substrate 13 sandwiched between them.

[0016] The first piezoelectric element 11 is, for example, a plate- or sheet-shaped piezoelectric element 111, with a plate- or sheet-shaped surface electrode 112 and a back electrode 113 disposed on both sides thereof, and terminals 114, 115 provided on the surface electrode 112 and the back electrode 113. The terminals 114, 115 receive the acoustic transmission signal output from the acoustic interface unit 2 via a signal cable.

[0017] Similar to the first piezoelectric element 11, the second piezoelectric element 12 also has, for example, a plate- or sheet-shaped surface electrode 122 and a back electrode 123 disposed on both sides of a plate- or sheet-shaped piezoelectric element 121, with terminals 124, 125 provided on the surface electrode 122 and the back electrode 123. The terminals 124, 125 output an acoustic wave reception signal corresponding to the acoustic wave received by the piezoelectric element 121 to the acoustic interface unit 2 via a signal cable.

[0018] The housing 14 is made of, for example, a plastic frame, and is arranged to surround the sensor 10 from all sides when the sensor 10 is placed directly on the object BX. The housing 14 may be a box with a closed top and an open bottom, or may be a U-shaped frame with no side surface. In the case of the box, a notch is provided on one side surface for leading out the signal cable of the sensor device 1.

[0019] (2-2) Acoustic interface unit 2 FIG. 3 is a block diagram showing the functional configuration of the acoustic interface unit 2 and the estimation device 3. As shown in FIG.

[0020] The acoustic interface unit 2 includes a sound wave control unit 21 and a sound wave amplification unit 22. The sound wave control unit 21 controls the transmission intensity of the sound wave transmission signal output from the estimation device 3 to a preset value, and supplies the controlled sound wave transmission signal to the first piezoelectric element 11 of the sensor 10. The sound wave amplification unit 22 amplifies the sound wave reception signal output from the second piezoelectric element 12 of the sensor 10, and outputs the amplified sound wave reception signal to the estimation device 3.

[0021] (2-3) Estimation device 3 The estimation device 3 is, for example, a personal computer. The estimation device 3 is a control device using a hardware processor such as a central processing unit (CPU). The control unit 31 is connected to a storage unit having a program storage unit 32 and a data storage unit 33, and an input / output interface (hereinafter, the interface will be referred to as I / F) unit 34 via a bus (not shown).

[0022] The input / output I / F unit 34 is connected to the acoustic interface unit 2 and the input / output device 4. The input / output device 4 includes an input unit having, for example, a keyboard and a mouse, and a display unit using, for example, a liquid crystal display. The input / output device 4 is used to input various control data for controlling the state estimation process to the estimation device 3, and to display estimation information and the like output from the estimation device 3.

[0023] The program storage unit 32 is, for example, a storage medium such as an SSD (Solid State Drive). It is configured by combining non-volatile memory that can be written to and read from at any time with non-volatile memory such as ROM (Read Only Memory), and stores middleware such as an OS (Operating System) as well as application programs required to execute various control processes according to the first embodiment. Hereinafter, the OS and each application program will be collectively referred to as the program.

[0024] The data storage unit 33 is, for example, a combination of a non-volatile memory such as an SSD that can be written to and read from at any time as a storage medium, and a volatile memory such as a RAM (Random Access Memory), and is equipped with an ultrasonic reception signal storage unit 331, a learning model storage unit 332, and an estimated information storage unit 333 as the main storage units required to implement the first embodiment.

[0025] The sonic wave reception signal storage unit 331 is used to store the sonic wave reception signal received from the acoustic interface unit 2 .

[0026] A learning model used to estimate the state of the object BX from the received sonic signal is stored in the learning model storage unit 332. As the learning model, for example, a Support Vector Machine (SVM) employing a supervised machine learning algorithm is used, but the present invention is not limited to this, and other models such as a Convolutional Neural Network (CNN) may also be used.

[0027] The estimated information storage unit 333 is used to store information indicating the estimated result of the state of the object BX obtained by the control unit 31.

[0028] The control unit 31 includes, as processing functions necessary for carrying out the first embodiment, a sound wave generation processing unit 311, a sound wave reception processing unit 312, a feature extraction processing unit 313, a model learning processing unit 314, a state estimation processing unit 315, and an estimated information output processing unit 316. These processing units 311 to 316 are all realized by causing a hardware processor of the control unit 31 to execute an application program stored in the program storage unit 32. Note that the application program may be stored in advance in the program storage unit 32, or may be downloaded when necessary from, for example, a server computer on the cloud.

[0029] The sound wave generation processing unit 311 generates a sound wave transmission signal in response to an instruction input from the input / output device 4, and outputs the generated sound wave transmission signal from the input / output I / F unit 34 to the acoustic interface unit 2.

[0030] The sonic wave reception processing unit 312 receives the sonic wave reception signal output from the acoustic interface unit 2 via the input / output I / F unit 34 and stores the received sonic wave reception signal in the sonic wave reception signal storage unit 331 .

[0031] The feature extraction processing unit 313 reads the received sonic wave signals from the received sonic wave signal storage unit 331 for a fixed period of time, and extracts feature amounts from the read received sonic wave signals. For example, a fast Fourier transform (FFT) is used to extract the feature amounts. The operation will be described in the operation example.

[0032] In a learning mode that is set prior to the actual estimation process, the model learning processing unit 314 performs processing to construct a learning model using the feature quantities extracted from the above-mentioned ultrasonic reception signal as explanatory variables and the correct label of the state estimation result entered by, for example, a system administrator as the objective variable.

[0033] In the estimation mode, the state estimation processing unit 315 inputs the features extracted from the above-mentioned ultrasonic reception signal into a trained learning model stored in the learning model memory unit 332, and then stores the resulting label output from the learning model in the estimation information memory unit 333 as information representing the state estimation result.

[0034] After the estimation process is completed, the estimated information output processing unit 316 reads information representing the estimation result of the state from the estimated information storage unit 333, and outputs the read estimated information from the input / output I / F unit 34 to the input / output device 4.

[0035] (Example of operation) Next, an example of the operation of the device configured as above will be described.

[0036] (1) Learning mode The control unit 31 of the estimation device 3 executes a learning process for the learning model prior to the state estimation process. Note that, here, an example will be described in which a learning model is constructed using a "hard floor" made of wood or resin, for example, and a "soft floor" covered with a sheet of cloth, urethane, or the like, as estimation targets.

[0037] FIG. 4 is a flowchart showing an example of the procedure and content of the learning process executed by the control unit 31 of the estimation device 3.

[0038] When a model learning processing request is input from the input / output device 4 and detected in step S10, the control unit 31 of the estimation device 3 sets the learning mode. Then, in this state, first, under the control of the sound wave generation processing unit 311, a sound wave transmission signal is generated in step S11, and the generated sound wave transmission signal is output from the input / output I / F unit 34 to the acoustic interface unit 2. Specifically, the sound wave generation processing unit 311 generates a sweep signal whose frequency varies in the range of 20-40 kHz, and outputs this sweep signal for 30 seconds each corresponding to two types of estimation targets, i.e., a "hard floor" and a "soft floor."

[0039] In response to this, when the acoustic interface unit 2 receives the sweep signal, the acoustic wave control unit 21 controls the sweep signal so that it has a predetermined transmission intensity. For example, the acoustic wave control unit 21 sets the sweep signal to an intensity determined by the rating of the piezoelectric element and such that a sufficient acoustic wave reception level is obtained. Then, the acoustic wave control unit 21 supplies the intensity-controlled sweep signal to the first piezoelectric element 11 of the sensor device 1 as an acoustic wave transmission signal.

[0040] In the sensor device 1, when the sound wave transmission signal is supplied, the first piezoelectric element 11 vibrates, thereby transmitting a sound wave. Then, in response to the transmission of the sound wave, a sound wave reflecting the state of the object BX, for example, the hardness or softness of the floor, is received by the second piezoelectric element 12, and a sound wave reception signal corresponding to the sound wave is input to the acoustic interface unit 2.

[0041] In the acoustic interface unit 2, the acoustic wave reception signal is amplified by the acoustic wave amplifier unit 22, and the amplified acoustic wave reception signal is output to the estimation device 3. At this time, the amplification degree is set to, for example, an optimum value for the estimation device 3 to convert the acoustic wave reception signal into a digital signal.

[0042] First, in step S12, under the control of the sonic reception processing unit 312, the control unit 31 of the estimation device 3 takes in the sonic reception signal converted into a digital signal by the input / output I / F unit 34, and stores this sonic reception signal in the sonic reception signal storage unit 331. At this time, the sampling rate of the sonic reception signal is set to, for example, 96 kHz.

[0043] Next, under the control of the feature extraction processing unit 313, the control unit 31 of the estimation device 3 reads the received sound wave signals from the received sound wave signal storage unit 331 for a fixed period of time in step S13, and inputs the read received sound wave signals into an FFT in step S14 to convert them into frequency domain data. For example, the feature extraction processing unit 313 reads the received sound wave signals for 8192 samples at a time and converts them into a power spectrum by FFT. The feature extraction processing unit 313 then generates a feature vector from the power spectrum.

[0044] Subsequently, in step S15, under the control of the model learning processing unit 314, the control unit 31 of the estimation device 3 constructs a learning model using the feature vectors as explanatory variables and the correct label input by the system administrator from, for example, the input / output device 4 as a target variable. In this example, "hard floor" is set as the correct label for the feature vectors obtained in the first 30 seconds, and "soft floor" is set as the correct label for the feature vectors obtained in the following 30 seconds.

[0045] Finally, in step S16, the control unit 31 of the estimation device 3 determines whether the learning process of the learning model has been completed. If not completed, the control unit 31 returns to step S11 to continue the learning process of the learning model; if completed, the control unit 31 ends the learning process.

[0046] (2) Estimation mode FIG. 5 is a flowchart showing an example of the processing procedure and processing content of the estimation processing executed by the control unit 31 of the estimation device 3.

[0047] When an estimation processing request is input from the input / output device 4 and detected in step S20, the control unit 31 of the estimation device 3 sets the estimation mode. Then, in this state, first, under the control of the sound wave generation processing unit 311, a sound wave transmission signal is generated in step S21, and the generated sound wave transmission signal is output from the input / output I / F unit 34 to the acoustic interface unit 2. In this case, as in the case of the learning mode, the sound wave generation processing unit 311 generates a sweep signal whose frequency changes in the range of 20-40 kHz, and outputs this sweep signal for a fixed period of time.

[0048] Then, the acoustic wave transmission signal is supplied to the first piezoelectric element 11 of the sensor device 1 after being controlled by the acoustic wave control unit 21 in the acoustic interface unit 2 so that the intensity thereof becomes a predetermined value.

[0049] When the sound wave transmission signal is supplied to the sensor device 1, the first piezoelectric element 11 vibrates, thereby transmitting a sound wave. In response to the transmission of the sound wave, a sound wave reflecting the state of the object BX on which the sensor device 1 is placed, such as hardness or softness, is received by the second piezoelectric element 12, and a sound wave reception signal corresponding to the sound wave is output to the acoustic interface unit 2.

[0050] In the acoustic interface unit 2 , the acoustic wave reception signal is amplified by the acoustic wave amplifier 22 , and the amplified acoustic wave reception signal is output to the estimation device 3 .

[0051] First, in step S22, the control unit 31 of the estimation device 3, under the control of the sonic reception processing unit 312, takes in the sonic reception signal converted into a digital signal by the input / output I / F unit 34, and stores this sonic reception signal in the sonic reception signal storage unit 331. In this case as well, the sonic reception signal is sampled at a sampling rate of 96 kHz, just as during learning.

[0052] Next, in step S23, the control unit 31 of the estimation device 3 reads the received sound signal for a certain period of time, for example, 8192 samples at a time, from the received sound signal storage unit 331 under the control of the feature extraction processing unit 313, and inputs the read received sound signal into an FFT to obtain a power spectrum. Then, the feature extraction processing unit 313 generates a feature vector from the power spectrum and outputs the generated feature vector to the state estimation processing unit 315.

[0053] When the feature vector is input, the state estimation processing unit 315 inputs the feature vector to the learning model as an explanatory variable in step S24. As a result, an estimation result (estimated label) corresponding to the feature vector is output from the learning model.

[0054] For example, if sensor device 1 is placed on a "hard floor," a label representing "hard floor" is output as the estimation result. If sensor device 1 is placed on a "soft floor," a label representing "soft floor" is output as the estimation result.

[0055] The state estimation processing unit 315 receives information representing the estimation result output from the learning model in step S25, and stores it in the estimation information storage unit 333.

[0056] In step 26, under the control of the estimated information output processing unit 316, the control unit 31 of the estimation device 3 reads information representing the estimation result from the estimated information storage unit 333, and outputs the read information representing the estimation result from the input / output I / F unit 34 to the input / output device 4. As a result, the input / output device 4 displays information representing the estimation result, for example, information representing "hard floor" or "soft floor", on, for example, the display unit.

[0057] Finally, in step S27, the control unit 31 of the estimation device 3 determines whether the learning process of the learning model has been completed. If not completed, the control unit 31 returns to step S21 to continue the estimation process, and if completed, the control unit 31 ends the estimation process.

[0058] (Actions and Effects) As described above, in the first embodiment, the sensor device 1 has a structure in which the first piezoelectric element 11 used as a speaker and the second piezoelectric element 12 used as a microphone are integrated back to back.

[0059] Therefore, it is possible to reduce the size of the sensor device 1 in the planar direction compared to when the piezoelectric elements 11 and 12 are arranged side by side on a plane, for example, and therefore it is possible to install it in a sensing unit with a limited installation area, such as a human finger or a robot probe.

[0060] In the first embodiment, the estimation device 3 first constructs a learning model in a learning mode in which the feature vector of the power spectrum extracted from the sound wave reception signal received by the sensor device 1 and the corresponding correct label of the state of the object BX are used as explanatory variables and objective variables, respectively, and then in an estimation mode, the feature vector is extracted from the sound wave reception signal received by the sensor device 1 and input into the learning model, thereby estimating the state of the object BX.

[0061] Therefore, the structure integrating the first piezoelectric element 11 and the second piezoelectric element 12 makes it possible to efficiently receive sound waves reflecting the state of the object BX without significant attenuation, thereby making it possible to estimate the state of the object BX with high accuracy.

[0062] Fig. 6 shows an example of the results of estimating floor conditions when the sensor device 1 is placed on a "hard floor" and a "soft floor" under the conditions described in the first embodiment. As shown in Fig. 6, the accuracy rate of the predicted labels output from the learning model was 99% for the "hard floor" and 100% for the "soft floor," confirming that this system is capable of estimating floor conditions with high accuracy.

[0063] Furthermore, in the first embodiment, the sensor device 1 is placed on the floor, which is the object BX, and is surrounded by the housing 14. This makes it possible to improve the reception frequency characteristics of sound waves for both a "hard floor" and a "soft floor."

[0064] Figure 8A shows an example of the power spectrum of the received sound waves in all frequency bands measured when the sensor device 1 is placed on a "hard floor" and surrounded by the housing 14, and Figure 8B shows an enlarged view of the power spectrum of the frequency band from which features are extracted.

[0065] FIG. 9A shows an example of the power spectrum of the entire frequency band of the received sound wave measured in the same manner when the sensor device 1 is placed on a "soft floor" and surrounded by the housing 14, and FIG. 9B shows an enlarged view of the power spectrum of the frequency band from which feature extraction is to be performed.

[0066] 10A and 10B are diagrams showing an example of the power spectrum in the entire frequency band of the received sound wave and the frequency band from which feature quantities are extracted, measured under the same conditions, with the sensor device 1 placed on a "hard floor" and not enclosed by the housing 14. Also, Fig. 11A and 11B are diagrams showing an example of the power spectrum in the entire frequency band of the received sound wave and the frequency band from which feature quantities are extracted, measured under the same conditions, with the sensor device 1 placed on a "soft floor" and not enclosed by the housing 14.

[0067] As is clear from comparing these figures, whether the sensor device 1 is placed on a "hard floor" or a "soft floor," surrounding the sensor device 1 with the housing 14 can make the change in acoustic characteristics relative to the floor, which is the target object BX, more noticeable. In other words, surrounding the sensor device 1 with the housing 14 can be expected to further improve the estimation accuracy.

[0068] [Second embodiment] In the first embodiment, an example was described in which a "floor" was used as an estimation target and its condition was estimated as "hard" or "soft." However, the present invention is not limited to this, and it may also be possible to estimate, for example, the contact state with an object.

[0069] For example, if a person wears sensor device 1 on the thenar eminence muscle on the back of their right hand and grasps a rod-shaped object BX in this state, the system of this invention can estimate the difference between a "tight grip" and a "weak grip" from the difference in acoustic characteristics. In this case, as in the first embodiment described above, by constructing a learning model in advance in learning mode, it becomes possible to estimate the contact state with object BX, i.e., whether the person grasped the object BX tightly or loosely. It is presumed that this is because the stiffness of the thenar eminence muscle of the person's hand changes depending on the strength of the grip, and the acoustic characteristics change in response to this change, making it possible to estimate the strength of the grip.

[0070] 7 shows an example of the results of estimating a rod-shaped object BX when it is gripped "tightly" and when it is gripped "weakly" with the sensor device 1 attached to the thenar eminence of a person's hand. As shown in Fig. 7, the accuracy rate of the predicted labels output from the learning model was 100% for both the "tightly gripped" and "weakly gripped" cases, confirming that the system according to the present invention can also estimate the floor condition with high accuracy when it comes to the contact condition of the object BX.

[0071] [Other embodiments] (1) In the first embodiment, the estimation result is displayed on the display unit of the input / output device 4. However, the present invention is not limited to this. For example, the estimation result may be transmitted from the communication interface unit via a network to a remote terminal or the like and displayed thereon.

[0072] (2) In the first embodiment, the sensor device 1 is placed on the floor, which is the object BX, to estimate the hardness of the floor, and in the second embodiment, the sensor device is attached to a finger to estimate the grip strength when gripping the object. However, in addition to this, the sensor device according to the present invention may be attached to, for example, a part of a robot that corresponds to a hand or a probe, and used to estimate the grip strength when the robot grasps an object, the hardness of the object when it is pushed, etc.

[0073] (3) The estimation device is composed of a server computer located on the web or cloud. In this case, for example, a terminal such as a personal computer or a smartphone is connected to the acoustic interface unit 2, and sonic transmission signals and sonic reception signals are transmitted to and from the server computer via this terminal. In addition, user operation information is sent from the terminal to the server computer, and estimation information obtained by the estimation device is received by the terminal and displayed on its display unit.

[0074] (4) In addition, the structure of the sensor of the sensor device, the structure of the housing surrounding it, the functions, processing procedures, and processing contents of the estimation device, etc. can be modified in various ways within the scope of the gist of this invention, and the type, shape, material, etc. of the object whose state is to be estimated or whose contact state is to be estimated can also be any type.

[0075] Although the embodiments of the present invention have been described in detail above, the above description is merely an example of the present invention in every respect. It goes without saying that various improvements and modifications can be made without departing from the scope of the present invention. In other words, when implementing the present invention, specific configurations according to the embodiments may be appropriately adopted.

[0076] In short, this invention is not limited to the above-described embodiments, and the components can be modified and embodied in practice without departing from the spirit of the invention. Furthermore, various inventions can be created by appropriately combining multiple components disclosed in the above-described embodiments. For example, some components may be omitted from all the components shown in each embodiment. Furthermore, components from different embodiments may be appropriately combined. [Explanation of symbols]

[0077] 1...Sensor device 2...Acoustic interface section 3…Estimation device 4...Input / output devices 10...Sensor 11...First piezoelectric element 12...Second piezoelectric element 13...Insulating substrate 14...Housing 21...Sound wave control unit 22...Sound wave amplifier 111,121...Piezo element 112,122…Surface electrode 113,123…back electrode 114, 115, 124, 125...Terminals 31...Control unit 32...Program memory section 33...Data storage unit 34...Input / output interface 311...sound wave generation processing unit 312...sound wave receiving processing unit 313...Feature extraction processing unit 314...Model learning processing unit 315...State estimation processing unit 316...Estimated information output processing unit 331...sound wave reception signal storage unit 332...Learning model memory unit 333... Estimated information storage unit

Claims

1. A state estimation system in which a sensing unit is attached to a muscle part whose hardness changes depending on the strength of a grip of a person's hand, and in this state, estimates the grip strength when an object is gripped by the person's hand, a sensor device in which a first piezoelectric element and a second piezoelectric element are integrally arranged in a state where they overlap each other, the sensor device being installed in the sensing unit; an acoustic interface unit that drives the first piezoelectric element to transmit a first sound wave and detects a reception signal corresponding to a second sound wave received by the second piezoelectric element in response to the transmission of the first sound wave; an estimation device that extracts a feature amount from the received signal and estimates the grip strength of the person's hand on the object based on the extracted feature amount; A state estimation system comprising:

2. The state estimation system according to claim 1 , wherein the sensor device is an integrated device in which the first piezoelectric element and the second piezoelectric element are arranged back to back.

3. The state estimation system according to claim 1 , wherein the sensor device further comprises a housing arranged to surround the integrated first and second piezoelectric elements.

4. 2. The state estimation system according to claim 1, wherein the estimation device further comprises a sound wave generation unit that generates a sweep signal whose frequency changes within a preset range and supplies the generated sweep signal to the acoustic interface unit to drive the first piezoelectric element.

5. 2. The state estimation system according to claim 1, wherein the estimation device converts the received signal into a frequency domain signal by fast Fourier transform processing at predetermined time intervals, and generates a feature vector from a power spectrum of the converted frequency domain signal.

6. A state estimation method in which a sensing unit is attached to a muscle part whose hardness changes depending on the strength of a grip of a person's hand, and in this state, the grip strength of an object is estimated when the person grips the object with the hand, a step of supplying a drive signal from an acoustic interface unit to a sensor device in which a first piezoelectric element and a second piezoelectric element are integrally arranged so as to overlap each other, thereby causing the first piezoelectric element to transmit a first sound wave; detecting, by the acoustic interface unit, a reception signal corresponding to a second sound wave received by the second piezoelectric element in response to the transmission of the first sound wave; extracting a feature amount from the received signal by an estimation device, and estimating the grip strength of the object by the person's hand based on the extracted feature amount; A state estimation method comprising:

Images