CN117769504A

CN117769504A - Estimation device, control device, estimation method, and estimation program

Info

Publication number: CN117769504A
Application number: CN202280051879.5A
Authority: CN
Inventors: 北野创; 若尾泰通; 篠原寿充; 江部一成; 鬼木良彦
Original assignee: Bridgestone Corp
Current assignee: Bridgestone Corp
Priority date: 2021-07-26
Filing date: 2022-07-12
Publication date: 2024-03-26
Also published as: JP2023017640A

Abstract

The estimation device detects, by the detection unit, electrical characteristics between a plurality of detection points of a vehicle seat provided with a flexible material that has conductivity and changes in electrical characteristics in accordance with changes in pressure applied thereto. The estimating unit estimates the posture state of the occupant of the vehicle seat from the electrical characteristics of the vehicle seat using the learning model. The learning model obtained by learning, using as learning data, the electrical characteristics when pressure is applied to the vehicle seat and posture state information indicating the posture state of the occupant applying pressure to the vehicle seat, and learning the electrical characteristics as input and outputting the posture state information is learned as input electrical characteristics and outputs the posture state of the occupant corresponding to the input electrical characteristics.

Description

Estimation device, control device, estimation method, and estimation program

Technical Field

The present disclosure relates to an estimation device, a control device, an estimation method, and an estimation program.

Background

In order to determine the posture of an occupant seated in a vehicle seat, the following processes have been performed: a shape change generated in the vehicle seat is detected, and the posture of the occupant is estimated using the detection result. In detecting a shape change of a vehicle seat, it is difficult to detect deformation of the vehicle seat without interfering with the deformation. Further, since strain sensors used for detecting rigid bodies such as metal deformation are difficult to use in vehicle seats, special detection devices are required to detect deformation of the vehicle seats. For example, a technique is known in which a deformation image is acquired by measuring displacement and vibration of an object obtained by a camera, and a deformation amount is extracted (for example, refer to japanese patent laid-open No. 2017-029905). Further, a technique related to a soft touch sensor that estimates a deformation amount from a light transmission amount is also known (for example, refer to japanese patent application laid-open No. 2013-101096).

In addition, in order to maintain the posture of an occupant seated in a vehicle seat in an appropriate state, the following technique is known (for example, refer to japanese patent application laid-open No. 2019-137286): the body pressure sensor detects the body pressure of an occupant seated in a vehicle seat of an automobile, and the servo member provided on the seatback, the seat cushion, and the like is driven based on the detected body pressure, thereby maintaining the body pressure distribution at an appropriate distribution state when seated.

Disclosure of Invention

Problems to be solved by the invention

However, in detecting a shape change of a vehicle seat, when a deformation amount such as displacement of an object is detected by using a camera and an image analysis method, a system including the camera and the image analysis is undesirable because the system increases in size, which leads to an increase in size of the device. In addition, in the optical method using a camera, a hidden portion that cannot be captured by the camera cannot be measured. Therefore, there is room for improvement in detecting deformation of the vehicle seat.

In addition, in the case of using vibration detected by a sensor such as a piezoelectric element, the sensor itself may interfere with deformation of the vehicle seat in order to maintain the posture of the occupant seated in the vehicle seat in an appropriate state. In addition, in the detection of the vibration, the seating state due to the deformation of the vehicle seat may not be considered, and there is room for improvement in estimating the seating state. In addition, the occupant sits on the sensor to feel unnatural, and the riding comfort is changed, so that there is a problem that measurement and riding comfort evaluation cannot be performed simultaneously.

An object of the present disclosure is to provide an estimation device, a control device, an estimation method, and an estimation program that can estimate posture state information indicating a posture state of an occupant of a vehicle seat using an electrical characteristic of the vehicle seat provided with a flexible material having conductivity without using a special detection device.

Solution for solving the problem

In order to achieve the above object, aspect 1 is an estimation device, comprising:

a detection unit that detects electrical characteristics between a plurality of detection points predetermined for a soft material of a vehicle seat provided with the soft material, the soft material having conductivity and the electrical characteristics changing in response to a change in applied pressure; and

an estimating unit that inputs the time-series electrical characteristic detected by the detecting unit and estimates posture state information indicating a posture state of the occupant corresponding to the input time-series electrical characteristic, with respect to a learning model obtained by learning, using as learning data, the time-series electrical characteristic when pressure is applied to the soft material and posture state information indicating a posture state of the occupant of the vehicle seat to which pressure is applied, the time-series electrical characteristic being input and the posture state information being output.

In the estimation device according to claim 1, according to claim 2,

the electrical characteristic is a volume resistance,

the vehicle seat includes at least one of a seat cushion, a seat back, a headrest, and an armrest,

the posture state includes a seating state in which an occupant is seated on the vehicle seat,

the learning model learns to output information representing a seating state of the occupant corresponding to the detected electrical characteristic as the posture state information.

In the 3 rd aspect, in the estimating apparatus according to the 2 nd aspect,

the vehicle seat includes a material obtained by imparting conductivity to at least a part of a polyurethane material having a structure of a skeleton of at least one of a fibrous shape and a mesh shape, or a structure in which a plurality of fine air bubbles are dispersed therein.

In the 4 th aspect, in the estimating device according to the 2 nd or 3 rd aspect,

the seating state includes a state related to breathing of an occupant of the vehicle seat, a state related to an action of a seating mode of the occupant, and a state related to tilting of the vehicle seat,

the learning model learns to output, as the posture state information, information indicating at least one of a state related to breathing of an occupant, a state related to an operation of a seating mode of the occupant, and a state related to tilting of the vehicle seat, corresponding to the detected electrical characteristic.

In the 5 th aspect, in the estimating apparatus according to any one of the 1 st to 4 th aspects,

the learning model includes a model generated by taking the soft material as a reservoir and learning using a network calculated based on the reservoir using the reservoir.

In the 6 th aspect, in the estimating apparatus according to any one of the 1 st to 5 th aspects,

the soft material has conductivity and electrical characteristics which change according to the applied pressure and the change of moisture,

the posture state information is time-series electrical characteristics when pressure and moisture are applied to the soft material and posture state information indicating a posture state of an occupant of the vehicle seat to which pressure and moisture are applied with water,

the estimating unit estimates posture state information indicating a posture state of the occupant accompanied by water.

In the estimation device according to claim 7, in the estimation device according to claim 6,

the posture state of the occupant accompanying water includes a posture state accompanying perspiration in the seating state of the occupant.

In the 8 th aspect, in the estimating device according to the 6 th or 7 th aspect,

The posture state includes a seating state with water in which an occupant sits on the vehicle seat,

the learning model learns to output, as the posture state information, information representing a seating state of the occupant accompanied by water content corresponding to the detected electrical characteristic.

In the 9 th aspect, in the estimating apparatus according to any one of the 1 st to 5 th aspects,

the posture state information is time-series electrical characteristics when pressure is applied to the soft material and posture state information indicating a posture state of an occupant of the vehicle seat to which pressure is applied with movement,

the estimating unit estimates posture state information indicating a posture state of the occupant accompanying movement.

In a 10 th aspect, in the estimating apparatus according to the 9 th aspect,

the posture state includes a seating state in which an occupant sits on the vehicle seat with movement,

the learning model learns to output, as the posture state information, information representing a seating state of the occupant accompanying movement corresponding to the detected electrical characteristic.

An 11 th aspect is a control device including:

the estimation device according to any one of the aspects 6 to 8; and

And a control unit that controls an environment adjustment device including at least one of a temperature adjustment device for the vehicle seat and an air conditioning device provided with a cabin of the vehicle seat, using the posture state information estimated by the estimation unit.

In a 12 th aspect, in the control device according to the 11 th aspect,

the control unit controls the environment adjustment device using the posture state information estimated by the estimation unit.

Mode 13 is a control device including:

an estimating device according to any one of the aspects 9 and 10; and

and a control unit that controls a vehicle device constituting a vehicle on which the vehicle seat is mounted, using the posture state information estimated by the estimation unit.

In a 14 th aspect, in the control device according to the 13 th aspect,

the estimating portion estimates the state of the occupant as a seating state accompanied by movement,

the control portion controls the vehicular apparatus based on the state of the occupant.

In a 15 th aspect, in the control device according to the 14 th aspect,

The state of the occupant is a state related to the physical and mental condition of the occupant, and the control unit controls a seat driving device as the vehicle device so that the position of the vehicle seat becomes a position corresponding to the state related to the physical and mental condition of the occupant.

In a 16 th aspect, in the control device according to the 14 th aspect,

the state of the occupant is a state related to a physical and mental condition of the occupant, and the control portion causes an acoustic output device as the vehicular apparatus to output an acoustic message corresponding to the state related to the physical and mental condition of the occupant.

In a 17 th aspect, in the control device according to the 14 th aspect,

the occupant's state is a state related to a seating mode operation, and the control unit controls a seat driving device as the vehicle device so that a position of the vehicle seat becomes a position corresponding to the seating mode operation.

An 18 th aspect is an estimation method in which a computer performs:

acquiring an electrical characteristic from a detection unit that detects the electrical characteristic between a plurality of detection points predetermined for a soft material of a vehicle seat provided with the soft material, the soft material having conductivity and the electrical characteristic changing in accordance with a change in applied pressure; and

For a learning model obtained by using, as learning data, time-series electrical characteristics when pressure is applied to the soft material and posture state information indicating a posture state of an occupant of the vehicle seat to which pressure is applied, and learning the time-series electrical characteristics as input and outputting the posture state information, the time-series electrical characteristics detected by the detection unit are input, and posture state information indicating the posture state of the occupant corresponding to the input time-series electrical characteristics is estimated.

Mode 19 is an estimation method for causing a computer to execute:

ADVANTAGEOUS EFFECTS OF INVENTION

According to the present disclosure, there is an effect that it is possible to estimate posture state information indicating a posture state of an occupant of a vehicle seat using an electrical characteristic of the vehicle seat provided with a flexible material having conductivity without using a special detection device.

Drawings

Fig. 1 is a diagram showing an example of the configuration of a posture state estimation device according to the first embodiment.

Fig. 2 is a diagram of a vehicle seat according to the first embodiment.

Fig. 3 is a diagram of a detection point of the conductive member according to the first embodiment.

Fig. 4 is a diagram of a member having conductivity according to the first embodiment.

Fig. 5 is a diagram of a member having conductivity according to the first embodiment.

Fig. 6 is a diagram of a member having conductivity according to the first embodiment.

Fig. 7 is a diagram relating to the learning process according to the first embodiment.

Fig. 8 is a flowchart showing an example of learning data collection processing according to the first embodiment.

Fig. 9 is a diagram relating to the learning process in the learning process section according to the first embodiment.

Fig. 10 is a flowchart showing an example of the flow of the learning process according to the first embodiment.

Fig. 11 is a diagram relating to the learning process in the learning process section according to the first embodiment.

Fig. 12 is a diagram showing an example of the configuration of the posture state estimation device according to the first embodiment.

Fig. 13 is a flowchart showing an example of the flow of the estimation process according to the first embodiment.

Fig. 14 is a diagram showing characteristics relating to the vehicle seat according to the first embodiment.

Fig. 15 is a diagram showing characteristics relating to the vehicle seat according to the first embodiment.

Fig. 16 is a diagram of a member having conductivity according to the first embodiment.

Fig. 17 is a diagram showing an example of the configuration of the posture state estimation device according to the second embodiment.

Fig. 18 is a diagram relating to the learning process according to the second embodiment.

Fig. 19 is a flowchart showing an example of learning data collection processing according to the second embodiment.

Fig. 20 is a diagram showing characteristics relating to the vehicle seat according to the second embodiment.

Fig. 21 is a diagram showing characteristics relating to a vehicle seat according to a second embodiment.

Fig. 22 is a diagram relating to learning processing in the learning processing section according to the second embodiment.

Fig. 23 is a flowchart showing an example of the flow of the learning process according to the second embodiment.

Fig. 24 is a diagram relating to learning processing in the learning processing section according to the second embodiment.

Fig. 25 is a flowchart showing an example of the flow of the estimation process according to the second embodiment.

Fig. 26 is a diagram showing an example of the configuration of the control device according to the third embodiment.

Fig. 27 is a flowchart showing an example of the flow of estimation/control processing according to the third embodiment.

Fig. 28 is a diagram showing an example of the configuration of the posture state estimation device according to the fourth embodiment.

Fig. 29 is a diagram relating to the learning process according to the fourth embodiment.

Fig. 30 is a flowchart showing an example of learning data collection processing according to the fourth embodiment.

Fig. 31 is a diagram relating to learning processing in the learning processing section according to the fourth embodiment.

Fig. 32 is a flowchart showing an example of the flow of the learning process according to the fourth embodiment.

Fig. 33 is a diagram relating to learning processing in the learning processing section according to the fourth embodiment.

Fig. 34 is a flowchart showing an example of the flow of the estimation process according to the fourth embodiment.

Fig. 35 is a diagram showing characteristics relating to a vehicle seat according to a fourth embodiment.

Fig. 36 is a diagram showing characteristics relating to the vehicle seat according to the fourth embodiment.

Fig. 37 is a diagram showing characteristics relating to a vehicle seat according to a fourth embodiment.

Fig. 38 is a diagram showing an example of the configuration of a control device according to the fifth embodiment.

Fig. 39 is a flowchart showing an example of the flow of estimation/control processing according to the fifth embodiment.

Detailed Description

Embodiments for implementing the techniques of the present disclosure are described in detail below with reference to the accompanying drawings.

In all the drawings, the same components and processes having the same effects on actions and functions are denoted by the same reference numerals, and overlapping descriptions may be omitted as appropriate. The present disclosure is not limited to the following embodiments, and can be implemented with modifications as appropriate within the scope of the object of the present disclosure. In addition, in the present disclosure, estimation of a physical quantity with respect to a non-linearly deformed member is mainly described, but it is needless to say that the present invention can be applied to estimation of a physical quantity with respect to a linearly deformed member.

< first embodiment >

In the present disclosure, the term "vehicle seat" is used as an example of a seat on which an occupant such as a driver's seat, a passenger's seat, and a rear seat of a second or subsequent row of a vehicle such as an automobile sits, and is a concept including a seat cushion that is a seat surface on which the occupant sits, a seat back that is a backrest that supports a back of the occupant, a headrest that supports a head of the occupant, and an armrest that allows the occupant to put an arm. The "vehicle seat" is a concept including a material at least a part of which is deformed such as deflected by an external force. As an example of the external force, pressure is given as a stimulus applied to the vehicle seat. The term "posture state of an occupant seated in the vehicle seat" is a concept including a state of an occupant that deforms a soft material and applies pressure to the soft material, and includes behaviors of the occupant such as posture and movement of the occupant seated in the vehicle seat. The posture state is a seating state in which an occupant is seated in the vehicle seat.

In the present disclosure, the term "soft material" is a concept including a material at least a part of which is deformed such as being deflected by an external force, and includes a soft elastic body such as a rubber material, a structure having a skeleton of at least one of a fibrous shape and a mesh shape, and a structure in which a plurality of fine air bubbles are dispersed. For an example of the external force, the pressure is exemplified. Examples of the structure having a skeleton of at least one of a fibrous shape and a mesh shape and a structure in which a plurality of fine air bubbles are dispersed include a polymer material such as a polyurethane material. The "soft material to which conductivity is imparted" is a concept including a material having conductivity, and includes a material obtained by adding a conductive material to a soft material in order to impart conductivity, and a material in which a soft material has conductivity. The soft material to which conductivity is imparted has a function of changing electrical characteristics according to deformation. Further, as an example of the physical quantity that causes the function of changing the electrical characteristics according to the deformation, a pressure value indicating a stimulus (hereinafter referred to as a pressure stimulus) based on the pressure applied to the soft material is exemplified. The soft material deforms according to the distribution of external force, for example, pressure stimulus, generated in the seating state of the occupant or the like. An example of the physical quantity showing the electrical characteristics that change according to the deformation is the resistance value. In addition, for other examples, voltage values or current values are exemplified. The resistance value can be understood as the volume resistance value of the flexible material.

The soft material exhibits electrical characteristics corresponding to deformation caused by pressure by imparting conductivity. That is, the electrical paths of the flexible material to which conductivity is imparted are combined in a complicated manner, and the electrical paths expand and contract according to the deformation. In addition, there are cases where the electrical path is temporarily cut off and the connection is made differently from the conventional one. Accordingly, the flexible material exhibits a behavior having different electrical characteristics depending on the magnitude and distribution of the applied force (e.g., pressure stimulus) between the positions (e.g., positions where the electrode-disposed detection points) spaced apart by a predetermined distance. Accordingly, the electrical characteristics change according to the magnitude and distribution of the force (e.g., pressure stimulus) applied to the soft material. Further, since the flexible material to which conductivity is imparted is used, it is not necessary to provide detection points such as electrodes at all the positions where pressure is applied to the flexible material by an object such as a person. Any detection points such as electrodes may be provided at any at least two positions of the flexible material between which the pressure is to be applied.

The estimating device of the present disclosure estimates the posture state of an occupant seated in a vehicle seat from the electrical characteristics of a soft material having conductivity, which is provided in the vehicle seat, using a learned learning model. The soft material can be disposed in a vehicle seat. The learning model uses, as learning data, time-series electrical characteristics when pressure is applied to a flexible material having conductivity and posture state information indicating a posture state of an occupant seated in a vehicle seat to which pressure is applied to the flexible material. The learning model learns to take as input the time-series electrical characteristics and to output posture state information indicating a posture state of an occupant seated in the vehicle seat corresponding to the time-series electrical characteristics.

In the following description, a case will be described in which a seat member (hereinafter referred to as conductive polyurethane) in which a conductive material is impregnated or mixed is disposed in all or at least a part of a polyurethane member as a vehicle seat, and the seat member is used as a soft material having conductivity. Further, the conductive polyurethane is provided to at least one of a seat cushion, a seat back, a headrest, and an armrest. The conductive polyurethane may be provided separately to the seat cushion, the seat back, the headrest, and the armrest, or may be provided integrally with the seat cushion, the seat back, the headrest, and the armrest. That is, at least one or more conductive polyurethane is provided in at least a part of the range that an occupant contacts when sitting in the vehicle seat. The thickness of the conductive polyurethane is preferably 1mm or more, for example. The volume resistance value of the conductive polyurethane is preferably 10, for example ⁷ Omega cm or less. As a physical quantity for deforming the conductive urethane, a value (pressure value) indicating pressure stimulus applied to the vehicle seat is applied, and as a posture state, a seating state of an occupant seated in the vehicle seat is applied. The case where the resistance value of the conductive urethane is applied is described as the physical quantity that changes according to the pressure stimulus.

Fig. 1 shows an example of a configuration of a posture state estimating device 1 as an estimating device of the present disclosure.

In the estimation process in the posture state estimation device 1, as the unknown posture state of the occupant seated in the vehicle seat, the learned learning model 51 is used to estimate and output the seating state of the occupant related to the posture and movement of the occupant OP seated in the vehicle seat 2. This makes it possible to determine the posture state of the occupant seated in the vehicle seat 2 without using a special device, a large-sized device, or without directly measuring the deformation of the soft material included in the vehicle seat 2. The learning model 51 takes as a tag the posture state (for example, seating state value) of the occupant with respect to the vehicle seat 2, and takes as input the electrical characteristics of the vehicle seat 2 in this posture state (that is, the resistance value of the conductive urethane disposed in the vehicle seat 2) to learn. Learning of the learning model 51 is described later.

In the present embodiment, for example, the vehicle seat 2 is configured to include a seat cushion 21A, a seat back 21B, and a headrest 21C. The conductive urethane 22A is disposed on the seat cushion 21A, the conductive urethane 22B is disposed on the seat back 21B, and the conductive urethane 22C is disposed on the headrest 21C. The conductive urethanes 22A, 22B, 22C are connected to an electrical characteristic detecting unit 76 (see fig. 3). The conductive polyurethane 22A, 22B, 22C are made of the same conductive polyurethane. The seat cushion 21A, the seat back 21B, and the headrest 21C are referred to as a whole as the seat 21. The conductive polyurethane 22A, 22B, and 22C are referred to as conductive polyurethane 22 as a whole. Further, it is assumed that the conductive polyurethane 22 is formed by any method of internally adding a conductive material and impregnating the conductive material, but the conductivity of the impregnation method is desirably higher than that of the internal addition method.

The vehicle seat 2 including the seat 21 provided with the conductive urethane 22 functions as a detection unit. As shown in fig. 2, the conductive urethane 22 may be disposed at least in a part of the seat 21, and may be disposed inside or outside. In fig. 2, the vehicle seat 2 is shown in a simple planar shape for simplicity of description. In a specific example, as shown in the section A-A of the vehicle seat 2 in the seat section 2-1, the entire interior of the seat 21 may be made of the conductive urethane 22. The conductive urethane 22 may be formed on the occupant side (front surface side) of the interior of the seat 21 as shown in the seat cross section 2-2, or the conductive urethane 22 may be formed on the opposite side (rear surface side) of the interior of the seat 21 as shown in the seat cross section 2-3. Further, as shown in the seat section 2-4, the conductive urethane 22 may be formed inside the seat 21.

The conductive urethane 22 may be disposed outside the seat 21 on the occupant side (front surface side) as shown in the seat cross section 2-5, or the conductive urethane 22 may be disposed outside the seat on the opposite side (rear surface side) as shown in the seat cross section 2-6. In the case where the conductive urethane 22 is disposed outside the seat 21, the conductive urethane 22 may be simply laminated with the seat 21, or the conductive urethane 22 and the seat 21 may be integrated by adhesion or the like. Even when the conductive urethane 22 is disposed outside the seat 21, the conductive urethane 22 is a conductive urethane member and does not interfere with the flexibility of the seat 21.

For simplicity of explanation, an example (seat cross section 2-5) in which the conductive urethane 22 is disposed outside the occupant side (front surface side) of the seat 21 to form the vehicle seat 2 will be described.

In the present embodiment, as shown in fig. 3, the electrical characteristics (that is, the volume resistance value as the resistance value) of the conductive urethane 22 can be detected from signals from a plurality of (two in fig. 3) detection points 75 arranged at a distance from each other. In the example shown in fig. 3, a first detection group #1 is shown in which resistance values are detected based on signals from a plurality of detection points 75 arranged at a diagonal position with a distance therebetween on the conductive urethane 22. The arrangement of the plurality of detection points 75 is not limited to the position shown in fig. 3, and any position may be used as long as the position can detect the electrical characteristics of the conductive urethane 22. Further, regarding the electrical characteristics of the conductive urethane 22, the electrical characteristic detecting unit 76 for detecting the electrical characteristics (that is, the volume resistance value as the resistance value) may be connected to the detection point 75, and the output of the electrical characteristic detecting unit 76 may be used.

The posture state described above includes a biasing state indicating the biasing of the occupant to the vehicle seat 2. The urging state is a state indicating the urging of the occupant to the vehicle seat 2, among the posture states of the occupant. For example, the resistance value detected in the vehicle seat 2 including the seat 21 including the conductive urethane 22 changes at least before and after the application of the pressure stimulus, depending on the deformation of the conductive urethane 22 when the pressure stimulus is applied to the vehicle seat 2. Thus, the resistance value changes before and after the application of force accompanied by pressure stimulus to the vehicle seat 2 by the occupant. Therefore, by detecting the resistance value in time series, that is, detecting a change in the resistance value from a state in which no pressure stimulus is applied to the vehicle seat 2 (for example, detecting a resistance value exceeding a predetermined threshold value), the urging force of the occupant to the vehicle seat 2 can be detected. Specifically, since pressure stimulus is accompanied even if the occupant contacts the vehicle seat 2, the urging state indicating the urging of the occupant against the vehicle seat 2 includes a contact state. Therefore, by disposing the conductive urethane 22 on the vehicle seat 2, contact between the occupant and the vehicle seat 2 can be detected. In addition, even if any one of the position, distribution, and magnitude of the pressure stimulus applied to the vehicle seat 2 changes, the resistance value changes. Thus, it is not impossible to detect the contact state including the contact position of the occupant with the vehicle seat 2 from the resistance values that change in time series.

The urging state is a seating state in which the occupant sits on the vehicle seat 2. That is, the seating state can be determined from the position, distribution, and magnitude of the pressure stimulus, and it is not impossible to detect the seating state including the start of seating from the resistance values that change in time series.

In the present embodiment, the sensor itself for detecting pressure stimulus given by the occupant is constituted by the conductive urethane 22 including the urethane member as the soft material having conductivity, and thus the feeling of unnaturalness when seated is extremely small compared with other conventional sensors. Therefore, the measurement and the riding comfort evaluation can be performed simultaneously without impairing riding comfort during the measurement. This is an advantage over conventional sensors that perform measurement and riding comfort evaluation independently, and is particularly advantageous in measurement evaluation over a long period of time that varies in time series.

In order to improve the accuracy of detecting the electrical characteristics of the conductive urethane 22, more detection points than the detection points (two) shown in fig. 3 may be used.

As an example, the conductive urethane 22 may be formed by arranging one or more rows of a plurality of conductive urethane sheets each having a detection point, and the electrical characteristics of the plurality of conductive urethane sheets may be detected. For example, the conductive urethane sheet 23 (fig. 4) may be arranged to constitute the conductive urethane 22 (fig. 5 and 6). The example shown in fig. 4 shows a first detection group #1 that detects resistance values from signals from detection points 75A arranged at diagonal positions with a distance therebetween, and a second detection group #2 that detects resistance values from signals from detection points 75B arranged at other diagonal positions. In addition, the following is shown in the example shown in fig. 5: the conductive urethane pieces 23 (fig. 4) are arranged (4×1) in the longitudinal direction of the seat 21 to constitute the conductive urethane 22, and the first detection group #1 to the eighth detection group #8 are sequentially constituted. Also, in the example shown in fig. 6, the following is shown: the conductive urethane sheet 23 (fig. 4) is configured by using the first detection group #1 and arranging (4×2) the conductive urethane sheet in the longitudinal direction and the width direction of the seat 21, and the first detection group #1 to the eighth detection group #8 are configured. As shown in fig. 16, the seat 21 may be configured by using conductive polyurethane 22 in which eight detection groups #1 to #8 are provided from the upper portion to the lower portion of the seat back 21B and from the rear portion to the front portion of the seating surface of the seat cushion 21A.

As another example, the detection range on the conductive urethane 22 may be divided, a detection point may be provided for each of the divided detection ranges, and the electrical characteristics may be detected for each of the detection ranges. For example, the region corresponding to the size of the conductive urethane sheet 23 shown in fig. 5 and 6 may be set as a detection range and set to the conductive urethane sheet 22, and the detection points may be arranged for each set detection range and the electrical characteristics may be detected for each detection range.

As shown in fig. 1, the posture state estimation device 1 includes an estimation unit 5. The time-series input data 4 indicating the magnitude of the resistance (resistance value) in the conductive urethane 22 is input to the estimating unit 5. The input data 4 corresponds to the seating state 3 of the occupant indicating the behavior of the occupant such as the posture and movement of the occupant seated in the vehicle seat 2. The estimation unit 5 outputs, as an estimation result, output data 6 indicating a physical quantity (seating state value) indicating the seating state of an occupant seated in the vehicle seat 2. The estimating unit 5 includes a learned model 51 after learning.

The learning model 51 is a model in which learning of the seating state of the occupant, that is, the seating state of the occupant (output data 6) indicating the behavior of the occupant such as the posture and movement of the occupant seated in the vehicle seat 2 is completed from the resistance (input data 4) of the conductive urethane 22 that changes according to the pressure stimulus according to the seating state 3 of the occupant. The learning model 51 is, for example, a model defining a neural network after learning, and is expressed as a set of information of weights (intensities) of connections between nodes (neurons) constituting the neural network.

The learning model 51 is generated by a learning process of the learning processing unit 52 (fig. 7). The learning processing unit 52 performs learning processing using the electrical characteristics (input data 4) in the conductive urethane 22 that change due to the pressure stimulus generated according to the seating state 3 of the occupant. That is, the seating state 3 of the occupant is set as a tag, and a large amount of data obtained by measuring the resistance in the conductive urethane 22 in time series is set as learning data. Specifically, the learning data includes a large number of sets of input data including resistance values (input data 4) and information (output data 6) representing the seating state 3 of the occupant corresponding to the input data. Here, the resistance value (input data 4) of the conductive urethane 22 is associated with time series information by giving information indicating the measurement time to each of them. In this case, the time-series information may be associated with the time-series resistance value set in the conductive urethane 22 while the seating state 3 of the occupant is determined, by applying information indicating the measurement time.

Next, the learning processing unit 52 will be described.

In the learning process performed by the learning processing unit 52, the vehicle seat 2 constituted by the seat 21 provided with the conductive urethane 22 described above is used as the detection unit, and the seating state 3 of the occupant and the resistance value (input data 4) of the conductive urethane 22 are used as the learning data. For example, a seating state indicating that the occupant OP is maintained in the vehicle seat 2 and shows a predetermined posture, movement, or the like, or an operation of the occupant OP in the vehicle seat 2 is instructed, and the resistance value at that time is detected and correlated with the seating state to be set as learning data.

Further, the seating state includes a state representing behavior of the occupant in time series. This state indicating the behavior of the occupant in time series includes a static state based on the posture of the occupant after the movement is stabilized in the vehicle seat 2, and a dynamic state based on a plurality of postures and movements that change in time series in the vehicle seat 2. Further, the electrical characteristic detecting unit 76 (fig. 3) may be connected to the detecting point 75 to detect the electrical characteristic (that is, the volume resistance value as the resistance value).

Specifically, the learning processing unit 52 may be configured as a computer having a CPU, not shown, and may execute learning data collection processing and learning processing. Fig. 8 shows an example of learning data collection processing executed by a CPU, not shown. The learning processing unit 52 instructs the occupant OP of the seating state on the vehicle seat 2 (the conductive urethane 22) in step S100, and acquires the resistance value that changes according to the pressure stimulus according to the seating state in time series in step S102. In the next step S104, the seating state 3 is given as a tag to the acquired time-series resistance values and stored. The learning processing unit 52 repeats the above-described processing until the set of the seating state 3 of the seated person and the resistance value of the conductive urethane 22 reaches a predetermined number or predetermined time (negative determination is performed in step S106 until affirmative determination is performed). Thus, the learning processing unit 52 can acquire and store the resistance values of the conductive urethane 22 in time series for each seating state of the occupant, and the stored set of the resistance values of the conductive urethane 22 in time series for each seating state of the occupant is used as learning data.

The seating state (posture) of the occupant OP can be determined based on at least a part of the relative positional relationship of each portion of the occupant OP with respect to the vehicle seat 2, the distribution, the change in the magnitude and the frequency of each physical quantity, the maintenance, and the like of the pressure stimulus given by each portion. Thus, it is considered that a part of these time-series physical quantities includes a feature representing the seating state (posture) of the occupant OP. In the present embodiment, by using the conductive urethane 22, the electrical characteristics (volume resistance) reflecting these physical quantities can be detected in time series.

Fig. 14 shows an example of electrical characteristics of the vehicle seat 2 including the seat 21 provided with the conductive urethane 22. Fig. 14 (a) to (H) are electrical characteristics of the vehicle seat 2 configured by the seat 21 provided with the conductive polyurethane 22 provided with the eight detection groups #1 to #8 as shown in fig. 16. Fig. 14 (a) to (D) show electrical characteristics of each position in the case where the seat back 21B is divided into four parts from the upper part to the lower part. Fig. 14 (E) to (H) show electrical characteristics of each position in the case where the seat cushion 21A is divided into four parts from the rear portion of the seat surface to the seat surface side. Fig. 14 (a) to (H) show the detection results of the detection groups #1 to #8 shown in fig. 16. The example shown in fig. 14 shows an example of a dynamic state, and shows an electrical characteristic of the occupant in a respiratory state that varies due to deep breathing of the occupant in the sitting state of the occupant.

The electrical characteristics in the sitting state of the occupant shown in fig. 14, that is, the electrical characteristics (time characteristics of the resistance values) detected by the detection groups #1 to #8 of the seat cushion 21A and the seat back 21B, respectively, are characteristic patterns corresponding to the state of the occupant regarding the breathing such as the deep breath, as the sitting state of the occupant.

Fig. 15 shows electrical characteristics of the occupant in a state of being seated, which relates to a series of seating-type operations in which the occupant sits deep on the seat cushion 21A, then sits back on the seat cushion 21A, and sits back deep on the seat cushion 21A. Fig. 15 (a) to (H) show the detection results of the detection groups #1 to #8 shown in fig. 16. Further, deep seated refers to the occupant sitting to the rear side of the seat cushion 21A. In addition, shallow seating means that the occupant sits to the front side of the seat cushion 21A. The electrical characteristics detected by the detection groups #1 to #8 of the seat cushion 21A and the seat back 21B are characteristic patterns corresponding to the state in which the occupant is subjected to a series of re-seating operations as the seated state. For example, the pattern of the detection result detected by the detection group #5 of the seat cushion 21A shown in (E) of fig. 15 is considered to be expressed as a characteristic pattern. The electrical characteristic at time T1 is considered to correspond to the characteristic electrical characteristic in the initially seated state. The electrical characteristic at time T2 is considered to correspond to the characteristic electrical characteristic in the state where the seating is attempted to be re-seated. The electrical characteristic at time T3 is considered to correspond to the characteristic electrical characteristic in the shallow state. The electrical characteristic at time T4 is considered to correspond to the characteristic electrical characteristic in the state of being seated again and being seated again. These feature patterns are also related to the patterns of detection results detected by other detection groups. Therefore, the electrical characteristics detected in time series by the conductive urethane 22 include characteristics of the seating state of the occupant.

The seating state value indicating the seating state may be associated with, for example, an information value indicating the order of deep seated, attempted to be re-seated, shallow seated, and re-deep seated, or may be associated with an information value obtained by extracting a part of deep seated, attempted to be re-seated, shallow seated, and re-deep seated. When a part is extracted as the sitting state, it is sufficient to associate each pattern (pattern of the electrical characteristics shown in fig. 14 (E) other than the electrical characteristics) in a range corresponding to the electrical characteristics (e.g., the electrical characteristics at time T3) of the pattern including at least a part of the features in which the sitting state occurs. These feature patterns are considered to effectively function in the learning process in correspondence with the seating state where the occupant OP is seated on the vehicle seat 2.

Accordingly, by changing the posture of the occupant OP in accordance with the instructed seating state, the pressure stimulus applied to the vehicle seat 2 is changed, and the electrical characteristics corresponding to the change in the pressure stimulus are acquired in time series, whereby the time-series electrical characteristics can be stored in accordance with the seating state (posture state) of the occupant. The time-series electrical characteristic and the set of seating state values representing the instructed seating state are set as learning data.

The state related to the seating mode operation includes a state related to the strength at the time of starting seating, such as forcefully sitting on the vehicle seat 2 and lightly sitting on the vehicle seat 2. The seating state includes various states in addition to a state related to breathing such as deep breathing and a state related to a seating mode operation. For example, the seating state includes a state related to the tilting of the vehicle seat 2, that is, a state related to an operation of changing the angle of the seat back 21B with respect to the seat cushion 21A. These sitting states can be determined by at least some of the relative positional relationship of each portion of the occupant OP with respect to the vehicle seat 2, the distribution, the magnitude, the frequency, and other changes and maintenance of each physical quantity of the pressure stimulus given by each portion, and the like, as in the sitting states (postures) of the occupant OP described above. Therefore, it is considered that a part of the physical quantity in time series includes characteristics indicating the respective sitting states (postures), and the electrical characteristics (volume resistance) reflecting the physical quantity in these states can be detected in time series by using the conductive urethane 22.

An example of the learning data is shown in table form. Table 1 is an example of static state data obtained by associating time-series resistance value data (r) with seating state values as learning data concerning seating states. Table 2 is an example of data obtained by associating the set of characteristic data (J) representing the time-series resistance values detected by each detection group with the seating state value shown in fig. 15. The arbitrary characteristic data (J) included in the set includes a characteristic pattern, which is a characteristic of the seating state. The respective characteristic data (J) are all used as learning data. For example, a plurality of characteristic data (J) and seating state values detected by the vehicle seat 2 are used as learning data.

TABLE 1

Resistance value data in time series	Seating state
		r11，r12，r13，...，r1n	Deep sitting
r21，r22，r23，·..，r2n	Attempting to re-seat
		r31，r32，r33，···，r3n	Shallow sitting
r41，r42，r43，……，r4n	Deep sitting again
		...	...
rk1，rk2，rk3，...，rkn	Deep breathing
		...	...

TABLE 2

Next, the learning process in the learning process section 52 will be described. Fig. 9 is a diagram showing functions of a CPU, not shown, of the learning processing unit 52 in the learning process.

The CPU (not shown) of the learning processing unit 52 includes functional units of a generator 54 and an arithmetic unit 56. The generator 54 has a function of generating an output in consideration of the front-rear relationship of the resistance values acquired in time series as an input.

The learning processing unit 52 holds, as learning data, a large number of sets of the input data 4 (resistance value) and the output data 6, the output data 6 being the seating state 3 of the occupant to which the pressure stimulus is applied to the conductive urethane 22.

The generator 54 includes an input layer 540, an intermediate layer 542, and an output layer 544 to form a well-known Neural Network (NN). The neural network itself is a well-known technology, and therefore detailed description is omitted, and the intermediate layer 542 contains a large number of node groups (neuron groups) having inter-node connections and feedback connections. Data from the input layer 540 is input to the intermediate layer 542, and data of the operation result of the intermediate layer 542 is output to the output layer 544.

The generator 54 is a neural network that generates the generated output data 6A representing the seating state of the occupant from the input data 4 (resistance) that is input. The generated output data 6A is data obtained by estimating the seating state of the occupant who has applied pressure stimulus to the conductive urethane 22 from the input data 4 (resistance). The generator 54 generates generated output data indicating a state close to the seating state of the occupant from the input data 4 (resistance) input in time series. The generator 54 can generate the generated output data 6A close to the seating state of the occupant who has applied pressure stimulus to the conductive urethane 22, which is the vehicle seat 2, by learning using a large amount of input data 4 (resistance). In other words, by capturing the electrical characteristics, which are the input data 4 input in time series, as a pattern and learning the pattern, it is possible to generate the generated output data 6A that approximates the seating state of the occupant who has applied the pressure stimulus to the conductive urethane 22, which is the vehicle seat 2.

The arithmetic unit 56 compares the generated output data 6A with the output data 6 of the learning data, and calculates an error of the comparison result. The learning processing unit 52 inputs the output data 6 that generates the output data 6A and the learning data to the arithmetic unit 56. Accordingly, the arithmetic unit 56 calculates an error between the generated output data 6A and the output data 6 of the learning data, and outputs a signal indicating the calculation result.

The learning processing unit 52 performs learning of the generator 54 to adjust the weight parameter of the connection between the nodes based on the error calculated by the arithmetic unit 56. Specifically, the weight parameters of the connection between the input layer 540 and the node of the intermediate layer 542, the weight parameters of the connection between the node within the intermediate layer 542, and the weight parameters of the connection between the intermediate layer 542 and the node of the output layer 544 in the generator 54 are fed back to the generator 54 by using, for example, a gradient descent method, an error back propagation method, or the like. That is, all the connections between nodes are optimized with respect to the output data 6 of the learning data so as to minimize the error between the generated output data 6A and the output data 6 of the learning data.

The learning model 51 is generated by the learning process of the learning process unit 52. The learning model 51 is represented as a set of information of weight parameters (weights or intensities) of the connection between nodes as a learning result of the learning processing section 52.

Fig. 10 shows an example of the flow of the learning process. The learning processing unit 52 acquires learning data, which is a result of time-series measurement, that is, input data 4 (resistance) having information indicating the seating state of the occupant OP as a tag in step S110. The learning processing unit 52 generates the learning model 51 using learning data that is a result of time-series measurement in step S112. That is, a set of information of weight parameters (weights or intensities) of the connections between nodes, which are learning results obtained by learning using a large amount of learning data as described above, is obtained. Then, in step S114, data representing a set of information of weight parameters (weights or intensities) of the connection between nodes as a learning result is stored as the learning model 51.

The generator 54 may use a recurrent neural network having a function of generating an output in consideration of a context of time-series input, or may use other methods.

In the posture state estimation device 1, the learned generator 54 (i.e., data representing a set of information of weight parameters of the connection between nodes as a learning result) generated by the method exemplified above is used as the learning model 51. If the learning model 51 obtained by sufficient learning is used, it is not impossible to determine the seating state of the occupant from the time-series resistance values in the vehicle seat 2, that is, the conductive urethane 22.

The processing of the learning processing unit 52 is an example of the processing of the learning model generating device of the present disclosure. The posture state estimation device 1 is an example of an estimation unit and an estimation device of the present disclosure. The output data 6, which is information indicating the sitting state 3, is an example of posture state information of the present disclosure.

As described above, the conductive urethane 22 combines the electrical paths in a complicated manner, and exhibits such actions as expansion and contraction, temporary cutting, and new connection of the electrical paths according to the deformation, and as a result, exhibits actions having different electrical characteristics according to the applied force (e.g., pressure stimulus). In this case, the conductive urethane 22 can be regarded as a reservoir for storing data concerning the deformation of the conductive urethane 22. That is, the posture state estimation device 1 can apply the conductive urethane 22 to a network model called physical pool computing (PRC: physical Reservoir Computing) (hereinafter referred to as prcn.). PRC and PRCN are known per se, and thus detailed description is omitted, that is, PRC and PRCN can be preferably applied to estimation of information about deformation of the conductive polyurethane 22.

Fig. 11 shows an example of the learning processing unit 52 that learns the vehicle seat 2 including the conductive urethane 22 as a pool that stores data related to deformation of the vehicle seat 2 including the conductive urethane 22. The conductive urethane 22 exhibits electrical characteristics (resistance values) corresponding to various pressure stimuli, and functions as an input layer for inputting the resistance values and as a reservoir layer for storing data relating to deformation of the conductive urethane 22. Since the conductive urethane 22 outputs different electrical characteristics (input data 4) according to the pressure stimulus applied according to the seating state 3 of the occupant, the applied pressure stimulus 3 (shape of the pressing member) can be estimated on the estimation layer according to the resistance value of the conductive urethane 22. Thus, in the learning process, the estimation layer may be learned.

The posture state estimation device 1 can be realized by, for example, causing a computer to execute a program indicating the functions described above.

Fig. 12 shows an example of a case where the posture state estimation device 1 is configured to include a computer as an execution device that executes processing for realizing various functions.

The computer functioning as the posture state estimation device 1 includes a computer main body 100 shown in fig. 12. 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 Device (HDD), and an input/output interface (I/O) 110. The CPU 102, RAM 104, ROM 106, auxiliary storage device 108, and input/output I/O110 are connected via a bus 112 so as to be able to mutually transmit data and commands. The input/output I/O110 is connected to a communication unit 114 for communicating with an external device, and an operation display unit 116 such as a display and a keyboard. The communication unit 114 functions to acquire input data 4 (resistance) with the vehicle seat 2 including the conductive urethane 22. That is, the communication unit 114 can acquire the input data 4 (resistance) from the electrical characteristic detection unit 76, which is a detection unit, including the vehicle seat 2 in which the conductive urethane 22 is disposed, and is connected to the detection point 75 in the conductive urethane 22.

The auxiliary storage device 108 stores a control program 108P for causing the computer main body 100 to function as the posture state estimation device 1, which is an example of the estimation device of the present disclosure. The CPU 102 reads the control program 108P from the auxiliary storage 108 and expands in the RAM 104 to execute processing. Thus, the computer main body 100 executing the control program 108P operates as the posture state estimating device 1, which is an example of the estimating device of the present disclosure.

Further, the learning model 108M including the learning model 51 and the data 108D including various data are stored in the auxiliary storage device 108. The control program 108P may be provided by a recording medium such as a CD-ROM.

Next, the estimation process in the posture state estimation device 1 implemented by a computer will be described.

Fig. 13 shows an example of a flow of the estimation process based on the control program 108P executed in the computer main body 100.

When the power is turned on to the computer main body 100, the estimation process shown in fig. 13 is executed by the CPU 102. That is, the CPU 102 reads the control program 108P from the auxiliary storage device 108 and expands in the RAM 104 to execute processing.

First, the CPU 102 acquires the learning model 51 by reading the learning model 51 from the learning model 108M of the auxiliary storage device 108 and expanding in the RAM 104 in step S200. Specifically, a network model (see fig. 9 and 11) to be a connection between nodes based on the weight parameters expressed as the learning model 51 is developed in the RAM 104. Thus, the learning model 51 that realizes the connection between the nodes based on the weight parameters is constructed.

Next, in step S202, the CPU 102 acquires unknown input data 4 (resistance) in time series via the communication unit 114, and the unknown input data 4 is a target for estimating the shape of the pressing member due to the pressure stimulus applied to the conductive urethane 22.

Next, the CPU 102 estimates output data 6 (unknown seating state) corresponding to the input data 4 (resistance) acquired in step S202 using the learning model 51 acquired in step S200 in step S204.

Then, in the next step S206, the output data 6 (seating state of the occupant) as the estimation result is output via the communication section 114, and the present processing routine is ended.

The estimation process shown in fig. 13 is an example of a process performed by the estimation method of the present disclosure.

As described above, according to the present disclosure, the seating state of an occupant can be estimated from the input data 4 (resistance) that varies in accordance with the pressure stimulus applied to the conductive urethane 22 based on the seating state 3. That is, the seating state of the unknown occupant can be estimated without using a special device, a large-sized device, or directly measuring the deformation of the soft member.

Further, since the electrical characteristics obtained by each detection group are changed according to the behavior of the occupant OP, and the seating state is reflected in the electrical characteristics (time-series resistance), the seating state of the occupant can be estimated from the time-series resistance value in the conductive urethane 22. That is, even in various seating states, the seating state of the occupant can be determined by using the learning model described above, and thus the seating state of the occupant can be estimated.

The posture state estimation device 1 using the learning model 51 learned by the learning process described above confirms that the electrical characteristics of the conductive urethane 22 in various seating states, which are unknown, are input, so that the corresponding seating state of the seated person can be estimated.

As described above, in the present disclosure, a case where conductive polyurethane is applied as an example of the flexible member is described, but the flexible member is not limited to conductive polyurethane, and it is needless to say.

The technical scope of the present disclosure is not limited to the scope described in the above embodiment. Various changes and modifications may be made to the above-described embodiments without departing from the spirit, and the manner in which such changes and modifications are made is also included in the technical scope of the present disclosure.

In the above embodiment, the case where the estimation process and the learning process are realized by the software configuration based on the process using the flowchart has been described, but the present invention is not limited to this, and the processes may be realized by a hardware configuration, for example.

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

< second embodiment >

The second embodiment is explained below. The second embodiment can follow the first embodiment. Therefore, the following description will be mainly made on portions different from those of the first embodiment, and the same reference numerals are given to the same portions as those of the first embodiment, and detailed description thereof will be omitted.

In the second embodiment, the "posture state of the occupant seated in the vehicle seat" is the "posture state accompanied by water of the occupant seated in the vehicle seat". Specifically, the term "posture state of an occupant sitting in a vehicle seat accompanied by water" is a concept including a state of an occupant that is deformed by a soft material, applies pressure to the soft material, and applies moisture generated by perspiration or the like, and includes behaviors of an occupant such as posture and movement of an occupant sitting in a vehicle seat. The posture state is a seating state in which an occupant is seated in the vehicle seat.

In the second embodiment, the electrical characteristics that change according to the deformation of the soft material are affected by the moisture (water content) applied by the perspiration of the occupant or the like. The physical quantity (resistance value) representing the electrical characteristic corresponding to the posture state of the occupant accompanied by water is changed with respect to the physical quantity (resistance value) representing the electrical characteristic corresponding to the posture state of the occupant not accompanied by water. That is, even if the occupant is in the same posture state, the electrical characteristics of the soft material are different depending on the moisture content, so that it is possible to determine whether the occupant sweats or the like contains moisture.

The estimating device according to the second embodiment estimates the posture state of the occupant sitting in the vehicle seat with water based on the electrical characteristics of the conductive soft material provided in the vehicle seat using the learned learning model. The posture state accompanied by water includes, for example, a posture state accompanied by perspiration, a posture state accompanied by water generated by wetting of at least one of the body and clothing of the occupant, and the like. The soft material can be disposed in a vehicle seat. The learning model uses, as learning data, time-series electrical characteristics when pressure and moisture are applied to a flexible material having conductivity, and posture state information indicating a posture state accompanied with moisture of an occupant seated in a vehicle seat when pressure and moisture are applied to the flexible material. The learning model learns to take as input a time-series electrical characteristic and to output posture state information indicating a posture state accompanied with water of an occupant seated in the vehicle seat corresponding to the time-series electrical characteristic.

In the second embodiment, a value indicating pressure stimulus (hereinafter, referred to as a moisture pressure value) including moisture applied to a vehicle seat is applied as a physical quantity for deforming the conductive polyurethane. The moisture pressure value in this case is generated by the posture state of the occupant of the vehicle seat accompanying water due to perspiration or the like. In the present embodiment, a seating state accompanied by sweating of an occupant seated in a vehicle seat is applied as the posture state. The resistance value of the conductive polyurethane is applied as a physical quantity that changes in response to pressure stimulation including moisture.

Fig. 17 to 19 and fig. 22 to 25 in the second embodiment correspond to fig. 1, fig. 7 to 11 and fig. 13 in the first embodiment. Fig. 17 to 19 and fig. 22 to 25 can be described by changing the "posture state" in the description of fig. 1, 7 to 11 and 13 of the first embodiment to the "posture state accompanied by water, changing the" pressure stimulus "to the" pressure stimulus including water ", and changing the" sitting state "to the" sitting state accompanied by water ".

Fig. 20 shows an example of electrical characteristics of the vehicle seat 2 including the seat 21 provided with the conductive urethane 22. The example of fig. 20 shows a case where the electrical characteristics (resistance value) are changed in time series in the case where water is sprayed to the conductive urethane 22 in a spray manner.

Fig. 21 shows another example of the electrical characteristics of the vehicle seat 2 constituted by the seat 21 provided with the conductive urethane 22. The example of fig. 21 shows a case where the electrical characteristics (resistance values) are changed in time series in the case where water is sprayed onto the conductive urethane 22 in a spray manner, as in the example of fig. 20. Fig. 21 (B) is an enlarged view of the X portion of fig. 21 (a), and fig. 21 (C) is an enlarged view of the Y portion of fig. 21 (a).

As shown in fig. 20 and 21, it was confirmed that the posture state of the occupant accompanied by water content was estimated from the resistance value which was changed in time series according to the water content of the conductive urethane 22, from the electrical characteristics of the conductive urethane 22, which was changed according to the water content of the conductive urethane 22, and the water content was reflected. That is, even in various water-containing posture states such as sweating, the results concerning the water-containing posture states of the occupant can be separated by using the learning model 51 obtained by learning, respectively, and the water-containing posture states of the occupant can be discriminated.

By changing the posture of the occupant OP in accordance with the instructed sitting state accompanied by water content to change the pressure stimulus containing water to be given to the vehicle seat 2, the electrical characteristics corresponding to the change in the pressure stimulus containing water content are acquired in time series, whereby the time-series electrical characteristics can be stored in association with the sitting state (posture state) accompanied by water content of the occupant. The set of the time-series electrical characteristics and seating state values indicating the instructed seating state accompanied by water content becomes learning data.

Next, an example of the learning data is shown in the following table. Table 3 is an example of static state data obtained by correlating time-series resistance value data (r) with seating state values as learning data concerning seating states accompanied by water.

TABLE 3

Resistance value data in time series	Seating state
		r11，r12，r13，...，r1n	Seating state 1
r21，r22，r23，...，r2n	Seating state 2
		r31，r32，r33，...，r3n	Seating state 3
r41、r42，r43...，r4n	Seating state 4
		...	...
rk1，rk2，rk3，...，rkn	Seating state k
		...	...

The learning processing section 52 shown in fig. 18 performs learning processing using the learning data as described above.

< third embodiment >

The third embodiment is explained below. The third embodiment can follow the second embodiment. Therefore, in the third embodiment, a description will be given mainly of a portion different from the second embodiment, and the same reference numerals are given to the same portions as those of the second embodiment, and detailed description thereof will be omitted.

Conventionally, it has been known that sweat is generated when an occupant moves. Thus, whether or not motion is active can be determined based on the degree of sweating of the occupant.

The control device according to the third embodiment estimates the posture state accompanied by water of the occupant seated in the vehicle seat based on the electrical characteristics of the conductive soft material provided in the vehicle seat using the learned learning model, and controls the environment adjustment device including at least one of the temperature adjustment device of the vehicle seat and the air conditioning device provided with the cabin of the vehicle seat using the estimated posture state information.

As described in the second embodiment, the posture state accompanied by water includes, for example, a posture state accompanied by perspiration, a posture state accompanied by water due to wetting of at least one of the body of the occupant and the clothing, and the like. In the third embodiment, the posture state accompanied by sweating includes a posture state accompanied by motion sickness.

Fig. 26 shows an example of the structure of the control device 10 according to the third embodiment. The control device 10 is configured by adding a control unit 7 and an environment adjustment device 8 to the estimation device 1 shown in fig. 17.

The control unit 7 in the control device 10 uses the posture state information output by the above estimation process to control the environment adjustment device 8 for adjusting the environment of the occupant OP. Here, adjusting the environment of the occupant OP means adjusting at least one of the temperature and the humidity in the vicinity of the occupant OP. The environment adjustment device 8 includes a temperature adjustment device 8A of the vehicle seat 2 and an air conditioning device 8B of the vehicle cabin. The environment adjustment device 8 may be provided with either one of the temperature adjustment device 8A and the air conditioning device 8B. The temperature adjustment device 8A is a device for adjusting at least one of temperature and humidity in the vicinity of an occupant OP seated in the vehicle seat 2, and includes, for example, at least one of a seat cooler for cooling the vehicle seat 2 and a seat heater for heating the vehicle seat 2.

The seat cooler is configured such that, for example, a fan (not shown) capable of changing the air volume level is incorporated in the vehicle seat 2, and the temperature and/or humidity in the vicinity of the occupant OP are adjusted by rotating the fan to blow cooled air toward the occupant OP. By blowing cooling air to suppress the temperature rise of the vehicle seat 2, it is possible to remove moisture generated by, for example, perspiration. The seat cooler may have the following structure: the fan is reversely rotated to suck air, thereby adjusting at least one of the temperature and the humidity in the vicinity of the occupant OP. The seat cooler may be implemented by circulating a refrigerant such as water, or by using other known methods.

The seat heater is, for example, a fan incorporated in the vehicle seat 2, and adjusts at least one of the temperature and the humidity in the vicinity of the occupant OP by blowing the heated wind toward the occupant OP. The seat heater may have the following structure: for example, by embedding an electric heating wire in the vehicle seat 2, and generating heat by passing an electric current through the electric heating wire, the vehicle seat 2 is heated, and at least one of the temperature and the humidity in the vicinity of the occupant OP is adjusted.

The air conditioner 8B is a device for adjusting at least one of the temperature and the humidity in the vehicle cabin in which the vehicle seat 2 is installed, and cools or heats the vehicle cabin by rotating a fan (not shown) to blow cooled air or heated air into the vehicle cabin. The air conditioner 8B has a dehumidification function and is controlled to have a set temperature and humidity.

When it is estimated that the posture state of the occupant OP is a posture state accompanied by sweating, the control unit 7 operates at least one of the temperature control device 8A and the air conditioning device 8B, for example, to reduce the temperature in the vicinity of the occupant OP. This suppresses the temperature rise of the vehicle seat 2, and moisture generated by perspiration is removed. That is, perspiration in a state where the occupant OP is seated on the vehicle seat 2 is suppressed, and a comfortable seated state can be obtained.

When the environment of the occupant OP is adjusted by the temperature control device 8A, for example, a data table in which the state information of the posture accompanied by sweating is correlated with the air volume level (for example, strong, medium, weak, etc.) of the fan of the temperature control device 8A is stored in advance, and the fan of the temperature control device 8A is operated at the air volume level corresponding to the estimated state of the posture accompanied by sweating.

When the environment of the occupant OP is adjusted by the air conditioner 8B, a data table is stored in advance, the data table being obtained by associating the state information of the posture accompanied by sweat with the set temperature and the set humidity of the air conditioner 8B, respectively, and the air conditioner 8B is operated at the set temperature and the set humidity corresponding to the estimated state of the posture accompanied by sweat. For example, when it is estimated that the posture state of the occupant OP sitting in the vehicle seat 2 is a posture state accompanied by perspiration, control is performed to operate the air conditioner 8B to reduce the temperature and humidity in the vehicle cabin. However, when the temperature in the vehicle cabin is high and the humidity is in an appropriate range, only the temperature in the vehicle cabin may be reduced. In addition, when the temperature in the vehicle cabin is in an appropriate range and the humidity is high, only the humidity in the vehicle cabin may be reduced. By this control, perspiration during sitting is suppressed, and a comfortable sitting state can be obtained.

In addition, when the estimated posture state accompanied by sweating is a posture state accompanied by motion sickness, the control unit 7 may control to output a message for notifying that the posture state is motion sickness from a speaker provided in the vehicle.

Fig. 27 shows an example of a flow of estimation/control processing based on the control program 108P executed in the computer main body 100.

The processing of steps S200 to S204 is the same as the processing of steps S200 to S204 of fig. 25.

In step S206, output data 6 (seating state of the occupant accompanied by water) as the estimation result is output to the auxiliary storage device 108.

Then, in step S208, the CPU 102 uses the output data 6 as the estimation result to execute environment adjustment control for adjusting the environment in the vicinity of the occupant OP. That is, for example, when it is estimated that the posture state of the occupant OP is a posture state accompanied by sweating, at least one of the temperature control device 8A and the air conditioning device 8B is operated to reduce the temperature in the vicinity of the occupant OP. Specifically, for example, the fan of the temperature control device 8A is operated at an air volume level corresponding to the estimated posture state accompanied by perspiration. The temperature and humidity set by the air conditioner 8B are set to be lower than the current temperature and humidity in the vehicle cabin, for example, to air-condition the vehicle cabin. By such environment adjustment control, perspiration during sitting is suppressed, and a comfortable sitting state can be obtained. In addition, when the estimated posture state accompanied by sweating is a posture state accompanied by motion sickness, the control unit outputs a message for notifying the state of motion sickness from a speaker provided in the vehicle. Thus, for example, the driver takes care of the driver taking account of the occupant in the motion sickness state and takes care of the driver, and motion sickness can be reduced.

As described above, according to the third embodiment, the sitting state of the occupant accompanied by water can be estimated from the input data 4 (resistance) that changes in accordance with the pressure stimulus including water applied to the conductive polyurethane 22 based on the sitting state 3 accompanied by water, and the environment of the occupant can be adjusted using the estimation result. That is, the seating state of the unknown occupant accompanied by water can be estimated without using a special device, a large-sized device, or directly measuring the deformation of the soft member, and the environment of the occupant can be adjusted using the estimation result.

< fourth embodiment >

The fourth embodiment is explained below. The fourth embodiment can follow the first embodiment. Therefore, in the fourth embodiment, a description will be given mainly of portions different from those of the first embodiment, and the same reference numerals are given to the same portions as those of the first embodiment, and detailed description thereof will be omitted.

In the fourth embodiment, the "posture state with movement of an occupant seated in the vehicle seat" is a concept including a state of an occupant that deforms a soft material and applies pressure to the soft material, and includes behaviors of the occupant such as posture and movement of the occupant seated in the vehicle seat. In the present embodiment, a description will be given of a case where the posture state is a seating state in which an occupant is seated in the vehicle seat. In this case, the "sitting state of the occupant of the vehicle seat accompanying the movement" is a concept including a state related to the breathing of the occupant of the vehicle seat, a state related to the movement of the seating mode of the occupant, and a state related to the tilting movement of the occupant of the vehicle seat. The state related to the breathing of the occupant of the vehicle seat includes a state indicating whether or not deep breathing is performed, a state indicating whether or not yawning is performed, and the like. The state related to the seating mode of the occupant of the vehicle seat means a state of a re-seating operation such as a deep seating operation on the vehicle seat, a re-seating operation on the vehicle seat, a state of shaking the body back and forth, a state related to the strength of the vehicle seat when sitting forcefully on the vehicle seat, a state of changing the pedal related to the driving of the vehicle such as stepping on the accelerator pedal and the brake pedal, a state of changing the left and right foot, a state of lifting one of the left and right legs, or the left and right legs, and the like. The state related to the tilting operation of the occupant of the vehicle seat includes a state indicating an operation of changing the angle of the seat back of the vehicle seat, and the like.

Fig. 28 to 34 in the fourth embodiment correspond to fig. 1, 7 to 11, and 13 in the first embodiment. Fig. 28 to 34 can be described by changing the "posture state" in the description of fig. 1, 7 to 11, and 13 of the first embodiment to the "posture state accompanied by exercise" and changing the "sitting state" to the "sitting state accompanied by exercise".

The sitting state accompanied by movement includes a state indicating time-series behavior of the occupant. The state representing the time-series behavior of the occupant includes a dynamic state based on a plurality of gestures and motions that change in time-series on the vehicle seat 2. The electrical characteristics (that is, the volume resistance value as the resistance value) may be detected by connecting the electrical characteristic detecting unit 76 (fig. 3) to the detection point 75.

The above-described respiration-related state includes a state indicating whether or not a yawning is made, in addition to a deep respiration. The state related to the sitting mode operation includes a state related to the strength at the time of starting sitting, such as forcefully sitting on the vehicle seat 2 or lightly sitting on the vehicle seat 2, a state indicating an operation of changing pedals related to driving of the vehicle, such as stepping on an accelerator pedal or a brake pedal, a state indicating an operation of changing the left foot and the right foot, and the like. Fig. 35 shows electrical characteristics of an occupant in a case where the occupant is yawed in a seating state accompanied by movement. Fig. 36 shows electrical characteristics when the occupant performs the operation of changing the pedal in a sitting state with the movement of the occupant. Fig. 37 shows electrical specification in the case where the occupant performs the operation of replacing the tilted left and right feet. Fig. 35 (a) to (H), fig. 36 (a) to (H), and fig. 37 (a) to (H) are the detection results of the detection groups #1 to #8 shown in fig. 16, respectively. The sitting state accompanied by movement includes various states in addition to a state related to breathing such as deep breathing and yawning and a state related to a sitting-mode operation. For example, the seating state accompanied by movement includes a state related to tilting operation of the occupant of the vehicle seat 2, that is, a state related to operation of changing the angle of the seat back 21B with respect to the seat cushion 21A. These sitting states accompanied by movement can be determined from at least some of the relative positional relationship of each portion of the occupant OP with respect to the vehicle seat 2, the distribution, the magnitude, the frequency, and other changes in each physical quantity, maintenance, and the like of the pressure stimulus generated by each portion, as in the sitting states (postures) accompanied by movement of the occupant OP described above. Therefore, it is considered that a part of the physical quantity in time series includes a feature indicating a sitting state (posture) in which each movement is accompanied, and by using the conductive urethane 22, it is possible to detect an electrical characteristic (volume resistance) reflecting the physical quantity in these states in time series.

According to the fourth embodiment, the seating state of the occupant accompanied by the movement can be estimated. This enables the comfort, fatigue, and motion sickness of the occupant to be estimated.

In addition to the above-described change in seating state, when the blood flow is not smooth due to sitting for a long period of time, and the buttocks and the like feel numbness, the occupant may want to move the buttocks back and forth and left and right to promote blood flow and eliminate numbness. Further, since the same posture is maintained for a long period of time and the muscles in the same portion are fatigued, not only the sitting state is changed, but also the body is required to be moved to relieve the rigidity of the muscles. In the present embodiment, since such physical movement of the occupant can be perceived, the present invention can be applied to determination of fatigue of the occupant.

< fifth embodiment >

A fifth embodiment will be described below. The fifth embodiment can be applied to the fourth embodiment. Therefore, in the fifth embodiment, a description will be given mainly of portions different from those of the fourth embodiment, and the same reference numerals are given to the same portions as those of the fourth embodiment, and detailed description thereof will be omitted.

In the fifth embodiment, the "sitting state of the occupant of the vehicle seat accompanying the movement" is a concept including a state related to the breathing of the occupant of the vehicle seat, a state related to the movement of the seating mode of the occupant, and a state related to the change in the posture of the occupant of the vehicle seat. The state related to the change in the posture of the occupant of the vehicle seat includes a state indicating an operation of changing the body position in a state where the occupant is seated in the vehicle seat, and the like.

Further, the state related to the respiration of the occupant is considered to be closely related to the state related to the physical and psychological condition of the occupant. Here, the state related to the physical and psychological condition of the occupant means a state indicating at least one of the fatigue degree, the tension degree, and the wakefulness degree of the occupant. For example, it is generally believed that the faster the breathing, the higher the tension of the occupant. As described above, since the state related to the respiration of the occupant is considered to be closely related to the state related to the physical and mental condition of the occupant, if the state related to the respiration of the occupant can be estimated as the seating state with the movement of the occupant seated in the vehicle seat, the state related to the physical and mental condition of the occupant can also be estimated. For example, by using table data representing a correspondence relationship between a state relating to the respiration of the occupant and a state relating to the physical and mental conditions of the occupant, the state relating to the physical and mental conditions of the occupant can be determined from the state relating to the respiration of the occupant.

The state related to the action of the seating system of the occupant is considered to be closely related to the state related to the physical and psychological condition of the occupant. For example, it is considered that the higher the frequency of the re-seating operation, the less calm the occupant and the higher the tension. As described above, since the state related to the action of the seating system of the occupant is considered to be closely related to the state related to the physical and mental condition of the occupant, if the state related to the seating system of the occupant can be estimated as the seating state accompanied by the movement of the occupant seated in the vehicle seat, the state related to the physical and mental condition of the occupant can also be estimated. For example, by using table data indicating a correspondence relationship between a state relating to the seating mode of the occupant and a state relating to the physical and mental conditions of the occupant, the state relating to the physical and mental conditions of the occupant can be determined from the state relating to the seating mode of the occupant.

In addition, the state related to the change in the posture of the occupant is considered to be closely related to the state related to the physical and psychological condition of the occupant. For example, when the posture of the occupant is hardly changed, it is considered that the wakefulness is low, that is, the state is close to dozing. As described above, since the state relating to the change in the posture of the occupant is considered to be closely related to the state relating to the physical and mental condition of the occupant, if the state relating to the change in the posture of the occupant can be estimated as the sitting state with the movement of the occupant sitting in the vehicle seat, the state relating to the physical and mental condition of the occupant can also be estimated. For example, by using table data representing a correspondence relationship between a state relating to a change in the posture of the occupant and a state relating to the physical and psychological condition of the occupant, the state relating to the physical and psychological condition of the occupant can be determined from the state relating to the change in the posture of the occupant.

In the fifth embodiment, the "sitting state of the occupant of the vehicle seat accompanying movement" includes a state related to the physical and psychological condition of the occupant.

Fig. 38 shows an example of the structure of the control device 11 according to the fifth embodiment. The control device 11 is configured by adding a control unit 71 and a vehicle device 81 to the estimation device 1 shown in fig. 28.

The control unit 71 in the control device 11 controls the vehicle device 81 constituting the vehicle on which the vehicle seat 2 is mounted, using the posture state information output by the estimation process. In the present disclosure, the vehicle device 81 includes a seat drive device 81A and a sound output device 81B as an example. The seat drive device 81A is a device for performing at least one of adjustment of the position of the seat cushion of the vehicle seat 2 with respect to the vehicle in the front-rear direction and adjustment of the angle (reclining angle) of the seat back. The sound output device 81B is a device that outputs various sound messages into the vehicle cabin to notify the occupant.

The control unit 71 controls the seat driving device 81A so that the position of the vehicle seat 2 is a position corresponding to the estimated state of mind and body of the occupant. For example, when the fatigue is estimated to be high as a state related to the physical and psychological condition of the occupant, the control unit 71 controls the seat drive device 81A so that the reclining angle is small, that is, the seatback is reclined. This reduces the fatigue of the occupant. In addition, the control unit 71 may cause the sound output device 81B to output a sound message prompting a rest when the fatigue degree is estimated to be high as a state related to the physical and psychological condition of the occupant.

In addition, for example, when the state of the occupant is estimated to be a low wakefulness, that is, a state close to dozing, the control unit 71 controls the seat drive device 81A so that the reclining angle is increased, that is, so that the seatback is raised. This can improve the degree of wakefulness of the occupant. The control unit 7 may cause the sound output device 81B to output a notice-urging sound message when the wakefulness is estimated to be low as a state related to the physical and mental conditions of the occupant. The control unit 71 may cause the sound output device 81B to output a sound message prompting attention to drive when the state of the occupant regarding the physical and mental condition is estimated to be low in tension, that is, relaxed.

It is considered that if the occupant is different, the state relating to the seating mode operation of the vehicle seat 2 is also different, for example. Therefore, the control unit 71 may control the seat drive device 81A to adjust the position of the vehicle seat 2 in accordance with the state related to the seating operation, for example, the operation to start seating the vehicle seat 2. Thus, the position of the vehicle seat 2 can be adjusted for each occupant, and therefore, for example, in the case where the vehicle seat 2 is a driver seat, the position of the vehicle seat 2 is automatically adjusted when the driver as an occupant sits on the driver seat, without the driver having to adjust the position of the vehicle seat 2 by himself.

Fig. 39 shows an example of a flow of estimation/control processing based on the control program 108P executed in the computer main body 100.

The processing of steps S200 to S204 is the same as the processing of steps S200 to S204 of fig. 34.

In step S206, output data 6 (seating state of the occupant with movement) as the estimation result is output to the auxiliary storage device 108.

Then, in step S208, the CPU 102 controls the vehicle apparatus 81 using the output data 6 as the estimation result. For example, as described above, the seat drive device 81A is controlled so that the position of the vehicle seat 2 is set to a position corresponding to the estimated state related to the physical and mental conditions of the occupant. For example, when the fatigue is estimated to be high as a state related to the physical and psychological condition of the occupant, the control unit 71 controls the seat drive device 81A so that the reclining angle is small, that is, the seatback is reclined. For example, when the fatigue is estimated to be high as a state related to the physical and psychological condition of the occupant, the voice output device 81B is caused to output a voice message prompting rest. In addition, for example, when the state of the occupant is estimated to be a low wakefulness, that is, a state close to dozing, the seat drive device 81A is controlled so that the reclining angle becomes large, that is, the seat back is raised. In addition, for example, when the wakefulness is estimated to be low as a state related to the physical and psychological condition of the occupant, the sound output device 81B is caused to output a notice-urging sound message. For example, when the state of the occupant regarding the physical and mental condition is estimated to be low in tension, that is, relaxed, the sound output device 81B is caused to output a sound message prompting attention to drive.

The estimation control process shown in fig. 39 is an example of a process executed by the control method of the present disclosure, and various kinds of control can be performed in addition to this.

For example, when it is estimated that the vehicle seat 2 other than the driver's seat, such as the passenger seat, is not seated, control may be performed to close the occupant airbag device of the vehicle seat 2, and when it is estimated that the vehicle seat 2 other than the driver's seat, such as the passenger seat, is seated, control may be performed to open the airbag device. This prevents the airbag device from operating uselessly.

The acquired posture state information may be collected and used for various evaluations. For example, as the sitting state accompanied by the exercise, a sitting state accompanied by a pedal operation including at least one of a brake pedal and an accelerator pedal may be estimated, and posture state information of the estimated sitting state accompanied by the pedal operation may be collected and used for evaluation of the driving technique.

The posture state information may be collected for each type of the vehicle seat 2 and used for evaluation of each type of the seat characteristics and riding comfort of the vehicle seat 2.

Further, the disclosures of Japanese patent application Nos. 2021-122012, 2021-122013, 2021-122014, 2021-122021 and 2021-122022 are incorporated herein by reference in their entirety. All documents, patent applications, and technical standards described in the present specification are incorporated in the present specification by reference to the same extent as if each document, patent application, and technical standard were specifically and individually described to be incorporated by reference.

Claims

1. An estimation apparatus comprising:

2. The estimation apparatus according to claim 1, wherein,

the electrical characteristic is a volume resistance,

3. The estimation apparatus according to claim 2, wherein,

4. An estimation device according to claim 2 or 3, wherein,

5. The estimation apparatus according to any one of claims 1 to 4, wherein,

6. The estimation apparatus according to any one of claims 1 to 5, wherein,

7. The estimation apparatus according to claim 6, wherein,

8. The estimation apparatus according to claim 6 or 7, wherein,

9. The estimation apparatus according to any one of claims 1 to 5, wherein,

10. The estimation apparatus according to claim 9, wherein,

11. A control apparatus comprising:

the estimation device according to any one of claims 6 to 8; and

12. The control device according to claim 11, wherein,

13. A control apparatus comprising:

the estimation device according to claim 9 or 10; and

14. The control device according to claim 13, wherein,

15. The control device according to claim 14, wherein,

16. The control device according to claim 14, wherein,

17. The control device according to claim 14, wherein,

18. An estimation method in which a computer performs the following processing:

19. An estimation program for causing a computer to execute: