Vehicle Riding Comfort Prediction Device, Method, and Program

Abstract

To provide a vehicle ride comfort prediction device, a method, and a program capable of more appropriately predicting the ride comfort of passengers.SOLUTION: A vehicle ride comfort prediction device 1000 according to the present invention predicts the ride comfort of passengers in a vehicle and includes: a vibration acquisition unit 1 which acquires vibration data representing the vibrations experienced by the passenger; a biological signal prediction unit 22 which utilizes a first machine learning model that has been machine-learned to predict a biological signal of the passenger on the basis of the vibration data acquired by the vibration acquisition unit 1; and a ride comfort prediction unit 23 which utilizes a second machine learning model that has been machine-learned to predict the ride comfort of the passenger on the basis of the biological signal of the passenger predicted by the biological signal prediction unit 22.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a vehicle ride comfort prediction device, a vehicle ride comfort prediction method, and a vehicle ride comfort prediction program for predicting the ride comfort of vehicle occupants.

Background Art

[0002] With the development of recent autonomous driving technology, in addition to route selection and safety, optimization of steering and acceleration / deceleration including the ride comfort felt by occupants has been under consideration. Among these, quantification of the subjective ride comfort and construction of a prediction model are major issues, and a ride comfort prediction model has been proposed, for example, in Non-Patent Document 1. The ride comfort prediction model disclosed in this Non-Patent Document 1 is a machine learning model that predicts ride comfort from vehicle vibration data using an emotion structure model based on Wundt's three-dimensional theory of emotion.

Prior Art Documents

Non-Patent Documents

[0003]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the ride comfort prediction model disclosed in Non-Patent Document 1, there is variation in the emotion structure vectors for each subject, and there is fluctuation in the evaluation values, so there is room for improvement.

[0005] The present invention has been made in view of the above circumstances, and an object thereof is to provide a vehicle ride comfort prediction device, a vehicle ride comfort prediction method, and a vehicle ride comfort prediction program that can more appropriately predict the ride comfort of occupants.

Means for Solving the Problems

[0006] As a result of various studies, the present inventors have found that the above object is achieved by the following present invention. That is, a vehicle ride comfort prediction device according to one aspect of the present invention is a device that predicts the ride comfort of a passenger in a vehicle, and includes a vibration acquisition unit that acquires vibration data representing the vibration of the passenger, and based on the vibration data acquired by the vibration acquisition unit, a biological signal prediction unit that predicts the biological signal of the passenger using a machine learning-trained first machine learning model, and a ride comfort prediction unit that predicts the ride comfort of the passenger using a machine learning-trained second machine learning model based on the biological signal of the passenger predicted by the biological signal prediction unit.

[0007] Ride comfort is considered to affect the biological signal. Since the above vehicle ride comfort prediction device predicts ride comfort via the biological signal from the vibration data, it can predict the ride comfort of the passenger more appropriately.

[0008] In another aspect, in the above vehicle ride comfort prediction device, the vibration data is three-axis acceleration data, and the biological signal is any one of an electrocardiogram, an RR interval which is an interval between R waves in the electrocardiogram, and a pulse wave.

[0009] According to this, a vehicle ride comfort prediction device can be provided that uses three-axis acceleration data for the vibration data and any one of an electrocardiogram, an RR interval, and a pulse wave for the biological signal.

[0010] In another aspect, in the above vehicle ride comfort prediction device, the explanatory variables of the first machine learning model are the first and second principal components obtained by principal component analysis of the respective conversion results obtained by wavelet-transforming each acceleration data of each axis in the three-axis acceleration data.

[0011] Such a vehicle ride comfort prediction device can simplify the first machine learning model because it compresses the explanatory variables in dimension by principal component analysis.

[0012] In another aspect, in the above-described vehicle ride comfort prediction device, the ride comfort is represented by an emotion vector representing the center of gravity point in each evaluation value of all items in Russell's annular model.

[0013] According to this, a vehicle ride comfort prediction device can be provided in which the ride comfort is represented by an emotion vector representing the center of gravity point in each evaluation value of all items in Russell's annular model.

[0014] A vehicle ride comfort prediction method according to another aspect of the present invention is a method for predicting the ride comfort of a vehicle occupant, including a vibration acquisition step of acquiring vibration data representing the vibration of the occupant, a biological signal prediction step of predicting the biological signal of the occupant using a trained first machine learning model based on the vibration data acquired in the vibration acquisition step, and a ride comfort prediction step of predicting the ride comfort of the occupant using a trained second machine learning model based on the biological signal of the occupant predicted in the biological signal prediction step. A vehicle ride comfort prediction program according to another aspect of the present invention is a program for predicting the ride comfort of a vehicle occupant, which causes a computer to function as a vibration acquisition unit that acquires vibration data representing the vibration of the occupant, a biological signal prediction unit that predicts the biological signal of the occupant using a trained first machine learning model based on the vibration data acquired by the vibration acquisition unit, and a ride comfort prediction unit that predicts the ride comfort of the occupant using a trained second machine learning model based on the biological signal of the occupant predicted by the biological signal prediction unit.

[0015] Such a vehicle ride comfort prediction method and vehicle ride comfort prediction program can predict the ride comfort of an occupant more appropriately because they predict the ride comfort via a biological signal from vibration data.

Advantages of the Invention

[0016] The vehicle ride comfort prediction device, vehicle ride comfort prediction method, and vehicle ride comfort prediction program according to the present invention can predict the ride comfort of an occupant more appropriately.

Brief Description of the Drawings

[0017]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Mode for Carrying Out the Invention

[0018] Hereinafter, one or more embodiments of the present invention will be described with reference to the drawings. However, the scope of the invention is not limited to the disclosed embodiments. In each figure, components denoted by the same reference numerals are the same components, and the description thereof will be omitted as appropriate. In this specification, when referring generically, reference numerals without subscripts are used, and when referring to individual components, reference numerals with subscripts are used.

[0019] The vehicle ride comfort device in the embodiment is a device that predicts the ride comfort of the vehicle occupants. This vehicle ride comfort prediction device includes a vibration acquisition unit, a biological signal prediction unit, and a ride comfort prediction unit. The vibration acquisition unit acquires vibration data representing the vibration of the occupant. The biological signal prediction unit predicts the biological signal of the occupant using a first machine learning model that has been machine-learned based on the vibration data acquired by the vibration acquisition unit. The ride comfort prediction unit predicts the ride comfort of the occupant using a second machine learning model that has been machine-learned based on the biological signal of the occupant predicted by the biological signal prediction unit. Hereinafter, such a vehicle ride comfort prediction device, as well as the vehicle ride comfort prediction method and vehicle ride comfort prediction program implemented thereon, will be described more specifically.

[0020] FIG. 1 is a block diagram showing the configuration of a vehicle ride comfort prediction device according to an embodiment. FIG. 2 is a diagram for explaining a test driving route as an example. FIG. 2A shows a driving route where bumps are in phase, FIG. 2B shows a driving route where bumps are out of phase, and FIG. 2C is a diagram for explaining the shape of a bump. FIG. 3 is a diagram showing a biological signal as an example. FIG. 3A shows a pulse wave, where the horizontal axis is time (elapsed time) and the vertical axis is magnitude. FIG. 3B shows an electrocardiogram, where the horizontal axis is time (elapsed time) and the vertical axis is electrocardiogram potential. FIG. 3C shows an RR interval, where the horizontal axis is time (elapsed time) and the vertical axis is magnitude. FIG. 4 is a diagram showing three-axis acceleration data as an example. FIG. 5 is a diagram for explaining Russell's annular model. The horizontal axis of FIG. 5 is pleasant / unpleasant, and the vertical axis is awake / calm. FIG. 6 is a diagram showing the wavelet transform result of an electrocardiogram as an example. The horizontal axis of FIG. 6 is time (elapsed time), and the vertical axis is frequency. FIG. 7 is a diagram showing the wavelet transform results of the acceleration data of each axis as an example. FIG. 7A shows the wavelet transform result of the acceleration data of the X axis, where the horizontal axis is time (elapsed time) and the vertical axis is frequency. FIG. 7B shows the wavelet transform result of the acceleration data of the Y axis, where the horizontal axis is time (elapsed time) and the vertical axis is frequency. FIG. 7C shows the wavelet transform result of the acceleration data of the Z axis, where the horizontal axis is time (elapsed time) and the vertical axis is frequency. FIG. 8 is a diagram showing the cumulative contribution rate of principal component analysis and the first and second principal components as an example. FIG. 8A shows the cumulative contribution rate, where the horizontal axis is each principal component from the first principal component and the vertical axis is the cumulative contribution rate. FIG. 8B shows the first and second principal components with respect to the target vehicle speed, where the horizontal axis is the first principal component and the vertical axis is the second principal component. FIG. 8C shows the first and second principal components with respect to the pattern of the driving route, where the horizontal axis is the first principal component and the vertical axis is the second principal component. FIG. 9 is a diagram showing the eigenvectors of the first and second principal components as an example. FIG. 9A is the eigenvector of the first principal component in the X-axis acceleration data, where the horizontal axis is frequency and the vertical axis is magnitude.FIG. 9B is the eigenvector of the first principal component in the Y-axis acceleration data, where the horizontal axis is the frequency and the vertical axis is the magnitude. FIG. 9C is the eigenvector of the first principal component in the Z-axis acceleration data, where the horizontal axis is the frequency and the vertical axis is the magnitude. FIG. 9D is the eigenvector of the second principal component in the X-axis acceleration data, where the horizontal axis is the frequency and the vertical axis is the magnitude. FIG. 9E is the eigenvector of the second principal component in the Y-axis acceleration data, where the horizontal axis is the frequency and the vertical axis is the magnitude. FIG. 9F is the eigenvector of the second principal component in the Z-axis acceleration data, where the horizontal axis is the frequency and the vertical axis is the magnitude. FIG. 10 is a diagram showing the regression characteristics in the first machine learning model of random forest as an example. FIG. 10A shows the regression characteristics when the wavelet transform result of the electrocardiogram is used for the biological signal, FIG. 10B shows the regression characteristics when the RR interval is used for the biological signal, and FIG. 10C shows the regression characteristics when the pulse wave is used for the biological signal. The horizontal axes of FIGS. 10A to 10C are evaluation values, and the vertical axes thereof are predicted values. FIG. 11 is a diagram for explaining the correlation between the biological signal and the emotion vector as an example. FIG. 11A shows the regression characteristics when the wavelet transform result of the electrocardiogram is used for the biological signal, where the horizontal axis is the evaluation value and the vertical axis is the predicted value. FIG. 11B shows the relationship with the pleasant / unpleasant / excited / calm axis when the wavelet transform result of the electrocardiogram is used for the biological signal, where the horizontal axis is the pleasant / unpleasant / excited / calm axis and the vertical axis is the importance distribution. FIG. 11C shows the regression characteristics when the RR interval is used for the biological signal, where the horizontal axis is the evaluation value and the vertical axis is the predicted value. FIG. 11D shows the relationship with the pleasant / unpleasant / excited / calm axis when the RR interval is used for the biological signal, where the horizontal axis is the pleasant / unpleasant / excited / calm axis and the vertical axis is the importance distribution. FIG. 11E shows the regression characteristics when the pulse wave is used for the biological signal, where the horizontal axis is the evaluation value and the vertical axis is the predicted value. FIG. 11F shows the relationship with the pleasant / unpleasant / excited / calm axis when the pulse wave is used for the biological signal, where the horizontal axis is the pleasant / unpleasant / excited / calm axis and the vertical axis is the importance distribution.

[0021] In an embodiment, a vehicle ride comfort prediction device 1000 includes, for example, as shown in FIG. 1, a vibration acquisition unit 1, a control processing unit 2, an input unit 3, an output unit 4, an interface unit (IF unit) 5, and a storage unit 6.

[0022] The vibration acquisition unit 1 is a device that is connected to the control processing unit 2 by wire or wirelessly and acquires vibration data representing the vibration of an occupant according to the control of the control processing unit 2. The vibration acquisition unit 1 is, for example, a three-axis acceleration sensor attached to the chest of the occupant, measures the vibration of the occupant on the X-axis, Y-axis, and Z-axis that are orthogonal to each other, acquires the acceleration of the X-axis, the acceleration of the Y-axis, and the acceleration of the Z-axis, and outputs these to the control processing unit 2. For example, the X-axis is set in the vertical direction, the Y-axis is set in the left-right direction, and the Z-axis is set in the front-back direction.

[0023] Note that the vibration acquisition unit 1 may be configured to include, for example, an interface circuit that inputs and outputs data to and from an external device. The external device is a storage medium such as a USB (Universal Serial Bus) memory and an SD card (registered trademark) that stores three-axis acceleration data representing the vibration of an occupant in three-axis acceleration in time series. The three-axis acceleration data includes time-series X-axis acceleration data (X-axis acceleration data (X-axis vibration data)) generated by sampling the X-axis acceleration at each sampling timing, time-series Y-axis acceleration data (Y-axis acceleration data (Y-axis vibration data)) generated by sampling the Y-axis acceleration at each sampling timing, and time-series Z-axis acceleration data (Z-axis acceleration data (Z-axis vibration data)) generated by sampling the Z-axis acceleration at each sampling timing. Alternatively, the external device is a drive device that reads data from a recording medium such as a CD-ROM (Compact Disc Read Only Memory), CD-R (Compact Disc Recordable), DVD-ROM (Digital Versatile Disc Read Only Memory), and DVD-R (Digital Versatile Disc Recordable) that records the three-axis acceleration data. Alternatively, the vibration acquisition unit 1 is, for example, a communication interface circuit that transmits and receives communication signals to and from an external device, and the external device is connected to the communication interface circuit via a network (WAN (Wide Area Network, including a public communication network)) or LAN (Local Area Network), and is a server device that manages the three-axis acceleration data. Here, when the vibration acquisition unit 1 is an interface circuit or a communication interface circuit, the vibration acquisition unit 1 may be used in combination with the IF unit 5 (that is, the IF unit 5 may be used as the vibration acquisition unit 1).

[0024] The input unit 3 is connected to the control processing unit 2 and is a device that inputs various commands such as commands for instructing the start of ride comfort prediction and various data necessary for operating the vehicle ride comfort prediction device 1000, such as the name of the passenger, into the vehicle ride comfort prediction device 1000. For example, it is a keyboard, a mouse, and a plurality of input switches assigned with predetermined functions. The output unit 4 is connected to the control processing unit 2 and is a device that outputs commands, data, and ride comfort of the prediction result input from the input unit 3 according to the control of the control processing unit 2. For example, it is a display device such as a CRT display, an LCD (liquid crystal display device), and an organic EL display, and a printing device such as a printer.

[0025] Note that the input unit 3 and the output unit 4 may be composed of a touch panel. When configuring this touch panel, the input unit 3 is a position input device that detects and inputs an operation position, such as a resistive film method or a capacitance method, and the output unit 4 is a display device. In this touch panel, the position input device is provided on the display surface of the display device, and one or a plurality of input content candidates that can be input to the display device are displayed. When the user touches the display position where the input content to be input is displayed, the position is detected by the position input device, and the display content displayed at the detected position is input to the vehicle ride comfort prediction device 1000 as the user's operation input content. In such a touch panel, since the user can easily understand the input operation intuitively, a vehicle ride comfort prediction device 1000 that is easy for the user to handle is provided.

[0026] The IF unit 5 is connected to the control processing unit 2 and is a circuit that inputs and outputs data to and from external devices according to the control of the control processing unit 2. For example, it is an interface circuit of RS-232C using a serial communication method, an interface circuit using the Bluetooth (registered trademark) standard, and an interface circuit using the USB standard. Further, the IF unit 5 may be a communication interface circuit that transmits and receives communication signals to and from external devices, such as a data communication card and a communication interface circuit according to the IEEE802.11 standard.

[0027] The memory unit 6 is a circuit connected to the control processing unit 2 and stores various predetermined programs and various predetermined data according to the control of the control processing unit 2. The various predetermined programs include, for example, a control processing program, and the control processing program includes, for example, a control program, a biological signal prediction program, and a ride comfort prediction program, etc. The control program is a program for controlling each part 1, 3 to 6 of the vehicle ride comfort prediction device 1000 according to the functions of the respective parts. The biological signal prediction program is a program for predicting the biological signal of the occupant using a machine learning-trained first machine learning model based on the vibration data acquired by the vibration acquisition unit 1. The ride comfort prediction program is a program for predicting the ride comfort of the occupant using a machine learning-trained second machine learning model based on the biological signal of the occupant predicted by the biological signal prediction program. The various predetermined data include, for example, the vibration data acquired by the vibration acquisition unit 1 and data necessary for executing these programs, such as the predicted ride comfort of the result.

[0028] Such a memory unit 6 includes, for example, a ROM (Read Only Memory) which is a non-volatile memory element, an EEPROM (Electrically Erasable Programmable Read Only Memory) which is a rewritable non-volatile memory element, etc. And the memory unit 6 includes a RAM (Random Access Memory) etc. which serves as a working memory of the so-called control processing unit 2 for storing data etc. generated during the execution of the predetermined program. Further, the memory unit 6 may be configured to include a hard disk device with a relatively large storage capacity.

[0029] The control processing unit 2 controls each part 1, 3 to 6 of the vehicle ride comfort prediction device 1000 according to the functions of the respective parts, and is a circuit for predicting the ride comfort via a biological signal from the vibration data acquired by the vibration acquisition unit 1. The control processing unit 2 is configured to include, for example, a CPU (Central Processing Unit) and its peripheral circuits. When the control processing program is executed in the control processing unit 2, a control unit 21, a biological signal prediction unit 22, and a ride comfort prediction unit 23 are functionally configured.

[0030] The control unit 21 controls each part 1, 3 to 6 of the vehicle ride comfort prediction device 1000 according to the functions of the respective parts, and is in charge of the overall control of the vehicle ride comfort prediction device 1000.

[0031] The biological signal prediction unit 22 predicts the biological signal of the occupant using a trained first machine learning model based on the vibration data acquired by the vibration acquisition unit 1.

[0032] The ride comfort prediction unit 23 predicts the ride comfort of the occupant using a trained second machine learning model based on the biological signal of the occupant predicted by the biological signal prediction unit 22.

[0033] In the present embodiment, for example, the vibration data is three-axis acceleration data, and the biological signal is any one of an electrocardiogram, an RR interval which is an interval between R waves in the electrocardiogram, and a pulse wave.

[0034] The explanatory variables of the first machine learning model are the first and second principal components obtained by principal component analysis of the conversion results obtained by wavelet-transforming each acceleration data of each axis in the three-axis acceleration data, and the target variable is the biological signal. The explanatory variable of the second machine learning model is the biological signal, and the target variable is the ride comfort. This ride comfort is represented by a vector (emotion vector) representing the centroid point of each evaluation value of all items in Russell's ring model.

[0035] For machine learning models, various models used in machine learning such as neural networks, support vector machines (SVMs), and random forests can be utilized.

[0036] The generation of such first and second machine learning models will be described in more detail.

[0037] To generate the training datasets for creating these first and second machine learning models, a test was conducted in which a vehicle carrying a passenger (subject) was driven under various driving conditions. The subject was equipped with a sensor on the chest that measures three-axis acceleration and further measures an electrocardiogram and a pulse wave. For such a sensor, for example, Silmee Bar type Lite manufactured by TDK Corporation was used. Note that Silmee is a registered trademark. The three-axis acceleration data, electrocardiogram, and pulse wave are measured synchronously with each other.

[0038] As shown in Fig. 2, on the travel path, three bump members (ramp-over fixtures), which are prismatic members with a trapezoidal cross-section and extend in a direction orthogonal to the vehicle's traveling direction, are provided at a predetermined interval in the traveling direction. As shown in Table 1, four first to fourth patterns PN1 to PN4 were set for the arrangement pattern of the bumps. In the first and second patterns PN1 and PN2, the height of the bumps was set to 15 [mm], and in the third and fourth patterns PN3 and PN4, the height of the bumps was set to 25 [mm]. In the first and fourth patterns PN1 and PN4, the bumps were arranged so that both the left and right wheels could cross the bumps simultaneously (Parallel), and in the second and third patterns PN2 and PN3, the bumps were arranged so that the left and right wheels could cross the bumps alternately with a 180-degree phase difference. The subject sat in the rear right seat of the vehicle, and the tests were conducted with and without blindfolds. The vehicle's traveling speed (target vehicle speed) was 10 [km / h], 20 [km / h], 30 [km / h], and 40 [km / h] for each test. Therefore, for one subject, the tests were conducted with these combinations of the first to fourth patterns PN1 to PN4, with and without blindfolds, and traveling speeds of 10 [km / h], 20 [km / h], 30 [km / h], and 40 [km / h]. During the test, the subject's three-axis acceleration was sampled at a predetermined sampling interval, three-axis acceleration data was generated, and the pulse wave and electrocardiogram were measured. An example of the pulse wave is shown in Fig. 3A, and an example of the electrocardiogram is shown in Fig. 3B. Fig. 3C shows the RR interval, which is the interval between R waves in the electrocardiogram, generated from the measured electrocardiogram. The R wave was extracted by Christov's algorithm. Since the RR interval is discretized at the R wave positions, spline interpolation was performed to match the number of data points with other data. An example of the acceleration data for each axis in the three-axis acceleration data is shown in Fig. 4.

[0039]

Table 1

[0040] After driving, the ride comfort was subjectively evaluated through an interview between the subject and the tester. The Russell's Ring Model was used as the subjective evaluation index, and the index value was generated by scoring on a 5-point Likert scale with reference to the ride comfort in the vehicle stationary state. In the Russell's Ring Model, as shown in FIG. 5, a total of 28 items of emotions and feelings were adopted, and each item was mapped (plotted) in a coordinate system with the horizontal axis of pleasure / displeasure (Positive Valence / Negative Valence) and the vertical axis of arousal / calmness (High Arousal / Low Arousal), and the intensity and direction of the emotion were represented by vectors. Then, the centroid of each point (each point of each evaluation value) of all these items was obtained, and the vector representing this centroid was obtained as the emotion vector indicating the ride comfort of the occupant. Thus, the ride comfort was quantified.

[0041] The vibration data of the three-axis acceleration data generated in this way and any one of the biological signals of the pulse wave, electrocardiogram, and RR interval were used as the learning data (first learning data) of the first machine learning model, and a plurality of such first learning data were provided to form a learning data set (first learning data set) of the first machine learning model. The biological signal of any one of the pulse wave, electrocardiogram, and RR interval and the emotion vector were used as the learning data (second learning data) of the second machine learning model, and a plurality of such second learning data were provided to form a learning data set (second learning data set) of the second machine learning model. However, in this embodiment, preprocessing was performed on the electrocardiogram and the three-axis acceleration data.

[0042] More specifically, the electrocardiogram is filtered with a band-pass filter having a passband of 3 to 45 [Hz] for noise removal, and then wavelet-transformed using the mother wavelet Φ(t) of Equation 1 below. The wavelet transform is an analysis method for simultaneously obtaining time information and frequency information included in the analysis target. By stretching and shrinking the mother wavelet in the time axis direction, a plurality of different wavelets are generated, and the frequency spectrum of the analysis target is obtained for each of the plurality of wavelets. An example thereof is shown in FIG. 6. The value (magnitude) obtained by the wavelet transform at each frequency is obtained by obtaining the sum of the frequency spectra at the corresponding frequency, and in FIG. 6, it is represented by brightness, and the brighter the value, the larger it is.

[0043]

Number

[0044] Each acceleration data of each axis in the three-axis acceleration data is wavelet-transformed using the mother wavelet ψ(t) of Equation 2 below. An example thereof is shown in FIG. 7. Similar to FIG. 6, in FIG. 7, each value is represented by brightness, and the brighter the value, the larger it is.

[0045]

Number

[0046] The explanatory variables of the first machine learning model are dimensionally compressed and are the first and second principal components obtained by principal component analysis of the conversion results of wavelet-transforming each acceleration data of each axis in the three-axis acceleration data. As shown in FIG. 8A, the cumulative contribution rate is approximately 70% when the number of dimensions N = 1 and approximately 90% when the number of dimensions N = 3. Also, the first and second principal components at each target vehicle speed are shown in FIG. 8B, and the eigenvectors of the first and second principal components for each axis are shown in FIGS. 9A to 9F. In all cases, the acceleration below 10 [Hz] in the X direction (vertical direction) is dominant, and the lower the frequency, the higher the contribution tends to be. In the first principal component, it is high around approximately 3.8 [Hz], and in the second principal component, it tends to be high around approximately 1.3 [Hz] and approximately 3 [Hz]. This is considered to correspond to the "fluffy feeling" and the "coarse feeling" respectively, and it has been shown that these also greatly affect the riding comfort in machine learning. Separation is generally achieved in the reduced space, and it can be confirmed that the first and second principal components can be adopted as explanatory variables.

[0047] In this embodiment, the first and second principal components are used as the explanatory variables of the first machine learning model, but the first to third principal components may be used as the explanatory variables of the first machine learning model.

[0048] In this embodiment, when the biological signal is the wavelet transform result of an electrocardiogram, the first learning data is configured to include the first and second principal components in the three-axis acceleration data and the wavelet transform result of the electrocardiogram. The first learning data set is configured to include a plurality of pieces of this first learning data. Alternatively, when the biological signal is an RR interval, the first learning data is configured to include the first and second principal components in the three-axis acceleration data and the RR interval. The first learning data set is configured to include a plurality of pieces of this first learning data. Alternatively, when the biological signal is a pulse wave, the first learning data is configured to include the first and second principal components in the three-axis acceleration data and the pulse wave. The first learning data set is configured to include a plurality of pieces of this first learning data.

[0049] When the biological signal is the wavelet transform result of an electrocardiogram, the second training data is composed of the wavelet transform result of the electrocardiogram and an emotion vector. The second training data set is composed of a plurality of pieces of this second training data. Alternatively, when the biological signal is an RR interval, the second training data is composed of an RR interval and an emotion vector. The second training data set is composed of a plurality of pieces of this second training data. Alternatively, when the biological signal is a pulse wave, the second training data is composed of a pulse wave and an emotion vector. The second training data set is composed of a plurality of pieces of this second training data.

[0050] The same type of biological signal is used for the biological signal of the first training data and the biological signal of the second training data, and the same type of biological signal is used for the biological signal of the first machine learning model and the biological signal of the second machine learning model.

[0051] Then, by using such a first training data set, the first machine learning model is machine-learned, a trained first machine learning model is generated, and this trained first machine learning model is stored in the storage unit 6. Similarly, by using such a second training data set, the second machine learning model is machine-learned, a trained second machine learning model is generated, and this trained second machine learning model is stored in the storage unit 6.

[0052] Such a trained first machine learning model is generated by random forest, and its regression characteristics are shown in FIG. 10. When the biological signal is the wavelet transform result of an electrocardiogram, the coefficient of determination R 2 is 0.571. When the biological signal is an RR interval, the coefficient of determination R 2 is 0.997. When the biological signal is a pulse wave, the coefficient of determination R 2 was 0.996. The tendency is generally reproduced for any type of biological signal.

[0053] On the one hand, an example of the correlation between the biological signal and the emotion vector is shown in FIG. 11. In the correlation between the biological signal and the emotion vector, random forest regression was used. When the biological signal is the wavelet transform result of the electrocardiogram, the coefficient of determination R 2 is 0.691. When the biological signal is the RR interval, the coefficient of determination R 2 is 0.035. When the biological signal is the pulse wave, the coefficient of determination R 2 was 0.017. For the first and second machine learning models, it is preferable to use the electrocardiogram as the biological signal.

[0054] These control processing unit 2, input unit 3, output unit 4, IF unit 5 and storage unit 6 can be configured by, for example, a computer such as a desktop type or a notebook type. As described above, when the vibration acquisition unit 1 is an interface circuit or a communication interface circuit, since the vibration acquisition unit 1 can be used in place of the IF unit 5, the vibration acquisition unit 1 can also be configured by a computer.

[0055] Next, the operation of this embodiment will be described. FIG. 12 is a flowchart showing the operation of the vehicle ride comfort prediction device.

[0056] When the power of the vehicle ride comfort prediction device 1000 having such a configuration is turned on, it initializes each necessary part and starts its operation. In the control processing unit 2, a control unit 21, a biological signal prediction unit 22, and a ride comfort prediction unit 23 are functionally configured by executing the control processing program.

[0057] In FIG. 12, when the sampling timing arrives, the vehicle ride comfort prediction device 1000 acquires the three-axis acceleration of the occupant at the current sampling timing from the vibration acquisition unit 1 by the control unit 21 of the control processing unit 2 and stores it in the storage unit 6 (S1).

[0058] Subsequently, the vehicle ride comfort prediction device 1000 predicts the occupant's biological signal based on the three-axis acceleration data acquired by the vibration acquisition unit 1 using the pre-trained first machine learning model in the biological signal prediction unit 22 of the control processing unit 2, and stores the prediction result in the storage unit 6 (S2). More specifically, the biological signal prediction unit 22 generates three-axis acceleration data from the three-axis acceleration acquired at the current sampling timing and the three-axis acceleration acquired at the sampling timing a predetermined time in the past, performs wavelet transform on the acceleration data of each axis in the generated three-axis acceleration data, inputs the first and second principal components into the pre-trained first machine learning model, and predicts the biological signal.

[0059] Subsequently, the vehicle ride comfort prediction device 1000 predicts the occupant's ride comfort based on the biological signal of the occupant predicted by the biological signal prediction unit 22 in the process S2 using the pre-trained second machine learning model in the ride comfort prediction unit 23 of the control processing unit 2, and stores the prediction result in the storage unit 6 (S3).

[0060] Subsequently, the vehicle ride comfort prediction device 1000 outputs the occupant's ride comfort predicted by the ride comfort prediction unit 23 in the process S3 to the output unit 4 by the control unit 21 of the control processing unit 2 (S4). Note that the control unit 21 may output the occupant's ride comfort to an external device via the IF unit 5 as necessary.

[0061] Then, the vehicle ride comfort prediction device 1000 determines whether the prediction is completed by the control unit 21 of the control processing unit 2 (S5). As a result of this determination, for example, when the input of a command instructing termination is received at the input unit 3, or when the power switch (not shown) is turned off, etc., if it is determined that the process is to be terminated (Yes), the vehicle ride comfort prediction device 1000 ends this process. On the other hand, as a result of the determination, if it is determined that the process is not to be terminated (No), the vehicle ride comfort prediction device 1000 returns the process to process S1.

[0062] By operating in this way, the vehicle ride comfort prediction device 1000 predicts the occupant's ride comfort at a predetermined sampling interval.

[0063] As described above, since the vehicle ride comfort prediction device 1000 in the embodiment, the vehicle ride comfort prediction method, and the vehicle ride comfort prediction program implemented thereon are considered to affect the ride comfort on the biological signal, the ride comfort is predicted via the biological signal from the vibration data (the three-axis acceleration data in the above example), so that the ride comfort of the occupant can be predicted more appropriately.

[0064] Since the above-described vehicle ride comfort prediction device 1000, vehicle ride comfort prediction method, and vehicle ride comfort prediction program dimensionally compress the explanatory variables into the first and second principal components by principal component analysis, the first machine learning model can be simplified.

[0065] According to the embodiment, a vehicle ride comfort prediction device 1000, a vehicle ride comfort prediction method, and a vehicle ride comfort prediction program that use three-axis acceleration data for vibration data and any one of an electrocardiogram, an RR interval, and a pulse wave for the biological signal can be provided, and a vehicle ride comfort prediction device 1000, a vehicle ride comfort prediction method, and a vehicle ride comfort prediction program that represent the ride comfort by an emotion vector representing the centroid point in each evaluation value of all items in Russell's annular model can be provided.

[0066] The above-described vehicle ride comfort prediction device 1000, vehicle ride comfort prediction method, and vehicle ride comfort prediction program can quantify the ride comfort by an emotion vector.

[0067] In order to express the present invention, the present invention has been appropriately and sufficiently described through the embodiments with reference to the drawings above. However, it should be recognized that those skilled in the art can easily make changes and / or improvements to the above-described embodiments. Therefore, as long as the modified or improved forms implemented by those skilled in the art do not depart from the scope of the claims described in the claims, the modified or improved forms are construed to be included in the scope of the claims of the claims.

Explanation of Reference Numerals

[0068] 1000 Vehicle ride comfort prediction device 1 Vibration acquisition unit 2 Control processing unit 3 Input unit 4 Output unit 5 Interface unit (IF unit) 6 Memory unit 21 Control unit 22 Biosignal prediction unit 23 Riding comfort prediction unit

Claims

1. A vehicle ride comfort prediction device for predicting the ride comfort of a vehicle occupant, comprising: a vibration acquisition unit that acquires vibration data representing the vibration of the occupant; a biological signal prediction unit that predicts the biological signal of the occupant using a pre-trained first machine learning model based on the vibration data acquired by the vibration acquisition unit; a ride comfort prediction unit that predicts the ride comfort of the occupant using a pre-trained second machine learning model based on the biological signal of the occupant predicted by the biological signal prediction unit. A vehicle ride comfort prediction device.

2. The vibration data is three-axis acceleration data, and the biological signal is any one of an electrocardiogram, an RR interval which is an interval between R waves in the electrocardiogram, and a pulse wave. The vehicle ride comfort prediction device according to Claim 1.

3. The explanatory variables of the first machine learning model are the first and second principal components obtained by principal component analysis of the conversion results obtained by wavelet-transforming each acceleration data of each axis in the three-axis acceleration data. The vehicle ride comfort prediction device according to Claim 2.

4. The ride comfort is represented by an emotion vector representing the center of gravity point in each evaluation value of all items in Russell's annular model. The vehicle ride comfort prediction device according to Claim 1.

5. A vehicle ride comfort prediction method for predicting the ride comfort of a vehicle occupant, comprising: a vibration acquisition step of acquiring vibration data representing the vibration of the occupant; a biological signal prediction step of predicting the biological signal of the occupant using a pre-trained first machine learning model based on the vibration data acquired in the vibration acquisition step; a ride comfort prediction step of predicting the ride comfort of the occupant using a pre-trained second machine learning model based on the biological signal of the occupant predicted in the biological signal prediction step. A vehicle ride comfort prediction method.

6. A vehicle ride comfort prediction program for predicting the ride comfort of a vehicle occupant, which causes a computer to function as a vibration acquisition unit that acquires vibration data representing the vibration of the occupant, a biological signal prediction unit that predicts the biological signal of the occupant using a pre-trained first machine learning model based on the vibration data acquired by the vibration acquisition unit, and a ride comfort prediction unit that predicts the ride comfort of the occupant using a pre-trained second machine learning model based on the biological signal of the occupant predicted by the biological signal prediction unit. A vehicle ride comfort prediction program for the above purpose.

Images