Information processing apparatus, information processing program, and information processing method

Abstract

To estimate the longitudinal force of a tire by using only a feature amount contributing to the estimation of the longitudinal force of the tire, and to reduce the replacement frequency of a battery.SOLUTION: The information processing device 10 acquires an acceleration signal from the acceleration sensor 4 attached to the tire 2, extracts a time domain feature quantity and a frequency domain feature quantity from the acquired acceleration signal, and estimates the longitudinal force of the tire 2 using the time domain feature quantity and the frequency domain feature quantity.SELECTED DRAWING: Figure 9

Description

[Technical Field]

[0001] The present disclosure relates to an information processing device, an information processing program, and an information processing method. [Background technology]

[0002] Patent Document 1 discloses an evaluation method for measuring the longitudinal forces acting on a tire and evaluating the running performance of a tire on snowy and icy road surfaces, which are road surfaces accompanied by phenomena caused by a drop in temperature, such as snowy and icy roads. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2015-137999 Summary of the Invention [Problem to be solved by the invention]

[0004] Various features obtained from tire acceleration signals are used to estimate tire longitudinal forces, but no research has been conducted to date to determine which of these features contributes to estimating tire longitudinal forces.

[0005] Therefore, various feature quantities obtained from tire acceleration signals are used as they are for estimating tire longitudinal forces without considering whether they contribute to estimating tire longitudinal forces.

[0006] Furthermore, if the number of feature values ​​used to estimate the longitudinal tire force is large, the amount of calculation required to calculate the feature values ​​increases as the number of feature values ​​increases, which in turn increases the power consumption of the battery that supplies power to the information processing device that calculates the feature values ​​and transmits the calculation results.Increased battery power consumption also increases the frequency of battery replacement, and for example, if the information processing device is attached to the wheel together with the battery, replacing the battery attached to the wheel requires the tire to be removed from the wheel, which is time-consuming.

[0007] The present disclosure aims to provide an information processing device, an information processing program, and an information processing method that estimate the longitudinal force of a tire using only those feature quantities that contribute to estimating the longitudinal force of a tire from among various feature quantities obtained from an acceleration signal, and that can reduce the frequency of battery replacement compared to estimating the longitudinal force of a tire from an acceleration signal without narrowing down the feature quantities that contribute to estimating the longitudinal force of a tire. [Means for solving the problem]

[0008] An information processing device according to a first aspect includes an acquisition unit that acquires acceleration signals from an acceleration sensor attached to an inner surface of a tire; an extraction unit that extracts, from the acceleration signals acquired by the acquisition unit, time domain features that are features related to the time domain of the acceleration signals and frequency domain features that are features related to the frequency domain of the acceleration signals; and an estimation unit that estimates longitudinal forces generated along a tangential direction at a contact patch of the tire with a road surface, using the time domain features and the frequency domain features extracted by the extraction unit.

[0009] An information processing device according to a second aspect is the information processing device according to the first aspect, wherein the extraction unit extracts the time domain feature and the frequency domain feature from the acceleration signal during a measurement period of the acceleration signal, the period being limited to a predetermined period including a point in time when the tread of the tire, which is in contact with the road surface, separates from the road surface.

[0010] An information processing device according to a third aspect is the information processing device according to the first or second aspect, wherein the acceleration sensor is a sensor that measures the acceleration signals along at least three directions, namely, a tangential direction of the tire, a width direction of the tire, and a direction of gravity, and the extraction unit extracts the time domain feature and the frequency domain feature from the acceleration signals along at least one direction of the tangential direction of the tire, the width direction of the tire, and the direction of gravity.

[0011] An information processing device according to a fourth aspect is the information processing device according to the third aspect, wherein the extraction unit extracts the time domain feature and the frequency domain feature from the acceleration signals along three directions: a tangential direction of the tire, a width direction of the tire, and a gravity direction.

[0012] An information processing device according to a fifth aspect is the information processing device according to any one of the first to fourth aspects, wherein the extraction unit extracts at least one of a standard deviation of the acceleration signal, a feature based on autocorrelation of the acceleration signal, and a feature based on non-stationarity of the acceleration signal from the acceleration signal as the time-domain feature.

[0013] An information processing device according to a sixth aspect is the information processing device according to the fifth aspect, wherein the extraction unit extracts, from the acceleration signal, a standard deviation of the acceleration signal, a feature based on autocorrelation of the acceleration signal, and a feature based on non-stationarity of the acceleration signal as the time-domain feature.

[0014] An information processing device according to a seventh aspect is an information processing device according to the fifth or sixth aspect, wherein the extraction unit extracts a time lag at which the value of the autocorrelation of the acceleration signal decreases to 1 / e compared to the autocorrelation of the acceleration signal at a time lag of 0 as a feature based on the autocorrelation of the acceleration signal.

[0015] An information processing device according to an eighth aspect is the information processing device according to the fifth or sixth aspect, wherein the extraction unit extracts the time lag at which the autocorrelation of the acceleration signal first becomes zero as a feature based on the autocorrelation of the acceleration signal.

[0016] An information processing device according to a ninth aspect is an information processing device according to any one of the fifth to eighth aspects, wherein the extraction unit extracts a value representing the degree to which the acceleration signal follows a unit root process as a feature based on the non-stationarity of the acceleration signal.

[0017] An information processing device according to a tenth aspect is the information processing device according to the ninth aspect, wherein the extraction unit sets a value representing the degree to which the acceleration signal follows a unit root process to be a ratio of the time lag from the point at which the time lag is 0 in the autocorrelation of the difference series of the acceleration signal until the autocorrelation of the difference series first becomes 0 to the time lag from the point at which the time lag is 0 in the autocorrelation of the acceleration signal until the autocorrelation of the acceleration signal first becomes 0.

[0018] An information processing device according to an eleventh aspect is the information processing device according to any one of the fifth to tenth aspects, wherein the extraction unit extracts a center frequency of a frequency distribution included in the acceleration signal as the frequency domain feature.

[0019] An information processing program according to a twelfth aspect is a program for causing a computer to execute a process of acquiring an acceleration signal from an acceleration sensor attached to an inner surface of a tire, extracting from the acquired acceleration signal time domain features that are features related to a time domain of the acceleration signal and frequency domain features that are features related to a frequency domain of the acceleration signal, and estimating a longitudinal force generated in a tangential direction at a contact patch of the tire with a road surface in association with rotation control of the tire, using the extracted time domain features and frequency domain features.

[0020] A non-transitory storage medium according to a thirteenth aspect is a non-transitory storage medium storing a program executable by a computer to execute processing for estimating longitudinal forces of a tire, the processing for estimating longitudinal forces including the steps of: acquiring acceleration signals from an acceleration sensor attached to an inner surface of the tire; extracting, from the acquired acceleration signals, time domain features that are features related to the time domain of the acceleration signals and frequency domain features that are features related to the frequency domain of the acceleration signals; and estimating longitudinal forces generated in a tangential direction at the contact patch of the tire with the road surface in association with rotation control of the tire, using the extracted time domain features and frequency domain features.

[0021] A computer program product according to a fourteenth aspect includes a program that causes a computer to execute a process of acquiring an acceleration signal from an acceleration sensor attached to an inner surface of a tire, extracting from the acquired acceleration signal time domain features that are features related to a time domain of the acceleration signal and frequency domain features that are features related to a frequency domain of the acceleration signal, and estimating a longitudinal force generated in a tangential direction at a contact patch of the tire with a road surface in association with rotation control of the tire, using the extracted time domain features and frequency domain features.

[0022] An information processing method according to a fifteenth aspect is a method in which a computer executes a process of acquiring an acceleration signal from an acceleration sensor attached to an inner surface of a tire, extracting from the acquired acceleration signal time domain features that are features related to the time domain of the acceleration signal and frequency domain features that are features related to the frequency domain of the acceleration signal, and estimating a longitudinal force generated in a tangential direction at a contact patch of the tire with a road surface in association with rotation control of the tire, using the extracted time domain features and frequency domain features. [Effects of the Invention]

[0023] According to the present disclosure, the longitudinal force of a tire is estimated using only those feature quantities that contribute to estimating the longitudinal force of a tire, out of the various feature quantities obtained from an acceleration signal, and this has the effect of reducing the frequency of battery replacement compared to estimating the longitudinal force of a tire from an acceleration signal without narrowing down the feature quantities that contribute to estimating the longitudinal force of a tire. [Brief explanation of the drawings]

[0024] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of an experimental device. [Figure 2] FIG. 10 is a diagram showing an example of an image captured of the movement of the tread. [Figure 3] FIG. 4 is a diagram illustrating an example of an acceleration signal. [Figure 4] FIG. 10 is a scatter diagram showing an example of the relationship between the longitudinal force of a tire and the standard deviation of an acceleration signal. [Figure 5] FIG. 10 is a scatter diagram showing an example of the relationship between the longitudinal force of a tire and the autocorrelation feature amount of an acceleration signal. [Figure 6] FIG. 1 is a diagram showing an example of a correlograph. [Figure 7] FIG. 10 is a scatter diagram showing an example of the relationship between the longitudinal force of a tire and the unit root process feature amount of an acceleration signal. [Figure 8] FIG. 4 is a scatter diagram showing an example of the relationship between the longitudinal force of a tire and the center frequency of an acceleration signal. [Figure 9] FIG. 2 is a diagram illustrating an example of a functional configuration of an information processing device. [Figure 10] FIG. 2 is a diagram illustrating an example of the configuration of a main part of an electrical system of an information processing device. [Figure 11] 10 is a flowchart showing an example of the flow of a longitudinal force estimation process. DETAILED DESCRIPTION OF THE INVENTION

[0025] Hereinafter, the present embodiment will be described with reference to the drawings. The same components and processes are denoted by the same reference numerals throughout the drawings, and duplicated explanations will be omitted. The dimensional proportions in the drawings are exaggerated for the sake of explanation, and may differ from the actual proportions.

[0026] The force applied in the tangential direction to the contact surface of the tire 2 (see Figure 1, for example) that is in contact with the road surface is called the "longitudinal force." The longitudinal force is an example of a parameter that represents the behavior of the tire 2 while in motion, and is used to analyze the driving state of the vehicle.

[0027] If the longitudinal force of tire 2 could be estimated in real time, the running state of a vehicle on snowy roads, for example, could be quantitatively obtained, and driving control could be performed to ensure stable running even on snowy roads. However, until now, it had not been clear what feature quantities obtained from the behavior of tire 2 should be focused on in order to estimate the longitudinal force of tire 2. Therefore, the inventors decided to observe the dynamics of tire 2 in detail through experiments in order to clarify the feature quantities that contribute to estimating the longitudinal force of tire 2.

[0028] <Knowledge regarding features that contribute to estimation of longitudinal forces> 1 is a diagram showing an example of the configuration of an experimental device 1. A tire 2 is attached to a rotating shaft 3, and is arranged so that a tread 2A of the tire 2 comes into contact with a belt 5 that simulates a road surface.

[0029] The rotating shaft 3 and the belt 5 are each connected to a drive unit (not shown), which controls the stopping and driving of the rotating shaft 3 and the belt 5. The moving speeds of the rotating shaft 3 and the belt 5 are variable. Furthermore, the drive unit can also drive the rotating shaft 3 and the belt 5 in the direction opposite to the direction in which they were previously driven.

[0030] The movement of the tread 2A of the tire 2 rotating on the belt 5 in this manner is photographed by a high-speed camera 7. To make it easier for the high-speed camera 7 to photograph the movement of the tread 2A, an experimental tire 2 is attached to the rotating shaft 3, with grooves formed in half of the tread 2A along the width direction of the tire 2 around the circumference of the tire 2 and half of the tread 2A removed. The high-speed camera 7 photographs the tread 2A from the side of the tire 2 from which the tread 2A has been removed. To make the movement of the tread 2A easier to see, white spots 6 are sprayed onto the side of the tread 2A. Images 15 (see Figure 2) taken by the high-speed camera 7 are stored in a computer 9.

[0031] Meanwhile, an acceleration sensor 4 is attached to the inner surface of the tire 2 located behind the tread 2A. The acceleration sensor 4 measures acceleration in at least three axial directions. The three axial directions in this disclosure refer to the tangential direction at the contact patch of the tire 2, the tire width direction, and the direction of gravity. The tangential direction at the contact patch of the tire 2 is also the direction along which the tire 2 rolls as the tire 2 rotates. Hereinafter, the tangential direction at the contact patch of the tire 2 will simply be referred to as the "tangential direction of the tire 2."

[0032] The tangential direction, width direction, and gravity direction of the tire 2 are perpendicular to each other. In Figure 1, the X axis represents the tangential direction of the tire 2, the Y axis represents the width direction of the tire 2, and the Z axis represents the gravity direction. The acceleration signal measured by the acceleration sensor 4 is transmitted to the data collection device 8.

[0033] The data collection device 8 records the acceleration signal measured by the acceleration sensor 4, and when the tire 2 and belt 5 are driven by a user's instruction, outputs a pulse signal indicating the start of driving to the high-speed camera 7. Upon receiving the pulse signal, the high-speed camera 7 starts capturing images. This synchronizes the acceleration signal with the image 15 capturing the movement of the tread 2A.

[0034] FIG. 2 is a diagram showing an example of an image capturing the movement of the tread 2A. In FIG. 2, arrow R1 indicates the position of the front edge of the tread 2A where it begins to come into contact with the road surface (the belt 5 in the example of FIG. 2) as the tire 2 rotates. Arrow R2 indicates the position of the rear edge of the tread 2A where it begins to separate from the road surface as the tire 2 rotates. Arrow R3 indicates the direction of rotation of the tire. The direction of movement of the belt 5 is the same as the direction of rotation of the tire.

[0035] In FIG. 2, image 15-1 shows the situation when the tire 2 is rotated so that the tread 2A moves at the same speed as the belt 5. In this case, the relative speed between the tread 2A and the belt 5 is 0, so the tread 2A moves without tilting relative to the belt 5. Hereinafter, the state in which this type of movement of the tread 2A occurs will be referred to as the "stationary state." In the stationary state, the longitudinal force of the tire 2 is 0.

[0036] Image 15-2 shows the situation when the tire 2 is rotated so that the tread 2A moves at a speed faster than the moving speed of the belt 5. In this case, the tire 2 attempts to move upstream in the moving direction of the belt 5, so the tread 2A moves while tilting relative to the belt 5 in the opposite direction to the rotating direction of the tire 2. In other words, a force that increases the moving speed of the tire 2 acts on the contact surface of the tire 2 with the belt 5. Hereinafter, the state in which this movement of the tread 2A occurs will be referred to as the "running state." The longitudinal forces of the tire 2 in the running state are represented by positive values.

[0037] Image 15-3 shows the situation when the tire 2 is rotated so that the tread 2A moves at a slower speed than the belt 5. In this case, the tread 2A is dragged by the movement of the belt 5, and moves while tilting relative to the belt 5 in the same direction as the rotation of the tire 2. In other words, a force that reduces the movement speed of the tire 2 acts on the contact surface of the tire 2 with the belt 5. Hereinafter, the state in which this movement of the tread 2A occurs is referred to as the "braking state." The longitudinal force of the tire 2 in the braking state is expressed as a negative value.

[0038] It can be seen from FIG. 2 that the closer the tread 2A is to the rear edge, the greater the inclination of the tread 2A becomes during running and braking.

[0039] Fig. 3 is a diagram showing an example of acceleration signals in the X-axis direction, Y-axis direction, and Z-axis direction measured by the acceleration sensor 4 during one rotation of the tire 2. The horizontal axis of Fig. 3 represents time, and the vertical axis represents the magnitude of the acceleration signal. Graph 16A represents the acceleration signal in the X-axis direction, graph 16B represents the acceleration signal in the Y-axis direction, and graph 16C represents the acceleration signal in the Z-axis direction.

[0040] As can be seen from FIG. 3, the magnitude of the acceleration signal in each axial direction was greatest when the acceleration sensor 4 reached the rear edge.

[0041] On the other hand, the movement of the tread 2A shown in Figure 2 indicates that the closer the tread 2A is to the rear edge, the greater the inclination of the tread 2A. The fact that the inclination of the tread 2A is maximum at the rear edge position means that when the tread 2A separates from the belt 5 at the rear edge, the force with which the tread 2A kicks up the belt 5 due to the restoring force of the tread 2A is also maximum. In other words, it is thought that the magnitude of the acceleration signal is maximum at the rear edge due to the vibrations generated when the tread 2A kicks up the belt 5.

[0042] The greater the magnitude of the acceleration signal, the greater the influence it has on the longitudinal force of tire 2. Therefore, when estimating the longitudinal force of tire 2, it has been discovered that it is preferable to focus on the acceleration signal at the rear edge, where the magnitude of the acceleration signal is greater than in other regions.

[0043] That is, it is preferable to use, within the measurement period of the acceleration signal of the tire 2 by the acceleration sensor 4, the acceleration signal for a predetermined period that includes the point in time when the acceleration sensor 4 reaches the rear edge to estimate the longitudinal force of the tire 2. Naturally, the predetermined period is limited to a period that includes the point in time when the acceleration sensor 4 reaches the rear edge and is shorter than the period for one rotation of the tire 2. The acceleration signal for the predetermined period that is used to estimate the longitudinal force of the tire 2 is referred to as the "rear edge acceleration signal."

[0044] Next, the feature quantities that contribute to the estimation of the longitudinal force of the tire 2 will be considered.

[0045] There are various types of feature quantities that can be obtained from acceleration signals. Among these, even if the longitudinal force of the tire 2 is estimated using a feature quantity that contributes less to the estimation of the longitudinal force of the tire 2 compared to other feature quantities, the amount of calculation simply increases, and the effect on the estimation accuracy of the longitudinal force of the tire 2 is small.

[0046] Therefore, the inventor analyzed the relationship between the longitudinal force of the tire 2 and various feature quantities while changing the rotational speed of the tire 2, and studied feature quantities related to the longitudinal force of the tire 2. As a result, the inventor discovered that the longitudinal force of the tire 2 can be estimated by using time domain feature quantities, which are feature quantities related to the time domain of the acceleration signal, and frequency domain feature quantities, which are feature quantities related to the frequency domain of the acceleration signal. Specifically, it was found that it is preferable to use at least one of the standard deviation of the acceleration signal, feature quantities based on the autocorrelation of the acceleration signal, and feature quantities based on the non-stationarity of the acceleration signal as the time domain feature. It was also found that it is preferable to use the center frequency of the frequency distribution included in the acceleration signal as the frequency domain feature.

[0047] FIG. 4 is a scatter diagram showing an example of the relationship between the longitudinal force of tire 2 and the standard deviation of the acceleration signal according to the moving speed of tire 2. The horizontal axis of the scatter diagram in FIG. 4 represents the longitudinal force of tire 2, and the vertical axis represents the standard deviation of the acceleration signal. In FIG. 4, with "0" on the horizontal axis as the reference point, the further to the right the positive longitudinal force, i.e., the driving force, increases, and the further to the left the negative longitudinal force, i.e., the braking force, increases. Furthermore, points SP1, SP2, and SP3 represent plot points for each moving speed of tire 2, and there is a magnitude relationship such that the moving speed of tire 2 corresponding to point SP1<the moving speed of tire 2 corresponding to point SP2<the moving speed of tire 2 corresponding to point SP3.

[0048] 4, it can be seen that the standard deviation of the acceleration signal tends to increase as the magnitude of the longitudinal force of the tire 2 increases in the positive direction. This is thought to be because the vibration caused by the kick-up of the belt 5 by the tread 2A occurring at the rear edge increases the standard deviation of the acceleration signal. It can also be seen that, for the same longitudinal force of the tire 2, the standard deviation of the acceleration signal tends to increase as the moving speed of the tire 2 increases.

[0049] 5 is a scatter plot showing an example of the relationship between the longitudinal force of the tire 2 and the autocorrelation feature of the acceleration signal according to the moving speed of the tire 2. The autocorrelation feature is a feature obtained based on a correlograph that shows the relationship between the autocorrelation of the acceleration signal and the time lag, and is an example of a feature based on the autocorrelation of the acceleration signal according to the technology of the present disclosure.

[0050] Fig. 6 is a diagram showing an example of a correlograph. The horizontal axis of the correlograph in Fig. 6 indicates the time lag, which represents the time shifted from the original data, and the vertical axis represents the autocorrelation. The time lag in Fig. 6 represents the time shifted from the original data using the number of sampling points of the acceleration signal.

[0051] From the scatter diagram in Fig. 5, it can be seen that the autocorrelation feature value is higher in the braking state than in the driving state, and tends to increase as the longitudinal force of the tire 2 increases in the negative direction. In other words, the change in the autocorrelation feature value represents the change trend of the longitudinal force of the tire 2. It can also be seen that, if the longitudinal force of the tire 2 remains the same, the autocorrelation feature value tends to decrease as the moving speed of the tire 2 increases.

[0052] As an autocorrelation feature having such characteristics, for example, the time lag at which the value of the autocorrelation of an acceleration signal decreases to 1 / e is used, where "e" is the base of the natural logarithm, i.e., Napier's constant.

[0053] The definition of the autocorrelation feature is not limited to the above example. For example, the time lag at which the autocorrelation of the acceleration signal first becomes 0 may be used as the autocorrelation feature.

[0054] 7 is a scatter plot showing an example of the relationship between the longitudinal force of the tire 2 and the unit root process feature of the acceleration signal according to the moving speed of the tire 2. The unit root process feature is a feature obtained based on the non-stationarity of the acceleration signal, in which the original series is non-stationary but the difference series is stationary, and is an example of a feature based on the non-stationarity of the acceleration signal according to the technology of the present disclosure.

[0055] More specifically, the unit root process feature represents the degree to which the acceleration signal follows a unit root process. The degree to which the acceleration signal follows a unit root process is expressed, for example, by the ratio of the time lag from the time lag of 0 to the time when the autocorrelation of the difference series of the acceleration signal first becomes 0, to the time lag from the time lag of 0 to the time when the autocorrelation of the acceleration signal first becomes 0.

[0056] From the scatter plot in FIG. 7, it can be seen that the unit root process feature value tends to increase as the longitudinal force of tire 2 increases in the positive direction. In other words, the change in the unit root process feature value represents the change trend of the longitudinal force of tire 2. It can also be seen that, if the longitudinal force of tire 2 remains the same, the unit root process feature value tends to increase as the moving speed of tire 2 increases.

[0057] 8 is a scatter diagram showing an example of the relationship between the longitudinal force of the tire 2 and the center frequency of the acceleration signal according to the moving speed of the tire 2. The center frequency of the acceleration signal is a frequency at which, in the frequency distribution included in the acceleration signal, the integral value of the power spectrum of the frequency components lower than that frequency is equal to the integral value of the power spectrum of the frequency components higher than that frequency.

[0058] From the scatter diagram in Figure 8, it can be seen that the center frequency of the acceleration signal tends to increase as the longitudinal force of the tire 2 increases in the positive direction. This is thought to be because vibrations caused by the kick-up of the belt 5 by the tread 2A at the rear edge increase the center frequency of the acceleration signal. In other words, changes in the center frequency represent the tendency for changes in the longitudinal force of the tire 2. It can also be seen that, if the longitudinal force of the tire 2 remains the same, the center frequency tends to increase as the moving speed of the tire 2 increases.

[0059] From the above considerations, it has been found that the longitudinal force of the tire 2 can be estimated by using the standard deviation of the acceleration signal, the autocorrelation feature of the acceleration signal, the unit root process feature of the acceleration signal, and the center frequency of the acceleration signal.

[0060] <Estimation of longitudinal forces> Next, we will explain the functional configuration of the information processing device 10 that uses feature quantities obtained from the acceleration signal to estimate the longitudinal force of the tire 2. The information processing device 10, together with a battery that drives the information processing device 10, is attached to the inner surface of the tire 2, for example, located behind the tread 2A.

[0061] 9 is a diagram illustrating an example of the functional configuration of the information processing device 10. As illustrated in FIG. 9, the information processing device 10 includes functional units, that is, an acquisition unit 11, an extraction unit 12, and an estimation unit 13, and an acceleration sensor 4.

[0062] The acquisition unit 11 acquires an acceleration signal from the acceleration sensor 4.

[0063] The extracting unit 12 extracts time domain features and frequency domain features from the acceleration signal of the rear edge acquired by the acquiring unit 11 .

[0064] The estimation unit 13 estimates the longitudinal force of the tire 2 using the time domain feature amount and the frequency domain feature amount extracted by the extraction unit 12 .

[0065] While FIG. 9 shows an example of an information processing device 10 having an acceleration sensor 4 built therein, the information processing device 10 does not necessarily have to have an acceleration sensor 4 built therein. In this case, an individual acceleration sensor 4 is attached to the inner surface of the tire 2 located behind the tread 2A. Because the individual acceleration sensor 4 measures the acceleration signal at the position of the tread 2A, an information processing device 10 that does not have an acceleration sensor 4 built therein may be attached anywhere on the inner surface of the tire 2. In the present disclosure, an information processing device 10 having an acceleration sensor 4 built therein is used as an example.

[0066] The acceleration sensor 4 and the acquisition unit 11 of the information processing device 10 are connected wirelessly or by wire.

[0067] The information processing device 10 shown in Fig. 9 is configured using, for example, a computer 20. Fig. 10 is a diagram showing an example of the configuration of the main parts of the electrical system of the information processing device 10 configured using the computer 20. The computer 20 is attached together with a battery 22 to, for example, the rim of a wheel on which a tire is attached.

[0068] The computer 20 includes a CPU (Central Processing Unit) 20A, which is an example of a processor, a RAM (Random Access Memory) 20B used as a temporary work area for the CPU 20A, a nonvolatile memory 20C, and an input / output interface (I / O) 20D. The CPU 20A, RAM 20B, nonvolatile memory 20C, and I / O 20D are all connected to each other via a bus 20E.

[0069] The CPU 20A executes the processing of each functional unit shown in FIG.

[0070] The nonvolatile memory 20C is an example of a storage device that maintains stored information even when power supplied to the nonvolatile memory 20C is cut off, and is, for example, a semiconductor memory (Solid State Drive: SSD). Information that must not be lost every time the power supply to the information processing device 10 is cut off, such as a program, is stored in the nonvolatile memory 20C.

[0071] The nonvolatile memory 20C does not necessarily have to be built into the computer 20, but may be, for example, a portable storage device that can be attached to and detached from the computer 20.

[0072] To the I / O 20D, for example, the acceleration sensor 4 and the communication unit 21 are connected.

[0073] The communication unit 21 has a communication protocol for transmitting and receiving data to and from an external device such as an on-board computer (Electronic Control Unit: ECU) installed in a vehicle. The communication unit 21 is connected to the external device wirelessly, but may be connected by wire depending on the situation. The communication unit 21 may also be built into the computer 20.

[0074] The battery 22 supplies power to the computer 20, the acceleration sensor 4, and the communication unit 21. When the power supplied from the battery 22 decreases with use, the battery 22 is replaced. Note that the battery 22 may not directly supply power to the acceleration sensor 4 and the communication unit 21, and the computer 20, which receives power from the battery 22, may supply power to the acceleration sensor 4 and the communication unit 21.

[0075] <Function of information processing device> Next, the operation of the information processing device 10 will be described. Fig. 11 is a flowchart showing an example of the flow of a longitudinal force estimation process that is executed when an estimation start instruction is received from a user, for example, via the communication unit 21. The longitudinal force estimation process is a process for estimating the longitudinal force of tires 2 of a vehicle that is actually traveling on a road surface. Note that an information processing device 10 with a built-in acceleration sensor 4 is attached to the inner surface of each tire 2 that is located behind the tread 2A of each tire 2 attached to the vehicle.

[0076] The CPU 20A of the information processing device 10 reads the information processing program stored in the nonvolatile memory 20C, develops it in the RAM 20B, and executes the process of estimating the longitudinal force.

[0077] The information processing device 10 performs processing to estimate the longitudinal force for each tire 2, but for ease of explanation, the following description will focus on processing to estimate the longitudinal force for one tire 2. That is, similar processing to estimate the longitudinal force is performed for the other tires 2.

[0078] In step S10, the CPU 20A acquires an acceleration signal from the acceleration sensor 4 attached to the inner surface of the tire 2. The acquired acceleration signal is an acceleration signal in at least one direction along the X-axis direction, the Y-axis direction, and the Z-axis direction. Here, as an example, the CPU 20A acquires an acceleration signal in any one direction. Specifically, the CPU 20A acquires an acceleration signal along the X-axis direction.

[0079] In step S20, the CPU 20A extracts a feature amount that contributes to estimating the longitudinal force from the acceleration signal of the rear edge of the tire 2 in each rotation, among the acceleration signals acquired by the process of step S10.

[0080] Based on the knowledge already described, the CPU 20A extracts the standard deviation of the acceleration signal, the autocorrelation feature of the acceleration signal, and the unit root process feature of the acceleration signal as time domain feature amounts from the acceleration signal of the rear edge of the tire 2 in each rotation.

[0081] Furthermore, the CPU 20A extracts the center frequency of the acceleration signal as a frequency domain feature from the acceleration signal of the rear edge at each rotation of the tire 2. The center frequency of the acceleration signal may be extracted using, for example, a short-term Fourier transform (STFT) or a holo-hilbert spectral analysis (HHSA).

[0082] In step S30, the CPU 20A estimates the longitudinal force of the tire 2 using the standard deviation of the acceleration signal, the autocorrelation feature amount of the acceleration signal, the unit root process feature amount of the acceleration signal, and the center frequency of the acceleration signal extracted by the process of step S20.

[0083] An estimation model that has been trained in advance using, for example, machine learning to determine the relationship between the feature values ​​and the longitudinal force of the tire 2 is used to estimate the longitudinal force of the tire 2. Specifically, the estimation model is generated by machine learning using training data in which the standard deviation of the acceleration signal, the autocorrelation feature value of the acceleration signal, the unit root process feature value of the acceleration signal, and the center frequency of the acceleration signal are used as input data, and the longitudinal force of the tire 2 when these feature values ​​are obtained is used as training data.

[0084] The CPU 20A estimates the longitudinal force of the tire 2 by inputting the standard deviation of the acceleration signal, the autocorrelation feature of the acceleration signal, the unit root process feature of the acceleration signal, and the center frequency of the acceleration signal extracted by the processing of step S20 into a trained estimation model pre-stored in the non-volatile memory 20C.

[0085] The method for estimating the longitudinal force of the tire 2 is not limited to the method using a trained estimation model. The CPU 20A may estimate the longitudinal force of the tire 2 using an estimation function that uses, as explanatory variables, for example, the standard deviation of the acceleration signal, the autocorrelation feature amount of the acceleration signal, the unit root process feature amount of the acceleration signal, and the center frequency of the acceleration signal, and the longitudinal force of the tire 2 as a response variable.

[0086] By executing the processing of steps S10 to S30 for each tire 2 of the vehicle, it is possible to estimate the longitudinal force for each tire 2. Note that the CPU 20A may execute the longitudinal force estimation processing only for the tire 2 specified by the user. With the above, the longitudinal force estimation processing shown in FIG. 11 is completed.

[0087] As described above, the information processing device 10 of the present disclosure estimates the longitudinal force of the tire 2 by extracting only feature quantities that contribute to estimating the longitudinal force from the acceleration signal. Therefore, compared to estimating the longitudinal force of the tire 2 by randomly using all kinds of feature quantities obtained from the acceleration signal without narrowing down the feature quantities that contribute to estimating the longitudinal force, the information processing device 10 can reduce the amount of calculation required to estimate the longitudinal force. Because the battery 22 that powers the information processing device 10 is attached to the inner surface of the tire 2, replacing the battery 22 requires removing the tire 2 from the wheel rim, which is a time-consuming task. Because the information processing device 10 can reduce the amount of calculation required to estimate the longitudinal force as described above, the power consumption required to estimate the longitudinal force of the tire 2 is also less than when estimating the longitudinal force of the tire 2 by using all kinds of feature quantities obtained from the acceleration signal. In other words, the information processing device 10 can suppress consumption of the battery 22 compared to a comparable device, thereby reducing the effort required for battery replacement.

[0088] In the longitudinal force estimation process shown in FIG. 11, an example has been described in which the longitudinal force of the tire 2 is estimated using an acceleration signal along the X-axis direction. However, the CPU 20A may estimate the longitudinal force of the tire 2 using an acceleration signal along the Y-axis direction or an acceleration signal along the Z-axis direction instead of the acceleration signal along the X-axis direction.

[0089] Furthermore, the CPU 20A may estimate the longitudinal force of the tire 2 using acceleration signals in two directions along any two of the X-axis, Y-axis, and Z-axis. Furthermore, the CPU 20A may estimate the longitudinal force of the tire 2 using acceleration signals in three directions along each of the X-axis, Y-axis, and Z-axis.

[0090] When estimating the longitudinal force of the tire 2 using acceleration signals in two directions, the CPU 20A may estimate the longitudinal force of the tire 2 using an estimation model that has previously learned the relationship between each feature amount obtained from the acceleration signals in two directions and the longitudinal force of the tire 2. Specifically, the CPU 20A inputs eight feature amounts, which are the standard deviation of the acceleration signals, the autocorrelation feature amount of the acceleration signals, the unit root process feature amount of the acceleration signals, and the center frequency of the acceleration signals, extracted from the acceleration signals along any two directions of the X-axis and Y-axis, the Y-axis and Z-axis, or the X-axis and Z-axis, into the estimation model, and estimates the longitudinal force of the tire 2.

[0091] Furthermore, when estimating the longitudinal force of the tire 2 using acceleration signals in three directions, the CPU 20A may estimate the longitudinal force of the tire 2 using an estimation model that has previously learned the relationship between each feature amount obtained from the acceleration signals in three directions and the longitudinal force of the tire 2. Specifically, the CPU 20A inputs 12 feature amounts, which are the standard deviation of the acceleration signals, the autocorrelation feature amount of the acceleration signals, the unit root process feature amount of the acceleration signals, and the center frequency of the acceleration signals, extracted from the acceleration signals along the X-axis, Y-axis, and Z-axis directions, into the estimation model, and estimates the longitudinal force of the tire 2.

[0092] 11, the longitudinal force of the tire 2 is estimated using four feature quantities: the standard deviation of the acceleration signal, the autocorrelation feature quantity of the acceleration signal, the unit root process feature quantity of the acceleration signal, and the center frequency of the acceleration signal. However, since each of these feature quantities alone is a value that reflects the longitudinal force of the tire 2, the CPU 20A may estimate the longitudinal force of the tire 2 using at least one of the four feature quantities. For example, the CPU 20A may estimate the longitudinal force of the tire 2 using only the standard deviation of the acceleration signal.

[0093] In addition, the CPU 20A may estimate the longitudinal force of the tire 2 using at least one time domain feature from among the standard deviation of the acceleration signal, the autocorrelation feature of the acceleration signal, and the unit root process feature of the acceleration signal, and the center frequency of the acceleration signal, which is an example of a frequency domain feature.

[0094] Furthermore, in the present disclosure, an example has been described in which the information processing device 10 includes the estimation unit 13, but the information processing device 10 does not necessarily include the estimation unit 13. In this case, the estimation unit 13 is provided in an external device other than the information processing device 10, for example, an ECU.

[0095] When the longitudinal force of the tire 2 is estimated by the estimation unit 13 of the ECU, the ECU wirelessly acquires the feature quantities used to estimate the longitudinal force from the information processing device 10. In this case, the information processing device 10 only needs to transmit the feature quantities that contribute to the estimation of the longitudinal force to the ECU. Therefore, compared to, for example, transmitting the acceleration signal itself to the ECU or blindly transmitting all kinds of feature quantities obtained from the acceleration signal to the ECU, the information processing device 10 can reduce the power consumption associated with data transmission. In other words, the information processing device 10 can suppress consumption of the battery 22 compared to the comparative device, and therefore can also reduce the effort associated with battery replacement.

[0096] While one form of the information processing device 10 has been described above using the embodiment, the disclosed form is merely an example, and the form of the information processing device 10 is not limited to the scope described in the embodiment. Various changes or improvements can be made to the embodiment without departing from the gist of the present disclosure, and forms of the information processing device 10 with such changes or improvements are also included in the technical scope of the disclosure.

[0097] In the above embodiment, as an example, the longitudinal force estimation process shown in FIG. 11 is implemented by software. However, the process equivalent to the flowchart of the longitudinal force estimation process may be executed by hardware. In this case, the processing speed can be increased compared to when the longitudinal force estimation process is implemented by software.

[0098] In the above embodiment, an example has been described in which the information processing program is stored in the nonvolatile memory 20C. However, the storage destination of the information processing program is not limited to the nonvolatile memory 20C. The information processing program may also be provided in a form recorded on a computer-readable storage medium.

[0099] For example, the information processing program may be provided in a form recorded on a portable semiconductor memory such as a USB (Universal Serial Bus) memory or a memory card. The non-volatile memory 20C, the USB, and the memory card are examples of non-transitory storage media.

[0100] Furthermore, the CPU 20A may download an information processing program from an external device via the communication unit 21 and store the downloaded information processing program in the nonvolatile memory 20C.

[0101] In the embodiments, CPU20A has been used as an example of a general-purpose processor, but in the embodiments, the term "processor" refers to a processor in a broad sense, and includes not only general-purpose processors such as CPU20A, but also dedicated processors (e.g., GPU: Graphics Processing Unit, ASIC: Application Specific Integrated Circuit, FPGA: Field Programmable Gate Array, programmable logic device, etc.).

[0102] Furthermore, the operation of the processor in the above-described embodiments may not only be performed by a single processor, but may also be performed by multiple processors working together, or may be performed by multiple processors located in physically separate locations working together.

[0103] The present disclosure is also applicable to programs and program products.

[0104] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[0105] [Contribution to the United Nations-led Sustainable Development Goals (SDGs)] The SDGs have been proposed to realize a sustainable society. One embodiment of the present invention is expected to contribute to the achievement of "No. 9 - Building infrastructure for industry and technological innovation." [Explanation of symbols]

[0106] 1. Experimental equipment 2 tires 2A Tread 3 Rotation Axis 4. Accelerometer 5 Belt 6 spots 7. High-speed camera 8 Data Collection Equipment 9. Computer 10. Information processing equipment 11 Acquisition Department 12 Extraction part 13 Estimation part 15 images 16A Graph of acceleration signal in the X-axis direction 16B Acceleration signal graph in the Y-axis direction 16C Acceleration signal graph in the Z-axis direction 20 Computer 20A CPU 20B RAM 20C Non-volatile Memory 20D I / O 20E Bus 21 Communication unit 22 Battery R1 Arrow indicating the position of the leading edge R2 Arrow indicating rear edge position R3 Arrow indicating the direction of tire rotation SP1, SP2, SP3 Plot points for each tire movement speed

Claims

1. an acquisition unit that acquires an acceleration signal from an acceleration sensor attached to an inner surface of the tire; an extracting unit that extracts, from the acceleration signal acquired by the acquiring unit, a time domain feature that is a feature related to a time domain of the acceleration signal and a frequency domain feature that is a feature related to a frequency domain of the acceleration signal; an estimation unit that estimates a longitudinal force generated along a tangential direction of a contact patch of the tire with a road surface, using the time domain feature amount and the frequency domain feature amount extracted by the extraction unit; An information processing device comprising:

2. The extraction unit extracts the time domain feature amount and the frequency domain feature amount from the acceleration signal during a measurement period of the acceleration signal, the period being limited to a predetermined period including a time point at which the tread of the tire in contact with the road surface separates from the road surface. The information processing device according to claim 1 .

3. the acceleration sensor is a sensor that measures the acceleration signals along at least three directions, namely, a tangential direction of the tire, a width direction of the tire, and a direction of gravity, The extraction unit extracts the time domain feature quantity and the frequency domain feature quantity from the acceleration signal along at least one of a tangential direction of the tire, a width direction of the tire, and a gravity direction. The information processing device according to claim 2 .

4. The extraction unit extracts the time domain feature amount and the frequency domain feature amount from the acceleration signal along three directions, namely, a tangential direction of the tire, a width direction of the tire, and a gravity direction. The information processing device according to claim 3 .

5. The extraction unit extracts, from the acceleration signal, at least one of a standard deviation of the acceleration signal, a feature based on autocorrelation of the acceleration signal, and a feature based on non-stationarity of the acceleration signal as the time-domain feature.

5. The information processing device according to claim 1.

6. The extraction unit extracts, from the acceleration signal, a standard deviation of the acceleration signal, a feature based on autocorrelation of the acceleration signal, and a feature based on non-stationarity of the acceleration signal as the time domain feature. The information processing device according to claim 5 .

7. The extraction unit extracts a time lag at which a value of the autocorrelation of the acceleration signal decreases to 1 / e compared to the autocorrelation of the acceleration signal at a time lag of 0 as a feature based on the autocorrelation of the acceleration signal. The information processing device according to claim 6 .

8. The extraction unit extracts a time lag at which the autocorrelation of the acceleration signal first becomes zero as a feature quantity based on the autocorrelation of the acceleration signal. The information processing device according to claim 6 .

9. The extraction unit extracts a value representing a degree to which the acceleration signal follows a unit root process as a feature quantity based on non-stationarity of the acceleration signal. The information processing device according to claim 5 .

10. The extraction unit determines a ratio of a time lag from a time point when the time lag is 0 in the autocorrelation of the difference series of the acceleration signal until the autocorrelation of the difference series first becomes 0 to a time lag from a time point when the time lag is 0 in the autocorrelation of the acceleration signal until the autocorrelation of the acceleration signal first becomes 0 as a value representing the degree to which the acceleration signal follows a unit root process. The information processing device according to claim 9 .

11. The extraction unit extracts a center frequency of a frequency distribution included in the acceleration signal as the frequency domain feature. The information processing device according to claim 5 .

12. The acceleration signal is acquired from an acceleration sensor attached to the inner surface of the tire. extracting, from the acquired acceleration signal, a time domain feature which is a feature relating to a time domain of the acceleration signal, and a frequency domain feature which is a feature relating to a frequency domain of the acceleration signal; and causing a computer to execute a process of estimating a longitudinal force generated along a tangential direction of a contact surface of the tire with a road surface in association with rotation control of the tire, using the extracted time domain feature amount and the extracted frequency domain feature amount. Information processing program.

13. The acceleration signal is acquired from an acceleration sensor attached to the inner surface of the tire. extracting, from the acquired acceleration signal, a time domain feature which is a feature relating to a time domain of the acceleration signal, and a frequency domain feature which is a feature relating to a frequency domain of the acceleration signal; A computer executes a process of estimating a longitudinal force generated along a tangential direction of a contact surface of the tire with a road surface in association with rotation control of the tire, using the extracted time domain feature amount and the extracted frequency domain feature amount. Information processing methods.

Images