JP5878612B2 - Road surface condition estimation method - Google Patents

Description

  The present invention relates to a method for estimating a road surface state during traveling, and more particularly to a method for subdividing and estimating a snow road state.

  In order to improve the driving safety of an automobile, it is required to accurately estimate the state of the road surface during traveling and feed it back to vehicle control. If it is possible to estimate the condition of the road surface during traveling, it is expected that, for example, more advanced control of the ABS brake and the like can be performed before the risk avoidance operation such as braking / driving and steering, and safety is further improved. Is done.

  As a method for estimating the road surface condition during traveling, for example, the vibration of the tire during traveling is detected, and the time region before the depression position (the region before depression) is detected from the time-series waveform of the detected tire vibration. Vibration level ratio R, which is the ratio of the magnitude of the vibration component in the high frequency region (3 kHz to 5 kHz band) to the magnitude of the vibration component in the low frequency region (1 kHz to 2 kHz band) of this vibration. A method for calculating and estimating whether the road surface state is a high μ road or a low μ road from the vibration level ratio R has been proposed (see, for example, Patent Document 1).

  Further, by detecting the tire generation sound generated from the running tire, calculating the average value of the sound pressure level within the set frequency range of the detected tire generation sound, and comparing this average value with the reference value , A method to estimate whether the road surface is very wet, slightly wet asphalt, dry asphalt or icy road, and the ratio of the average value of sound pressure levels in multiple frequency ranges Compared to the above, there has also been proposed a method for estimating whether the road surface is a snowy road, a flooded road, an icy road, or a dry road (see, for example, Patent Documents 2 and 3).

WO 2006/135090 A1 JP-A-6-174543 JP-A-8-261993

  However, with the conventional method for estimating the road surface condition, it is possible to determine whether the running road surface is a compressed snow road or an icy road, but the snow road is a snow covered road with fresh snow, or the surface is a sherbet. I could not classify the snow road more finely, such as whether it was a snow road.

  The present invention has been made in view of the conventional problems, and an object of the present invention is to provide a road surface state estimation method capable of further subdividing and estimating a running snow road.

This onset Ming is a method of estimating the state of the road surface during running, the time-series waveform of the vibration of the tire during running, the previous depression front region R1 than the peak of the leading side appearing in the leading edge, Form a stepping region R2 forming the stepping side peak, a region R3 before kicking between the stepping side peak and the kicking side peak appearing at the kicking end, and the kicking side peak A step (a) for dividing the kick-out area R4 into a kick-out area R5 after the kick-out peak, and a specific frequency set in advance for each of the areas R1 to R5 from the time-series waveform. and step (b) to obtain the bandwidth value is the magnitude of the vibration component of the band, on the basis of the bandwidth value, possess a step (c) for estimating the road surface condition during travel, in the step (c), Said stepping Wherein the bandwidth value front region R1, the band value of the depression region R2, by using the band value of region R5 after out the kicking, that the road surface condition during travel is determined whether the snow road And
Thereby, since it can be estimated that the road surface is a snowy road on which fresh snow is piled up, the state of the snowy road can be estimated in more detail.
In the step (c) of the present invention, the band value of the pre-kicking region R3, the band value of the kicking region R4, and the band value of the post-kicking region R5 are used, or the kicking The road surface condition during running is a sherbet-like snow using the band value of the post-pick-up area R5 and the band value of the area spanning the pre-kick-up area R3, the kick-out area R4, and the post-kick-up area R5. It is characterized by determining whether it is a road.
Thereby, since it can be estimated that the road surface is a sherbet-like snow road, the state of the snow road can be estimated more finely.
According to the present invention, in the step (c), a band value of a region straddling the kicking region R4 and the post-kicking region R5, or a band value of the kicking region R4 and a band of the post-kicking region R5 Using the value, it is determined whether or not the running road surface state is a frozen road.
Thereby, since it can be estimated that the road surface is a frozen road, the state of a snowy road can be estimated more finely.
In the step (c) of the present invention, the band value of the pre-stepping area R1, the band value of the stepping area R2, the band value of the pre-kicking area R3, and the band of the kicking area R4 Using the value and the band value of the post-kicking region R5, it is determined whether or not the running road surface condition is a snow-capped road.
Thereby, since it can be estimated that the road surface is a snow-capped road, the state of the snow road can be estimated more finely.
Further, according to the present invention, a step (d) of detecting a road surface temperature and a tire generated sound during traveling between the step (b) and the step (c), and a low frequency band of the tire generated sound and step (e) for obtaining a band value P _B of the band value P _a and the high frequency band, and a band value P _B of the band value P _a and the high frequency band of the said surface temperature low frequency band, inclusions on the road surface Step (f) for determining whether or not there is an object, and in step (f), when it is determined that there is an inclusion on the road surface, The band value of the kicking area R4 and the band value of the post-kicking area R5 are used, or the band value of the post-kicking area R5, the pre-kicking area R3, the kicking area R4, and the kicking Using the band value of the area that spans the post-out area R5, Road surface condition or shallow sherbet-like snowy of and judging whether shallow wet road surface.
As a result, for the “sorbet-like snow road” and “WET road”, “shallow sherbet-like snow road”, “deep sherbet-like snow road”, “shallow WET road” and “deep WET road” respectively. Therefore, the estimation accuracy of the road surface condition can be further improved.
In the present invention, when it is determined in step (f) that there is no inclusion on the road surface, the band value of the pre-depression region R1, the band value of the depressing region R2, and the kicking The band value of the pre-development region R3, the band value of the kick-out R4, and the band value of the post-pick-up region R5 are used, or the band value of the pre-depression region R1 and the band of the depressing region R2 Whether or not the road surface condition during running is a smooth dry road using the value, the band value of the area R3 before kicking, and the band value of the area straddling the kicking area R4 and the post-kicking area R5 It is characterized by determining.
As a result, it is possible to estimate whether the running road surface is a smooth dry road, a snow-capped road, or a rough dry road, so that the estimation accuracy of the road surface condition can be further improved.
The invention according to claim 7 is characterized in that the time series waveform of the tire vibration is a time series waveform of the tire vibration detected by an acceleration sensor installed in the tire.
The invention according to claim 8 is characterized in that the tire vibration is vibration in a tire circumferential direction or vibration in a tire width direction of a running tire.
According to the ninth aspect of the present invention, in the step (c), the band value obtained in the step (b) is substituted into an identification function representing a relationship between the road surface state and the band value obtained in advance. A road surface condition during traveling is determined based on the obtained function value.

The band value P _ij may be an average value of vibration levels in a designated frequency band, or an average value of a region having a vibration level that has a particularly large difference depending on conditions in the designated frequency band. There may be. Alternatively, it may be an average value or a sum of one or a plurality of vibration levels set in advance in a designated frequency band.
In addition, the discrimination function used for the determination in the present invention is a function for discriminating two or more sets. As shown in FIG. 18, a certain condition A (in this case, an inclusion on the road surface such as water or snow) P ₁₁ ≧ P ₁₂ in the case of satisfying the condition that the tire and the tire collide), and in the case of not satisfying the condition A, if P ₁₁ <P ₁₂ , the band value satisfying the condition A (P ₁₁ , P ₁₂ ) and the boundary line between the set of band values (P ₁₁ , P ₁₂ ) that do not satisfy the condition A is P ₁₁ = P ₁₂ . In this case, identification function _{F1 = P 11 -P 12 (w} 11 = 1, w 12 = -1, K1 = 0), the boundary line becomes F1 = 0. That is, when substituting (P _11, P ₁₂₎ into F1, when F1 ≧ 0 is a P ₁₁ ≧ P _12, when F1 <0 is a P ₁₁ <P _12.
Common identification function, with _{F1 = w 11 · P 11 +} w 12 · P 12 -K1, w 11, w 12, K1 is set actually obtained bandwidth value for the condition A (P _11, P ₁₂₎ Can be calculated using a known method such as the least square method, Mahalanobis distance, or SVM.
Not only the two-variable discriminant function F2, but also the multi-variable discriminant functions F3, F4, F7, and F8 can be obtained in the same manner.
The function F′1 = P ₁₂ −P ₁₁ is also an identification function that can perform the same identification as the identification function F1 = P ₁₁ −P ₁₂ . When this discriminant function F′1 is used, the sign of determination is reversed. The general discriminant function F = w ₁ · P ₁ + w ₂ · P ₂ +... + W _n · P _n −K is also the discriminant function F ′ = w ′ ₁ · P ₁ + w ′ ₂ · P ₂ +. + W ′ _n · P _n −K ′ (w ′ _k = −w _k , K ′ = − K) can be identified in the same manner, and when the discrimination function F ′ is used, the sign of determination is reversed. To do.

  The summary of the invention does not list all necessary features of the present invention, and sub-combinations of these feature groups can also be the invention.

It is a functional block diagram which shows the structure of the road surface state estimation system which concerns on this Embodiment. It is a figure which shows arrangement | positioning of a sensor etc. FIG. It is a functional block diagram which shows the structure of the arithmetic unit which estimates a road surface state. It is a figure which shows the area before stepping in, the stepping area, the area before kicking out, the area kicking out, and the area after kicking out in the time series waveform of tire vibration. It is a figure which shows the octave distribution waveform of a sound pressure signal. It is a flowchart which shows the estimation method of the road surface state by this invention. It is a figure which shows an example of the frequency spectrum of the area before stepping on. It is a figure which shows an example of the frequency spectrum of a stepping area. It is a figure which shows an example of the frequency spectrum of the area | region after kicking out. It is a figure which shows the other example of the frequency spectrum of the area | region after kicking out. It is a figure which shows an example of the frequency spectrum of the area | region before kicking out. It is a figure which shows the other example of the frequency spectrum of the area | region after kicking out. It is a figure which shows the other example of the frequency spectrum of the area | region before stepping on. It is a figure which shows the other example of the frequency spectrum of a stepping area. It is a figure which shows the other example of the frequency spectrum of the area | region before kicking out. It is a figure which shows an example of the frequency spectrum of a kick-out area | region. It is a figure which shows the other example of the frequency spectrum of the area | region after kicking out. It is a figure for demonstrating an identification function.

  Hereinafter, the present invention will be described in detail through embodiments, but the following embodiments do not limit the invention according to the claims, and all combinations of features described in the embodiments are included. It is not necessarily essential for the solution of the invention.

FIG. 1 is a functional block diagram showing a configuration of a road surface state estimation system 1 according to an embodiment of the present invention.
Each component of the road surface state estimation system 1 is distributed and installed in a tire 10 mounted on a vehicle, a vehicle body 20 and a vehicle interior 20M, and a management center 40 that accumulates and manages data such as estimated road surface state. Is done.
In the tire 10, an acceleration sensor 11, an amplifier 12, an A / D converter 13, a tire side transmission device 14, a power reception device 15, and a power supply device 16 are arranged.
The vehicle body 20 and the interior 20M include a road surface thermometer 21, a microphone 22, a wheel speed sensor 23, a monitoring acceleration sensor 24, a camera 25, a GPS 26, a computing device 27, a road surface information recording unit 28, A monitor 29, a receiving device 30, a vehicle-side transmitting device 31, a power feeding device 32, and an alarm device 33 are installed.
In the management center 40, a data server 41 and a display device 42 are installed.

As shown in FIG. 2, the acceleration sensor 11 is arranged at a substantially central portion on the tire air chamber 10 b side of the inner liner portion 10 a of the tire 10 of the front wheel 20 </ b> F, and is input to the tread 10 c of the tire 10 from the road surface R. Is detected. In this example, the acceleration sensor 11 is arranged so that the detection direction is the tire circumferential direction.
The amplifier 12 includes a low-pass filter, removes high-frequency noise components from the time series waveform (acceleration waveform) of the tire circumferential vibration detected by the acceleration sensor 11, and outputs it to the A / D converter 13.
The A / D converter 13 outputs the A / D converted acceleration waveform to the tire-side transmission device 14.
The tire side transmission device 14 transmits the A / D converted acceleration waveform to the reception device 30 provided on the vehicle body 20 side by radio.
The power receiving device 15 receives the RF signal for power supply transmitted from the power supply device 32 and outputs it to the power supply device 16. In this example, the power receiving device 15 is constituted by a receiving coil, and a plurality of receiving coils are arranged on the circumference of the tire 10 at equal intervals, so that the RF signal is received almost continuously when the tire 10 rotates. I can do it.
The power supply device 16 converts the RF signal received by the power receiving device 15 into power and charges the capacitor, and supplies the power to the amplifier 12, the A / D converter 13, and the tire side transmission device 14.
The tire-side transmitting device 14 and the power receiving device 15 may be disposed on the inner liner portion 10a integrally with the acceleration sensor 11, but may be configured to be attached to the valve 10v of the tire 10.

The road surface thermometer 21 is composed of, for example, an infrared temperature sensor, and is installed in the lower part of the front bumper 20a of the vehicle body 20 as shown in FIG. To measure the temperature of the road surface R. The temperature data measured by the road surface thermometer 21 is output to the arithmetic unit 27 and the road surface information recording means 28.
The microphone 22 is attached to the lower part of the frame 20b in front of the rear wheel 20R of the vehicle body 20, and detects the sound pressure signal of the tire generated sound. Tire generated sound is generated near the tire contact surface when the tire 10 of the rear wheel 20R contacts the road surface R when the vehicle is traveling. The sound pressure signal of the tire generated sound detected by the microphone 22 is output to the arithmetic unit 27.
The wheel speed sensor 23 detects the rotational speed of the wheel (hereinafter referred to as the wheel speed). In this example, a rotor having a gear formed on the outer peripheral portion thereof and rotating together with the wheel, and a yoke constituting the rotor and a magnetic circuit, And a coil that detects a change in magnetic flux of the magnetic circuit, and uses a known electromagnetic induction type wheel speed sensor that detects a rotation angle of a wheel (here, the front wheel 20F). The yoke and the coil are attached to a knuckle 52 that is rotatably attached to the axle 51. The wheel speed data detected by the wheel speed sensor 23 is output to the arithmetic unit 27 and the road surface information recording means 28.
The monitoring acceleration sensor 24 is attached to the knuckle 52 and detects tread vibration (acceleration of the suspension part) transmitted from the tire 10 to the lower part of the vehicle spring via the wheel 53. The vibration data under the vehicle spring detected by the monitoring acceleration sensor 24 is output to the arithmetic unit 27.

The camera 25 is composed of, for example, a CCD color camera, and is installed on the roof 20c of the vehicle body 20 as shown in FIG. 2, and photographs the state of the road surface (road surface in front of the vehicle) R on which the vehicle is to pass.
The GPS 26 is installed in a driver's seat within the vehicle 20M and measures the absolute position of the vehicle on the ground.
The image data of the video imaged by the camera 25 and the vehicle position data measured by the GPS 26 are output to the road surface information recording means 28.
The computing device 27 includes tire vibration data detected by the acceleration sensor 11, road surface temperature data detected by the road surface thermometer 21, tire generated sound data detected by the microphone 22, and wheels detected by the wheel speed sensor 23. Estimate the road surface condition while driving from the speed data. The tire vibration data is transmitted from the tire-side transmitting device 14 to the receiving device 30 and is output from the receiving device 30 to the arithmetic device 27.
Details of the arithmetic unit 27 will be described later.
The road surface information recording means 28 is measured by the road surface thermometer 21, the road surface state during traveling estimated by the computing device 27, the image data of the video taken by the camera 25, the vehicle position data measured by the GPS 26, and the road surface thermometer 21. Display image data obtained by synthesizing the temperature data and the wheel speed data detected by the wheel speed sensor 23 is generated and output to the monitor 29 and the vehicle-side transmitter 31.
The monitor 29 displays the display image data input from the road surface information recording means 28 on the display screen.
The arithmetic device 27 and the road surface information recording means 28 are each constituted by microcomputer software, and are installed in the vicinity of the driver's seat together with the monitor 29.

The receiving device 30 receives tire vibration data transmitted from the tire-side transmitting device 14 and outputs the tire vibration data to the arithmetic device 27.
The vehicle-side transmission device 31 includes information on the vehicle (estimated road surface state information, road surface image data, vehicle position data, road surface temperature data, and wheels) input to the road surface information recording means 28. Speed data) is transmitted to the data server 41 of the management center 40.
The power feeding device 32 includes a high frequency generation unit and a power supply coil, and transmits the high frequency generated by the high frequency generation unit to the power reception device 15 as an RF signal for power supply from the power supply coil. The power feeding device 32 is installed in the tire house 20 d of the vehicle body 20 at a position facing a receiving coil (not shown) of the power receiving device 15 provided in the tire 10.
The alarm device 33 is installed in the vicinity of the driver's seat, and when it is estimated that the road surface is a deep WET state or a frozen road, the driver should be cautiously driven by lighting or blinking an alarm LED. Warning.
Note that an alarm buzzer may be driven to cause the driver to recognize that the occurrence of a hydroplaning phenomenon is predicted by an alarm sound, or an alarm buzzer and an LED may be used in combination.

The data server 41 of the management center 40 aggregates and stores vehicle information sent from the vehicle on which the road surface state estimation system 1 is mounted, and from the aggregated data, for example, a road surface in a predetermined area. Create statistical data, such as combining the state data with the map data of the area.
The display device 42 displays statistical data or vehicle data of a specific vehicle on the display screen.
The management center 40 and each vehicle can communicate with each other, and the statistical data created by the data server 41 is fed back to each vehicle, so that warnings and cautions for vehicles passing the road surface in a predetermined area can be provided. If prompted, the traveling safety of the vehicle can be further improved.

As shown in FIG. 3, the computing device 27 includes a tire vibration data processing unit 27A, a sound data processing unit 27B, and road surface state estimating means 27C.
The tire vibration data processing unit 27A includes a vibration waveform detection unit 27a, a region signal extraction unit 27b, and a band value calculation unit 27c. The tire vibration data processing unit 27A converts tire vibration data detected by the acceleration sensor 11 into a band value P _ij and the road surface state. It outputs to the estimation means 27C.
The vibration waveform detecting means 27a calculates the time for one revolution of the tire from the peak position on the depression side or the peak position on the kicking side of the acceleration signal that is the output signal of the acceleration sensor 11 as shown in FIG. At the same time, an acceleration waveform for one rotation of the tire is detected using the wheel speed detected by the wheel speed sensor 23.
As shown in FIG. 4 (b), the region signal extraction means 27b generates an acceleration waveform for one rotation of the tire before the stepping side peak R1 that appears at the stepping end and the stepping side peak. A kick-in area R3 between the step-on peak and the kick-out peak appearing at the kick-out end, and a kick-out area R4 forming the kick-out peak. Then, it is divided into a post-kick region R5 after the peak on the kick side, and a time series waveform of vibration levels in each region R1 to R5 is extracted.

The band value calculating means 27c passes the time series waveform of each of the regions R1 to R5 through a bandpass filter, and calculates a band value P _ij that is a magnitude of a vibration component in a predetermined frequency region as shown below. The subscript i indicates a region, and the subscript j indicates an extracted frequency region.
A band pass filter is provided for each frequency region.
P ₁₁ : Band value selected from the 2 kHz to 8 kHz band in the pre-stepping region R 1 P ₁₂ ; Band value selected from the 0.5 kHz to 1.5 kHz band in the pre-stepping region R 1 P ₁₃ ; bandwidth value selected from the range of P _21; band value P chosen from 1kHz~3kHz band of depression region R2 _22; bandwidth value selected from 2kHz~4kHz band of depression region R2 P _23; 4 kHz depression region R2 Band value P ₃₁ selected from the 10 kHz band P ₃₁ ; Band value P ₃₂ selected from the 7 kHz to 10 kHz band in the pre-kicking region R 3; Band value P ₃₃ selected from the 2 kHz to 4 kHz band in the pre-kicking region R 3; Band value P ₄₁ selected from the 4 kHz to 10 kHz band in the region R3 before kicking; 7 kHz to 10 in the kicking region R4 Band value P ₄₂ selected from the kHz band; Band value P ₅₁ selected from the 1 kHz to 4 kHz band in the kick-out area R4; Band value P ₅₂ selected from the 1 kHz to 4 kHz band in the kick-out area R5; Band value selected from the 2 kHz to 4 kHz band in the rear region R5 P ₅₃ ; Band value selected from the 7 kHz to 10 kHz band in the rear region R5

The sound data processing unit 27B includes a frequency analysis unit 27d, a sound pressure level calculation unit 27e, and a sound pressure level ratio calculation unit 27f, and calculates a sound pressure level ratio Q = (P _B / P _A ) to calculate the road surface. It outputs to the state estimation means 27C.
The frequency analyzing means 27d analyzes the sound pressure signal of the tire generated sound by 1 / N octave to obtain a sound pressure signal distribution waveform (octave distribution waveform) as shown in FIG. The octave distribution waveform is obtained by measuring the sound pressure level (band power) for each octave band divided into 1 / N octave bands. In this example, N = 3.
Sound pressure level calculation means 27e from octave distribution waveform, a low frequency band (e.g., 500 Hz) band power values P _A and the high frequency band (e.g., 9000 Hz) calculates a band power value P _B with.
Sound pressure level ratio calculating means 27f calculates the sound pressure level calculated value from a band power value P _B of band power values P _A and the high frequency band of the low frequency band. In this example, the sound pressure level calculation value is a sound pressure level ratio Q which is a ratio of the band power value P _{B in the} high frequency band to the band power value P _A in the low frequency band. Q = (P _B / P _A ).

The road surface state estimating means 27C is configured to calculate the tire circumferential vibration band value P _ij input from the tire vibration data processing unit 27A, the sound pressure level ratio Q input from the sound data processing unit 27B, and the road surface thermometer 21. Using the road surface temperature T data that is input, the road surface is “snow-covered road”, “shallow sherbet-like snow road”, “deep sherbet-like snow road”, “shallow WET road”, “deep WET road”, It is determined which of the eight states of “frozen road”, “pressed snow road”, and “dry road”, and the determination result is output to the road surface information recording means 28. It is also possible to determine whether the “drying path” is a “rough drying path” or a “smooth drying path”.
When it is determined that the vehicle is in a deep WET state or a frozen road, the determination result is output to the alarm device 33.
When the acceleration value of the suspension detected by the monitoring acceleration sensor 24 exceeds a preset threshold value, the road surface state estimation unit 27C stops the road surface determination.
Below, the estimation method of a road surface state is demonstrated with reference to the flowchart of FIG.
First, it is determined whether or not there are inclusions such as water and snow on the road surface (step S1).
When there are inclusions such as water and snow on the road surface, the water film, snow, and tire collide when stepping on, so the time series waveform in the region R1 before stepping in FIG. As shown in the obtained frequency spectrum, the band value P ₁₁ selected from the 2 kHz to 8 kHz band in the pre-step-down region R1 is a value on a WET road or a sherbet-like snow road from a value on a dry road (dry). Also grows. However, the bandwidth value P ₁₂ is selected from 0.5kHz~1.5kHz band of depression front region R1 is small difference due to road conditions. Therefore, using the band value P ₁₂ as a reference value, a difference (dB difference) P ₁₁ -P ₁₂ between the band value P ₁₁ and the band value P ₁₂ is obtained, and the magnitude of this difference P ₁₁ -P ₁₂ is obtained. set the threshold value K1, it is determined and when the magnitude of the difference P ₁₁ -P ₁₂ is the threshold value K1 or more, there are inclusions such as water or snow on the road surface. When the difference P ₁₁ -P ₁₂ is smaller than the threshold value K1, there is no inclusion such as water or snow on the road surface, or the water film or the sherbet-like snow layer is thin. Is determined.
In this example, the relationship between the band value P ₁₁ and the band value P ₁₂ in various road surface conditions is experimentally obtained in advance, and the discriminant function F1 = w ₁₁ · P ₁₁ + w ₁₂ · P ₁₂ −K1 is set. calculated bandwidth value P ₁₁ and band value P ₁₂ and the identification function F1 function value f1 obtained by substituting the found depending on whether meets f1 ≧ 0, inclusions such as water or snow on the road surface It was determined whether there was. The coefficient w _{11 of the} discrimination function is about +1 and the coefficient w ₁₂ is about -1.

If f1 ≧ 0 in step S1, the process proceeds to step S2 to determine whether the inclusions on the road surface are soft with fresh snow.
When the inclusion is fresh snow, the impact when stepping on is reduced by the snow, and the road surface becomes slippery. Accordingly, as shown in the frequency spectrum of the depression region R2 in FIG. 8, the value of the bandwidth value P ₂₁ is selected from 1kHz~3kHz band of depression region R2 in Snowy path is smaller than the value in the WET road or a dry road made with, as shown in the frequency spectrum of the region R5 after kicking 9, also decreases bandwidth value P ₅₁ is selected from 1kHz~4kHz band after region R5 kicking.
Therefore, a threshold value is set for the band value P ₂₁ , the band value P ₅₁ , or the sum P _{2151 of} the band value P ₂₁ and the band value P _51, and P ₂₁ , P ₅₁ , or the sum P ₂₁₅₁ is greater than the threshold value. Is smaller, it can be determined that the road surface is a snow covered road with fresh snow.
In this example, the relationship between the band value P ₂₁ and the band value P ₅₁ in various road surface conditions is experimentally obtained in advance, and the discrimination function F 2 = w ₂₁ · P ₂₁ + w ₂₂ · P ₅₁ −K 2 is set. calculated bandwidth value P ₂₁ and band value P ₅₁ and the identification function F2 function value f2 which is obtained by substituting the you are, depending on whether satisfies f2 <0, determine whether the road surface is snowy road did. Note that the coefficient w ₂₁ and the coefficient w ₂₂ of the discrimination function are about +1.

If f2 ≧ 0 in step S2, the process proceeds to step S3 to determine whether the inclusion on the road surface is water or snow, that is, whether the road surface is a deep WET road or a deep sherbet-like snow road. To do.
When the inclusion is a sherbet-like snow, the vibration component in the high-frequency region of the kicking vibration becomes larger than the case where the inclusion is water, while the road surface becomes slippery. The vibration component in the low region becomes large. Therefore, as shown in the frequency spectrum of the post-kick region R5 in FIG. 10, the value of the band value P ₅₂ selected from the 2 kHz to 4 kHz band of the post-kick region R5 is smaller in the sherbet-like snow road, and is 7 kHz. The value of the band value P ₅₃ selected from the -10 kHz band is larger on the sherbet-like snow road. On the other hand, as shown in the frequency spectrum of the trailing front region R3 of FIG. 11, the value of the bandwidth value P ₃₁ is selected from 7kHz~10kHz bandwidth before kicking region R3 is towards the sherbet-like snowy road becomes larger. Moreover, the value of the band value P ₄₁ selected from the 7 kHz to 10 kHz band of the kick-out area R4 is also increased.
Therefore, the inclusions are water or sherbet-like snow from the band values P ₅₂ , P ₃₁ , P ₄₁ , P ₅₃ , or the difference between the band values P ₅₂ and the band values P ₃₁ , P ₄₁ , P _53. It can be determined whether there is. When the number of parameters increases as described above, the discriminant function F′3 = w ′ ₃₁ · P ₅₂ + w is previously applied to the band values P ₅₂ , P ₃₁ , P ₄₁ , and P ₅₃ in various road surface conditions. _{'32 · P 31 + w'} 33 · P 41 + w ' set the ₃₄ · P ₅₃ -K'3 road surface it is preferable to determine whether the sherbet-like snowy road. That is, when the function value f′3 obtained by substituting the band values P ₅₂ , P ₃₁ , P ₄₁ , and P ₅₃ into the discrimination function F′3 is f′3 ≧ 0, it is a deep WET path. If f′3 <0, it is determined that the road surface is a sherbet-like snow road. As a result, the band values P ₅₂ , P ₃₁ , P ₄₁ , P ₅₃ , or the difference between the band value P ₅₂ and the band values P ₃₁ , P ₄₁ , P ₅₃ can be obtained with higher accuracy. it can.
Further, instead of the band values P ₃₁ , P ₄₁ , and P ₅₃ , the band value selected from the 7 kHz to 10 kHz band of the region R345 straddling the pre-kicking region R3, the kicking region R4, and the post-kicking region R5. seeking P _345, band value P ₃₄₅ and band value P ₅₂ previously obtained discriminant function and _{F3 = w 31 · P 52 +} w 32 · P 345 -K3 function value f3 obtained by substituting the can, f3 When <0, the traveling road surface may be determined to be a sherbet-like snow road.

When f1 <0 in step S1, the process proceeds to step S4 to determine whether or not the road surface is slippery. Since a slippery road surface other than a snowy road, a sherbet-like snow road, or a deep WET road is a frozen road, it is determined in step 4 whether or not the road surface is a frozen road.
When the road surface is slippery, in particular, as shown in the band value P ₄₂ selected from the 1 kHz to 4 kHz band of the kicking region R4 and the frequency spectrum of the post kicking region R5 in FIG. 12, the post kicking region R5 The band value P ₅₁ selected from the 1 kHz to 4 kHz band becomes smaller. Therefore, the bandwidth value P _42, P ₅₁ the previously obtained discriminant function _{F'4 = w '41 · P 42} + w' 42 · P 51 -K'4 to assign to the obtained function value f'4 However, when f′4 <0, it is determined that the traveling road surface is a frozen road, and when f′4 ≧ 0, the process proceeds to step S5.
Further, in place of the band values P ₄₂ and P ₅₁ , the discriminant function obtained in advance is a band value P ₄₅₀ selected from the 1 kHz to 4 kHz band of the region R450 straddling the kicking region R4 and the post-kicking region R5. F4 = w ₄₁ · P ₄₅₀ -K4 function value f4 obtained by substituting the can may be road traveling case where f4 <0 is determined to be a frozen road.

In step S5, it is determined whether or not there is a thin water film or shallow sherbet-like snow on the road surface. If the water film or sherbet-like snow layer is thin, it is determined in step S1 that there is no water film or sherbet-like snow. Using the calculated sound pressure level ratio Q = (P _B / P _A ), it is determined whether there is a thin water film or shallow sherbet-like snow on the road surface.
Specifically, the road surface temperature T data is compared with a preset reference temperature T _{0. If} the measured road surface temperature T is equal to or higher than the reference temperature T ₀ , the water on the road surface is in a liquid state. It is determined whether or not it can be taken. If the road surface temperature T is lower than the reference temperature T ₀ , the process immediately proceeds to step S7 without determining whether the water on the road surface can take a liquid state. In this example, the reference temperature T ₀ is T ₀ = −3 ° C.
Whether the water on the road surface can take a liquid state is determined using the sound pressure level ratio Q = (P _B / P _A ). A low frequency band (e.g., 500 Hz) band power value P _A in the water on the road surface regardless of whether a liquid state, change the speed. On the other hand, the band power value P _{B in} a high frequency band (for example, 9000 Hz) also changes depending on the speed, but increases when a sound of water splashing water is detected. Accordingly, when the calculated sound pressure level ratio Q = (P _B / P _A ) is 1 or more, the water on the road surface can be in a liquid state.
That is, when the road surface temperature T is lower than the reference temperature T ₀ or the sound pressure level ratio Q = (P _B / P _A ) is less than 1, inclusions (water or snow) are present on the road surface. If the road surface temperature T is equal to or higher than the reference temperature T ₀ and the sound pressure level ratio Q = (P _B / P _A ) is equal to or higher than 1, the inclusion on the road surface is determined. The process proceeds to step S6.
In step S6, it is determined whether the inclusion on the road surface is water or snow, that is, whether the road surface is a shallow WET road or a shallow sherbet-like snow road. This determination is the same as the determination in step S3, and the function value f′3 obtained by substituting the band values P ₅₂ , P ₃₁ , P ₄₁ , and P ₅₃ into the discrimination function F′3 is f′3 ≧ 0. Is determined to be a shallow WET road, and when f′3 <0, it is determined that the road surface is a shallow sherbet-like snow road. The determination may be made using the function value f3 obtained by substituting the band values P ₅₂ and P ₃₄₅ for the discriminant function F3.
Finally, in step S7 and step S8, it is determined whether the road surface is a snowy road or a dry road (DRY).
As shown in the frequency spectrum of the depression front region R1 in FIG. 13, the bandwidth value P ₁₃ is selected from the bandwidth value P ₁₁ and 1kHz or less of the frequency band selected from 2kHz~8kHz band of depression front region R1, the road surface When the roughness of is rough, it becomes large. That is, the band value P ₁₁ and the band value P ₁₃ are larger on the snowy road than on the smooth dry road. Although not shown, the band value P ₁₁ and the band value P ₁₃ are larger in the rough drying path than in the smooth drying path.
On the snowy road, the impact during stepping is alleviated by snow. Therefore, as shown in the frequency spectrum of stepping region R2 in FIG. 14 and the frequency spectrum of region R3 before kicking in FIG. The band value P ₂₂ selected from the 2 kHz to 4 kHz band and the band value P ₃₂ selected from the 2 kHz to 4 kHz band in the pre-kicking region R3 are smaller on the snow-capped road than on the dry road. However, on a snowy road, a minute slip in the tread surface due to snow occurs, so the band value P ₂₃ selected from the 4 kHz to 10 kHz band in the stepping area R2 and the 4 kHz to 10 kHz band in the pre-kicking area R3 are selected. the bandwidth value P _33, towards the snow-packed road is greater than a dry road.
Furthermore, since the shearing force at the time of kicking out decreases when slippery, the kicking region R4 as shown in the frequency spectrum of the kicking region R4 in FIG. 16 and the frequency spectrum of the post-kicking region R5 in FIG. Since the band value P ₄₂ selected from the 1 kHz to 4 kHz band and the band value P ₅₁ selected from the 1 kHz to 4 kHz band in the post-kick region R5 are reduced, the compressed snow road or the dry road is also reduced. Can be determined.

In step S7, the road surface during running smooth dry road and snow-packed road or rough dry road identification function for identifying the _{F'7 = w '71 · P 11} + w' 72 · P 13 + w '73 · P 22 + w' _{74 · P 23 + w '75} · P 32 + w' 76 · P 33 + w '77 · P 42 + w' 78 · P 51 -K'7 set the bandwidth value _{P 11, P 13, P 22} , P 23, If _{P 32, P 33, P 42} , P 51 the previously obtained obtained by substituting the discriminant function F'7 had been function value F'7 is a f'7 ≧ 0, the road surface during running Is determined to be a smooth drying path. On the other hand, if f′7 <0, it is determined that the road is a snowy road or a rough dry road, and the process proceeds to step S8.
Further, instead of the band values P ₄₂ and P ₅₁ , a band value P ₄₅₀ selected from the 1 kHz to 4 kHz band of the region R450 straddling the kicking region R4 and the post-kicking region R5 is obtained, and the band values P ₁₁ , P ₁₃ , P ₂₂ , P ₂₃ , P ₃₂ , P ₃₃ , P ₄₅₀ are obtained in advance. The discriminant function F7 = w ₇₁ · P ₁₁ + w ₇₂ · P ₁₃ + w ₇₃ · P ₂₂ + w ₇₄ · P ₂₃ + w ₇₅ · P ₃₂ + w ₇₆ · P ₃₃ + w ₇₇ · P _{450 When} the function value f7 obtained by substituting for K7 is f7 ≧ 0, it is estimated that the running road surface is a smooth dry road. Also good.
In step S8, the identification function that identifies the road surface or compacted snow road or rough dry road _{F8 = w 81 · P 11 +} w 82 · P 13 + w 83 · P 22 + w 84 · P 23 + w 85 · P 32 + w 86 · P 33 + W ₈₇ · P ₄₂ + w ₈₈ · P ₅₁ -K8 is set, and the discriminant function F8 obtained in advance for the band values P ₁₁ , P ₁₃ , P ₂₂ , P ₂₃ , P ₃₂ , P ₃₃ , P ₄₂ , P ₅₁ When the function value f8 obtained by substituting into is f8 ≧ 0, it is estimated that the running road surface is a rough dry road, and when f8 <0, the running road surface is a compressed snow road. Judge that there is.
The discriminant function F7 and discriminant function F8 are band values (P ₁₁ , P ₁₃ , P ₂₂ , P ₂₃ , P ₃₂₎ obtained by actually running the vehicle on a smooth dry road, a rough dry road, and a snowy road. , P ₃₃ , P ₄₂ , P ₅₁ ) calculated using a known method such as the least square method, Mahalanobis distance, or SVM from the set data, and the coefficient w _mn of the discriminant function F7 and the discriminant function F8 Of course, the coefficient w _mn is different.

As described above, in the present embodiment, the acceleration sensor 11, the road surface thermometer 21, and the microphone 22 detect the tire circumferential vibration, the road surface temperature T, and the tire generated sound of the tire 10, respectively. From the vibration data, the band values P ₁₁ , P ₁₂ , and P _{13 in the} pre-depression region R1, the band values P ₂₁ , P ₂₂ , and P _{23 in} the depressing region R2, and the band values P ₃₁ and P in the pre-kicking region R3 are obtained. _32, the P _33, and the calculated bandwidth value P _41, P ₄₂ of the kicking region R4, and a band value P _51, P _52, P ₅₃ after kicking area R5, a low frequency band from the data of the tire generated sound _A sound pressure level ratio Q = (P _A / P _B ), which is a ratio between the band power value P _A of the current band and the band power value P _B of the high frequency band, is calculated, the band value P _ij , the road surface temperature T data, Using the sound pressure level ratio Q and wheel speed data, the road Since so as to estimate a state, it is possible to estimate by subdividing the state of the snowy road, the state of the road surface can be accurately estimated.
In addition, since the discrimination function Fk for determination is obtained in advance and the road surface state is determined based on the sign of the function value fk obtained by substituting the band value P _ij for this Fk, the estimation accuracy of the road surface state Can be significantly increased.

In the embodiment described above, the band value P _ij is obtained by passing the time series waveform of each region R1 to R5 of the tire circumferential vibration through the band pass filter. However, the band value P _ij is obtained by performing FFT analysis on the time series waveform. The band value P _ij may be obtained from the obtained frequency spectrum.
In the above example, the band value P _ij and a band value P _ij of the tire circumferential direction vibration, change the detection direction of the acceleration sensor 11 in the tire width direction, obtains a bandwidth value P _'ij in the tire width direction vibration The road surface state may be estimated using the band value P ′ _ij . However, since the vibration in the tire width direction has a smaller amplitude than the vibration in the tire circumferential direction, it is preferable to use the vibration in the tire circumferential direction to increase the estimation accuracy as in this example.
In the above example, power is supplied from the power supply device 32 on the vehicle body side to the power supply device 16. However, power may be supplied to the power supply device 16 by power generation in the tire. As an apparatus for generating power in the tire, for example, a rotor that is magnetized to rotate by rolling of the tire 10, a stator made of a high permeability material adjacent to the rotor, and a magnetic circuit including the rotor and the stator are provided. And a power generation device including the generated power generation coil.

  In the above example, the road surface state is estimated by detecting the vibration of the tire 10, the road surface temperature T, and the tire generated sound. However, even if the road surface state is estimated using only the tire circumferential vibration data, the snow road Can be subdivided and estimated. In this case, a shallow sherbet-like snow road is determined to be a deep sherbet-like snow road, a dry road or a snowy road, and a shallow WET road is determined to be a deep WET road, a snowy road or a dry road, but “deep” If the setting value for distinguishing “shallow” is changed to create the discriminant function F1 and discriminant function F2 used for the inclusion determination in step S1, the estimation accuracy will not be reduced. Further, when such road surface determination is performed, steps S5 and S6 may be omitted in the flowchart shown in FIG.

It is also possible to perform estimation by subdividing only the snow road condition using the road surface condition estimation method according to the present invention.
That is, in order to estimate whether or not the road surface is a snowy road, the band values P ₁₁ and P ₁₂ of the tire circumferential vibration detected by the acceleration sensor 11 are used as the discriminant function F1 = w ₁₁ · P obtained in advance. ₁₁ + w and ₁₂ · P ₁₂ -K1 function values f1 obtained by substituting, the said zone values P _21, P ₅₁ discriminant function obtained in advance to _{F2 = w 21 · P 21 +} w 22 · P 51 -K2 It is determined whether or not the function value f2 obtained by substituting for f1 satisfies f1 ≧ 0 and f2 <0. If f1 ≧ 0 and f2 <0, the road surface being traveled is a snowy road What is necessary is just to determine that there exists. In the case of f1 <0 or f2 ≧ 0, the road surface is other than a snowy road, so that the following snow road condition estimation may be continued.
In order to estimate whether or not the road surface is a sherbet-like snow road, the discriminant function F′3 = w ′ ₃₁ · P in which the band values P ₅₂ , P ₃₁ , P ₄₁ , and P ₅₃ are obtained in advance. _{52 + w '32 · P 31} + w' 33 · P 41 + w '34 · P 53 -K'3 function value F'3 obtained by substituting the found decision whether to satisfy the F'3 <0 When f′3 <0, it may be determined that the traveling road surface is a sherbet-like snow road.

Also, to estimate whether the road surface is freezing path, bandwidth value P _42, discriminant function obtained in advance the _{P 51 F'4 = w '41 ·} P 42 + w' 42 · P 51 -K It is determined whether or not the function value f′4 obtained by substituting for “4” satisfies f′4 <0. If f′4 <0, the running road surface is a frozen road. Can be determined.
Further, the road surface is estimated whether the snow-packed road, bandwidth value _{P 11, P 13, P 22} , P 23, P 32, P 33, P 42, discriminant function obtained in advance the P ₅₁ F'7 = w _'71 , P ₁₁ + w' ₇₂ , P ₁₃ + w _'73 , P ₂₂ + w' ₇₄ , P ₂₃ + w _'75 , P ₃₂ + w' ₇₆ , P ₃₃ + w _'77 , P ₄₂ + w' ₇₈ , and P ₅₁ -K'7 to assign to the obtained function value F'7, our previously seeking bandwidth values _{P 11, P 13, P 22} , P 23, P 32, P 33, P 42, P 51 There identification function _{F8 = w 81 · P 11 +} w 82 · P 13 + w 83 · P 22 + w 84 · P 23 + w 85 · P 32 + w 86 · P 33 + w 87 · P 42 + w substituted in ₈₈ · P ₅₁ -K8 It is determined whether or not the function value f8 obtained in this way satisfies f′7 <0 and f8 <0. If f′7 <0 and f8 <0, the road surface being traveled is a compressed snow road. What is necessary is just to determine that it is.
[Example]

表１からわかるように、本発明による路面推定方法を用いることにより、雪路の状態を細分化して推定できることが確認された。シャーベット状の雪路と凍結路とでは正答率が低いが、これは、なお、基準となる路面状態が車内の目視によるため、圧雪路上の浅いシャーベット状の雪と凍結状態との判別がつけにくいことによるものと考えられる。 An infrared temperature sensor is attached to the bumper of the four-wheel drive vehicle that is the test vehicle, an acceleration sensor is attached to the inner liner of the left front wheel, and a microphone is attached to the lower part of the vehicle in front of the left rear wheel. The road surface condition was estimated by driving on a general road.
Table 1 shows the correct answer rates of the estimation results when the road surface condition visually determined in the running vehicle is “correct”. The tire size is 265 / 65R17, and the running speed is 60 km / h.

As can be seen from Table 1, it was confirmed that by using the road surface estimation method according to the present invention, snow road conditions can be subdivided and estimated. The correct answer rate is low between the sherbet-like snowy road and the frozen road, but this is because it is difficult to distinguish between the shallow sherbet-like snow on the snowy road and the frozen state because the standard road surface condition is visually observed inside the vehicle. This is probably due to this.

  As mentioned above, although this invention was demonstrated using embodiment, the technical scope of this invention is not limited to the range as described in the said embodiment. It will be apparent to those skilled in the art that various modifications or improvements can be added to the embodiment. It is apparent from the claims that the embodiments added with such changes or improvements can be included in the technical scope of the present invention.

  As described above, according to the present invention, it is possible to accurately estimate the state of the road surface during traveling and to subdivide the snow road, so that the driver can recognize the estimated road surface state. If the running state of the vehicle is controlled based on the estimated road surface state, the running safety of the vehicle can be improved.

1 road surface condition estimation system, 10 tires, 11 acceleration sensor, 12 amplifier,
13 A / D converter, 14 tire side transmitting device, 15 power receiving device,
16 power supply device, 20 car body, 20M in the car, 21 road surface thermometer,
22 microphone, 23 wheel speed sensor, 24 acceleration sensor for monitoring,
25 cameras, 26 GPS, 27 arithmetic units, 28 road surface information recording means,
29 monitor, 30 receiver, 31 vehicle-side transmitter, 32 power feeder,
33 alarm device, 40 management center, 41 data server, 42 display device.

Claims (9)

A time series waveform of the vibration of the running tire is expressed by a step region R1 before the step side peak appearing at the step end, a step region R2 forming the step side peak, and the step side peak. A kick-out region R3 between the kick-out peak appearing at the kick-out end, a kick-out region R4 forming the kick-out peak, and a kick-out after the kick-out peak A step (a) for dividing the region R5;
For each of the regions R1 to R5, a step (b) for obtaining a band value that is a magnitude of a vibration component in a predetermined specific frequency band;
On the basis of the bandwidth value, it possesses a step (c) for estimating the road surface condition during travel,
In the step (c), the running road surface state is a snowy road using the band value of the pre-depression region R1, the band value of the depressing region R2, and the band value of the post-kicking region R5. A road surface state estimating method characterized by determining whether or not . A time series waveform of the vibration of the running tire is expressed by a step region R1 before the step side peak appearing at the step end, a step region R2 forming the step side peak, and the step side peak. A kick-out region R3 between the kick-out peak appearing at the kick-out end, a kick-out region R4 forming the kick-out peak, and a kick-out after the kick-out peak A step (a) for dividing the region R5;
For each of the regions R1 to R5, a step (b) for obtaining a band value that is a magnitude of a vibration component in a predetermined specific frequency band;
(C) estimating a road surface condition during traveling based on the band value;
In the step (c), the band value of the pre-kicking region R3, the band value of the kicking region R4, and the band value of the post-kicking region R5 are used, or the post-kicking region R5 and bandwidth value, the front region R3 out the kick using the bandwidth value of the region spanning the the region R5 after kicking the said trailing region R4, the road surface condition during travel is a sherbet-like snowy road road surface condition estimating how to and judging whether. A time series waveform of the vibration of the running tire is expressed by a step region R1 before the step side peak appearing at the step end, a step region R2 forming the step side peak, and the step side peak. A kick-out region R3 between the kick-out peak appearing at the kick-out end, a kick-out region R4 forming the kick-out peak, and a kick-out after the kick-out peak A step (a) for dividing the region R5;
For each of the regions R1 to R5, a step (b) for obtaining a band value that is a magnitude of a vibration component in a predetermined specific frequency band;
(C) estimating a road surface condition during traveling based on the band value;
In the step (c), a band value of a region straddling the kicking region R4 and the post-kicking region R5 or a band value of the kicking region R4 and a band value of the post-kicking region R5 is used. Te, road surface condition estimating how to characterized in that the road surface condition during travel is determined whether frozen road. A time series waveform of the vibration of the running tire is expressed by a step region R1 before the step side peak appearing at the step end, a step region R2 forming the step side peak, and the step side peak. A kick-out region R3 between the kick-out peak appearing at the kick-out end, a kick-out region R4 forming the kick-out peak, and a kick-out after the kick-out peak A step (a) for dividing the region R5;
For each of the regions R1 to R5, a step (b) for obtaining a band value that is a magnitude of a vibration component in a predetermined specific frequency band;
(C) estimating a road surface condition during traveling based on the band value;
In the step (c), the band value of the pre-depressing region R1, the band value of the depressing region R2, the band value of the pre-kicking region R3, the band value of the kicking region R4, and the kicking-out by using the band value of the rear region R5, the road surface condition during travel is you and judging whether snow-packed road road surface state estimating method. A time series waveform of the vibration of the running tire is expressed by a step region R1 before the step side peak appearing at the step end, a step region R2 forming the step side peak, and the step side peak. A kick-out region R3 between the kick-out peak appearing at the kick-out end, a kick-out region R4 forming the kick-out peak, and a kick-out after the kick-out peak A step (a) for dividing the region R5;
For each of the regions R1 to R5, a step (b) for obtaining a band value that is a magnitude of a vibration component in a predetermined specific frequency band;
A step (c) of estimating a road surface condition during traveling based on the band value;
Provided between said step (b) and said step (c), and step (d) for detecting the road surface temperature and the tire noise generated during running, bandwidth value P _A of the low frequency band of the tire generated sound whether and determining the bandwidth value P _B of the high frequency band (e), and a bandwidth value P _B of the band value P _a and the high frequency band of the said surface temperature low frequency band, there is inclusion on the road surface the and a step (f) determining whether,
If it is determined in step (f) that there are inclusions on the road surface,
The band value of the pre-kick region R3, the band value of the kick region R4, and the band value of the post-kick region R5 are used, or the band value of the post-kick region R5 and the pre-kick region It is determined whether the running road surface state is a shallow sorbet-like snow road or a shallow wet road surface by using the band value of the region extending over the region R3, the kicking region R4, and the post-kicking region R5. road surface condition estimating how to said.   A time series waveform of the vibration of the running tire is expressed by a step region R1 before the step side peak appearing at the step end, a step region R2 forming the step side peak, and the step side peak. A kick-out region R3 between the kick-out peak appearing at the kick-out end, a kick-out region R4 forming the kick-out peak, and a kick-out after the kick-out peak A step (a) for dividing the region R5;
For each of the regions R1 to R5, a step (b) for obtaining a band value that is a magnitude of a vibration component in a predetermined specific frequency band;
A step (c) of estimating a road surface condition during traveling based on the band value;
A step (d) of detecting a road surface temperature and a tire-generated sound during traveling provided between the step (b) and the step (c); and a band value P of a low-frequency band of the tire-generated sound. _A And the band value P of the high frequency band _B Step (e), and the road surface temperature and the band value P of the low frequency band _A And the band value P of the high frequency band _B And (f) determining whether there is an inclusion on the road surface,
If it is determined in step (f) that there are no inclusions on the road surface,
The band value of the pre-depressing region R1, the band value of the depressing region R2, the band value of the pre-kicking region R3, the band value of the kicking R4, and the band value of the post-kicking region R5 Used or straddles the band value of the pre-depression region R1, the band value of the depressing region R2, the band value of the pre-kick region R3, the kick region R4 and the post-kick region R5 A road surface state estimation method that determines whether or not a road surface state during traveling is a smooth dry road using a band value of a region.   The road surface state estimation method according to any one of claims 1 to 6, wherein the time series waveform of the tire vibration is a time series waveform of a tire vibration detected by an acceleration sensor installed in the tire. .   The road surface state estimation method according to any one of claims 1 to 7, wherein the tire vibration is vibration in a tire circumferential direction or vibration in a tire width direction of a running tire.   In the step (c), the vehicle travels based on the function value obtained by substituting the band value obtained in the step (b) into the discriminant function representing the relationship between the road surface condition and the band value obtained in advance. The road surface state estimation method according to any one of claims 1 to 8, wherein an inside road surface state is determined.

Images