JP5241556B2 - Road surface condition estimation device - Google Patents

Description

  The present invention is provided on the inner side of the tire, a tire internal sound measuring means for measuring the sound in the tire, an arithmetic means for performing an operation for estimating a road surface state based on a sound measurement signal from the tire internal sound measuring means, It is related with the road surface state estimation apparatus provided with.

  It is extremely meaningful to detect in real time the condition of the road surface on which the traveling vehicle is in contact with the ground. By conveying this information to the driver, safety during driving is improved, and by knowing the road surface condition. It is possible to produce an effect that the antilock brake system can be effectively operated. As a method of detecting the road surface condition, a method based on a friction coefficient obtained by measuring a slip ratio has been proposed, but the obtained friction coefficient is extremely low reliability especially when the slip ratio is small. As a countermeasure, the sound in the tire is measured by using the fact that the sound in the tire changes depending on the road surface. A method for estimating the road surface state by comparing with a reference sound corresponding to the road surface state (more specifically, comparing the intensity of sound in a predetermined frequency region between the measurement sound and the reference sound) and determining. It has been proposed (see, for example, Patent Document 1).

Japanese Utility Model Publication No. 6-174543

  However, in the road surface state estimation method disclosed in Patent Document 1, first, the sound should change greatly depending on the traveling speed of the vehicle. Compared, secondly, as a comparison method, they are limited to one frequency region and compared, and even if the set frequency region shows the same level of strength, another frequency region However, the intensity may be completely different. As a result, this estimation method could hardly be practical in terms of the estimation accuracy.

  The present invention has been made in view of such problems, and a road surface capable of estimating a road surface state with high accuracy on the basis of a sound measurement signal in a tire sound measuring means for measuring sound in the tire. An object is to provide a state estimation device.

<1> is based on a vehicle speed measuring means for measuring the vehicle speed, a tire internal sound measuring means for measuring a sound in the tire provided inside the tire, and a sound measurement signal from the tire internal sound measuring means. In a road surface state estimation device comprising: a calculation means for performing a calculation for estimating a road surface state;
Predetermined from each of the sound measurement signals obtained by running the vehicle under I x J conditions combining the predetermined I road surface conditions and the predetermined J vehicle speed regions. I x J x K reference characteristic values P0 _ijk (i = 1, 2,..., I, j = 1, 2) obtained as intensities or physical quantities P correlated to the K frequency bands ,..., J, k = 1, 2,...
The calculation means inputs a sound measurement signal from the tire internal sound detection means and a vehicle speed signal from the vehicle speed measurement means in real time while the vehicle is running, and when the sound measurement signal is input based on the vehicle speed signal. The vehicle speed region V _n is estimated from the J vehicle speed regions, and the intensity of each of the K frequency bands F _k from the sound measurement signal or the physical quantity P correlated therewith is input to the vehicle when the sound measurement signal is input. calculated as the measured characteristic values P1 _nk corresponding to the velocity region V _n, these and the K measured characteristic values P1 _nk, the K reference characteristics corresponding to each of the I-number of road conditions belonging to the vehicle speed range V _n By comparing the values P0 _ink (i = 1, 2,..., I, k = 1, 2,..., K), the road surface traveling at the time of the sound measurement signal measurement Estimate the road surface condition ,
Upon comparing the with the K measured characteristic values P1 _nk, and the K reference characteristic value P0 _ink corresponding to each of the I-number of road conditions belonging to the vehicle speed range V _n, the K measured characteristic values P1 _{nk Is} represented by a point Q _n (P1 _n1 , P1 _n2 ,..., P1 _nK ) in the K-dimensional space , belongs to the vehicle speed region V _n, and K reference characteristic values P0 corresponding to the i-th road surface state _in is represented by a point Q _in (P0 _in1 , P0 _in2 ,..., P0 _inK ) _in the K-dimensional space, and the distance D _i between the point Q _n and the I points Q _in is expressed by Equation (1). A road surface state estimating device that estimates the road surface state corresponding to the closest distance among these I distances as the road surface state of the road surface that was traveling at the time of the sound measurement signal measurement It is.

According to <1>, the measured characteristic values P1 _nk, is compared to a reference characteristic value P0 _ink vehicle speed range V _n Field of the vehicle speed during the sound measurement, i.e., on the combined identical vehicle speed Therefore, the measured sound and the reference sound are compared, and in the comparison between the measured characteristic value P1 _nk and the reference characteristic value P0 _ink , a comprehensive comparison is made in a plurality of frequency regions. The road surface state can be estimated with high accuracy.

Also, according to < 1 >, in the comparison between the measurement characteristic value P1 _nk and the reference characteristic value P0 _ink , the distance between these points when they are made to correspond to the points in the K-dimensional space Since it is determined which road surface condition is closest to the reference characteristic value, the road surface state can be estimated with extremely high accuracy.

It is a block diagram which shows the structure of the road surface state estimation apparatus of embodiment which concerns on this invention. It is a conceptual diagram which shows one process of a calculating means. . It is a conceptual diagram which shows the arithmetic processing at the time of setting a reference | standard characteristic value. and the point Q _n in n-dimensional space is a conceptual diagram representing the distance between the point Q _in.

  Embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a road surface state estimating device according to an embodiment of the present invention. The road surface state estimating device 10 includes a vehicle speed measuring unit 1 that measures a vehicle speed, a tire internal sound measuring unit 2 that is provided inside the tire 20 and measures a sound in the tire 20, and a tire internal sound measuring unit 2. The calculation means 3 which performs the calculation for estimating a road surface state based on this sound measurement signal, and the memory | storage part 4 which stores the data of a predetermined value are provided.

  The vehicle measuring means 1 calculates the average speed of the vehicle based on the wheel speeds obtained from the well-known wheel speed sensors 11a and 11b provided on each wheel and the wheel speed sensors 11a and 11b, for example. The vehicle speed calculation unit 12 can generate a speed signal. The vehicle speed calculation unit 12 is provided in the control unit 5 that controls the entire vehicle together with the calculation unit 3 and the storage unit 4, for example. A vehicle speed signal from the vehicle speed calculation unit 12 is input to the calculation means 3 in real time at a predetermined sampling time.

  Further, the tire internal sound measuring means 2 can be constituted by, for example, a microphone module with a transmitting antenna attached to the inner surface of the tire 20 or attached to the rim. A sound measurement signal is transmitted from the measuring means 2 at a predetermined sampling time, and this signal is received by the receiving antenna 13 provided on the vehicle body side near the tire and input to the computing means 3.

  In the present invention, a characteristic value is calculated from the sound measurement signal measured by the tire internal sound measuring means 2, and the measured characteristic value is collated with a reference characteristic value that is previously collected and stored for each road surface state, This is one of the methods for estimating the road surface state corresponding to the reference characteristic value closest to the value as the road surface state corresponding to the measured estimated value. It is a point to compare against the reference characteristic value collected under speed conditions, and this makes it possible to compare the road surface condition with high reliability compared with the case of comparing with the reference characteristic value for which the vehicle speed condition is not estimated. Secondly, as a characteristic value when comparing the measured characteristic value and the reference characteristic value, a plurality of frequency components are extracted from the sound measurement signal, and those based on each component are used. Is that this And makes it possible to estimate the road surface condition with a higher reliability for the case of comparing the characteristic value with each other based on a single frequency component.

The processing of the calculation means 3 will be described more specifically. The calculation means 3 performs the process of estimating the road surface state of the sound measurement signal X _n (m) that is continuously input at a predetermined sampling time over a period T, and FIG. It is a conceptual diagram which shows 1 process. As the period T, for example, one rotation of the tire is taken, and m (m = 1, 2, 3,...) Is a sampling point within this period.

  As described above, one of the features of the present invention is that it is compared with the reference characteristic value collected under the condition of the vehicle speed that is substantially the same as the measured characteristic value. It is necessary to know the speed of the vehicle in the period T corresponding to this process for each road surface state estimation process. The vehicle speed signal at one or more sampling points in the period T is input, and based on this signal, A process of estimating which of the J vehicle speed regions Vj (j = 1, 2, 3,..., J) that the vehicle speed at the sampling point belongs to is performed.

Here, as the J vehicle speed range, for example, the first speed range is less than 10 km / h, the second speed range is 10 km / h or more and less than 20 km / h, ... The area can be set to be (J-1) * 10 km / h or more and less than J * 10 km / h. In the following description, the vehicle speed region corresponding to the processing for estimating the road surface condition is assumed to be V _n. (The above description for estimating the vehicle speed region V _n is not shown in FIG. 2)

As shown in FIG. 2, the input sound measurement signal X _n (m) is a predetermined number of K band filters f _k (z ⁻¹ ) (k = 1, 2, 3,..., K ), Each frequency band component Y _nk (m) is extracted. Here, the suffix n means that the vehicle speed region corresponding to the time when the sound measurement signal is sampled is V _n , and the suffix k is a code for distinguishing the frequency band. Examples of the band filter f _k (z ⁻¹ ) include the following frequency band F _k with K = 4.
Frequency band F _{1 of} band filter f ₁ (z ^-1 ): center frequency 31.5 Hz, 1/3 octave width,
Frequency band F _{2 of} band filter f ₂ (z ⁻¹ ): center frequency 125 Hz, 1/3 octave width,
Frequency band F _{3 of} band filter f ₃ (z ^-1 ): center frequency 250 Hz, 1/3 octave width,
Frequency band F _{4 of} band filter f ₄ (z ^-1 ): Center frequency 3150Hz, 1/3 octave width

Each frequency band component Y _nk (m) of the sound measurement signal X _n (m) is averaged in order to suppress variations due to minute changes in time, tire position, and road surface. As the averaging process, for example, as shown in equations (2), (3), and (4), it is preferable to perform a square average over a period T and further perform a moving average.

Here, the averaging is preferably performed for the period T corresponding to 5 to 10 rotations of the tire. If the period T is smaller than the period corresponding to the 5 rotations of the tire, Y _nk (m) within the period T It is difficult to sufficiently suppress variations due to sampling points. On the other hand, if the period T is longer than the length corresponding to 10 revolutions, it becomes difficult to estimate when the road surface changes. Further, when the period T is, for example, 10 laps of the tire, the period T corresponding to 10 laps of the tire varies depending on the rotation speed of the tire, but the wheel rotation speed sensor 11a of the vehicle speed measuring means 1 provided separately is provided. By correcting using information, this period T can be set almost accurately for each process.

Zave _nk (T) calculated by the above averaging process is the K measurement characteristic values P1 _nk corresponding to the vehicle speed region V _n described above, and the calculation means 3 is as shown in FIG. to, after the averaging processing, these measurements characteristic values P1 _nk, preset, K pieces of the reference characteristic value P0 _ink a corresponding I-number of road conditions belonging to the vehicle speed range V _n (i = 1, 2,..., I, k = 1, 2,.

Reference characteristic values P0 _ijk (i = 1, 2,..., I, j = 1, 2,..., J, k = 1, 2,..., K) are shown in FIG. It can be represented by Zave _ijk (T) obtained by performing the calculation as shown. That is, in order to obtain the reference characteristic value P0 _ijk , among the predetermined I road surface conditions, the road surface of the i-th road surface state is set so that the speed becomes the vehicle speed region Vj, the vehicle is run, At that time, the K band filters f _k (z ⁻¹ ) (k = 1, 2, 3,..., K from the sound measurement signal X _ij (m) from the tire internal sound measurement means 2. ) To extract each frequency band component Y _ijk (m).

Next, each frequency band component Y _ijk (m) is averaged according to, for example, equations (5), (6), and (7). The averaging process can be performed in the same manner as described above with respect to equations (2), (3), and (4).

Now, in comparison operation processing shown in FIG. 2, and the K measured characteristic values P1 _nk (i.e. Zave _{nk (T)),,} the K reference characteristic value P0 _ink corresponding to each of the I sets of road conditions ( In other words, the process is compared with Zave _ink (T)), but if the i-th road surface condition, the actual measurement conditions are perfectly the same as the conditions when the reference characteristic value P0 _ink was obtained. If this is the case, for this i, the K measured characteristic values P1 _nk should all be the same as the corresponding K reference characteristic values P0 _ink . However, this is not the case because the conditions are slightly different in practice. In such a case, it is reasonable to estimate i that is closest to the K measured characteristic values P1 _nk as a whole when compared to the corresponding K reference characteristic values P0 _ink as the current road surface condition. is there.

The K measurement characteristic values P1 _nk in the K-dimensional space are expressed as the total perspective of the K measurement characteristic values P1 _nk and the corresponding K reference characteristic values P0 _ink . This is represented by a point Q _n (P1 _n1 , P1 _n2 ,..., P1 _nK ), and K reference characteristic values P0 _nk corresponding to the i-th road surface state are represented by points Q _in (P0 _in1 , P0 _in2 ,..., P0 _inK ), using the distance between the point Q _n and the point Q _in and the road surface corresponding to the point Q _in that is the closest to the point Q _n The state i is estimated as the road surface currently running. That is, the road surface state i corresponding to i having the smallest distance represented by the above-described equation (1) is estimated as the currently traveling road surface.

Note that the distance D _i represented by the equation (1) represents the total perspective of the K measurement characteristic values P1 _nk and the corresponding K reference characteristic values P0 _ink. Is apparent from FIG. 4 showing a point Q _n and, for example, two points Q _in (i = 1, 2) in a three-dimensional space with K = 3.

However, the present invention is, as an indication of the distance degree, the distance is not limited to D _i, for example, such as represented by the formula (8), and the K measured characteristic values P1 _nk, corresponding reference it is also possible to use the total C _i of the absolute values of respective differences between the characteristic values P0 _ink.

  A vehicle equipped with a road surface estimation device according to the above description is run at various speeds on road surfaces with different road surface conditions, the road surface is estimated, the road surface state estimated by the road surface estimation device, and the actual road surface state And the accuracy rate for each speed was calculated and summarized in Table 1.

The road surface I drove
1) Paved asphalt road surface 2) Cracked asphalt road surface 3) Cobblestone road surface 4) Low friction coefficient road surface. The speed was changed at seven levels of 20km / h, 30km / h, 40km / h, 50km / h, 60km / h, 70km / h and 80km / h.

As the band filter, f ₁ (z ⁻¹ ), f ₂ (z ⁻¹ ), f ₃ (z ⁻¹ ), and f ₄ (z ⁻¹ ) exemplified above were used. The frequency band of each band filter is as follows.
Frequency band F _{1 of} band filter f ₁ (z ⁻¹ ): center frequency 31.5 Hz, 1/3 octave,
Frequency band F _{2 of} band filter f ₂ (z ^-1 ): center frequency 125 Hz, 1/3 octave,
Frequency band F _{3 of} band filter f ₃ (z ^-1 ): center frequency 250 Hz, 1/3 octave,
Frequency band F _{4 of} band filter f ₄ (z ^-1 ): Center frequency 3150Hz, 1/3 octave

  In the averaging process, the equations (2) to (4) were followed. In obtaining the reference characteristic value, the equations (5) to (7) were followed. Further, the formula (1) was used for the determination. In calculating the correct answer rate, the correct answer rate was determined for 50 times for each condition.

DESCRIPTION OF SYMBOLS 1 Vehicle speed measuring means 2 Tire internal sound measuring means 3 Calculation means 4 Memory | storage part 5 Control part 10 Road surface state estimation apparatus 11a, 11b Wheel rotational speed sensor 12 Vehicle speed calculating part 13 Receiving antenna 20 Tire

Claims (1)

A vehicle speed measuring means for measuring the vehicle speed, a tire internal sound measuring means for measuring a sound in the tire provided inside the tire, and a road surface state is estimated based on a sound measurement signal from the tire internal sound measuring means. In a road surface state estimating device comprising a calculation means for performing calculation,
Predetermined from each of the sound measurement signals obtained by running the vehicle under I x J conditions combining the predetermined I road surface conditions and the predetermined J vehicle speed regions. I x J x K reference characteristic values P0 _ijk (i = 1, 2,..., I, j = 1, 2) obtained as intensities or physical quantities P correlated to the K frequency bands ,..., J, k = 1, 2,...
The calculation means inputs a sound measurement signal from the tire internal sound detection means and a vehicle speed signal from the vehicle speed measurement means in real time while the vehicle is running, and when the sound measurement signal is input based on the vehicle speed signal. The vehicle speed region V _n is estimated from the J vehicle speed regions, and the intensity of each of the K frequency bands F _k from the sound measurement signal or the physical quantity P correlated therewith is input to the vehicle when the sound measurement signal is input. calculated as the measured characteristic values P1 _nk corresponding to the velocity region V _n, these and the K measured characteristic values P1 _nk, the K reference characteristics corresponding to each of the I-number of road conditions belonging to the vehicle speed range V _n By comparing the values P0 _ink (i = 1, 2,..., I, k = 1, 2,..., K), the road surface traveling at the time of the sound measurement signal measurement Estimate the road surface condition ,
Upon comparing the with the K measured characteristic values P1 _nk, and the K reference characteristic value P0 _ink corresponding to each of the I-number of road conditions belonging to the vehicle speed range V _n, the K measured characteristic values P1 _{nk Is} represented by a point Q _n (P1 _n1 , P1 _n2 ,..., P1 _nK ) in the K-dimensional space , belongs to the vehicle speed region V _n, and K reference characteristic values P0 corresponding to the i-th road surface state _in is represented by a point Q _in (P0 _in1 , P0 _in2 ,..., P0 _inK ) _in the K-dimensional space, and the distance D _i between the point Q _n and the I points Q _in is expressed by Equation (1). A road surface state estimating device that estimates the road surface state corresponding to the closest distance among these I distances as the road surface state of the road surface that was running at the time of the sound measurement signal measurement .

Images