JP3428620B2 - Road surface friction coefficient determination device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a judgement apparatus of a road surface friction coefficient that can accurately judge a road surface friction coefficient even if case of the road surface canted in lateral direction has initial displacement in sensor value of a lateral-G sensor. SOLUTION: Differential values dGyc , dGy of calculated lateral acceleration and actual lateral acceleration are respectively determined (a differential processing block 12) in sensor signal processing in a ECU 8, and a road surface μis adjusted from a μ-judgement map judgement block 26) based upon a deviation-effective value dGyer , of between the differential values dGyc , dGy . The road surface μ is judged from magnitude of a value normalizing the deviation-effective value dGyer .

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a road surface friction coefficient determination device for determining whether a road surface on which a vehicle travels has a high friction coefficient or a low friction coefficient.

[0002]

Related Background Art As a road surface friction coefficient determining device of this type, for example, a road surface friction coefficient detecting device disclosed in Japanese Utility Model Publication No. 6-18276 is known. This known friction coefficient detection device determines the road surface friction coefficient by comparing the calculated lateral acceleration calculated based on the vehicle speed and the steering angle with the detected actual lateral acceleration when the vehicle turns. Specifically, if the actual lateral acceleration is substantially equal to the calculated lateral acceleration and the ratio of these is 1 or more, it is determined that the road surface is a high friction coefficient road that is relatively unlikely to skid. If the acceleration is smaller than the calculated lateral acceleration and the ratio of these is smaller than 1, it is determined that the road surface is a low friction coefficient road that is likely to skid.

[0003]

In the known friction coefficient detecting device described above, attention is paid to the fact that when the vehicle actually skids, the actual lateral acceleration detected becomes smaller, and the actual lateral acceleration and the calculated lateral acceleration are noted. It is considered that whether the friction coefficient of the road surface is high or low is determined by comparing the magnitude with the acceleration.

However, when the vehicle travels on a road surface (bank road) inclined in the lateral direction, for example, lateral acceleration (sin θ of gravitational acceleration) corresponding to the inclination angle θ is generated in the vehicle regardless of steering. Therefore, the initial value is given to the actual lateral acceleration accordingly. In this case, the detected actual lateral acceleration shows a value that is larger by the initial value on the one hand, and conversely shows a value that is smaller by the initial value on the other hand. In such a situation, in the above-described friction coefficient detection device, it is not possible to accurately determine the road surface friction coefficient by simply comparing the detected actual lateral acceleration and the calculated lateral acceleration. There is a risk of doing it.

The present invention has been made under the circumstances described above, and an object of the present invention is to provide a road surface friction coefficient judging device capable of accurately judging the friction coefficient regardless of the inclination of the road surface. is there.

[0006]

In order to achieve the above-mentioned object, a road surface friction coefficient determining device according to a first aspect of the present invention provides a differential value of a calculated lateral acceleration to be generated in a vehicle body and a detected actual lateral acceleration.
The output phase difference is obtained from the deviation from the differential value, and the determination means determines the road surface friction coefficient based on the output phase difference. According to the road surface friction coefficient determination device of claim 2,
Means for determining the differential value of the calculated lateral acceleration and
Output phase difference by the effective value of the deviation from the differential value of the actual lateral acceleration
And determine the road friction coefficient based on this output phase difference
To do. Furthermore, in the road surface friction coefficient determination device of claim 3,
The output phase difference has been subjected to normalization processing.
In the road surface friction coefficient determining device according to claim 4,
The effective value of the differential value of the actual lateral acceleration is higher than the upper limit value.
When it is large, or the effective value of the differential value of the actual lateral acceleration is
If it is less than the lower limit, the above judgment shall be prohibited.
ing.

That is, according to the road surface friction coefficient determining apparatus of claim 1, when the output phase difference between the calculated lateral acceleration and the actual lateral acceleration is relatively small, it is determined that the road surface friction coefficient is high. On the other hand, when the output phase difference is relatively large, it is determined that the road surface friction coefficient is low. In other words, the road surface having a high road friction coefficient (high μ road) and the road surface having a low road friction coefficient (low μ road) have different temporal output follow-up characteristics of the actual lateral acceleration with respect to the calculated lateral acceleration. Since μ is determined, it is not necessary to consider the influence of the initial value of the actual lateral acceleration described above. In addition, the road surface friction coefficient judgment of claim 2
According to the measuring device, the magnitude of the output phase difference is quantitatively obtained.
The determination accuracy is improved, and the road surface friction coefficient determination according to claim 3
The device suppresses the effects of sudden changes in steering,
According to the road surface friction coefficient determination device of claim 4, erroneous determination is made.
The possibility of
improves.

[0008]

BEST MODE FOR CARRYING OUT THE INVENTION Referring to FIG. 1, there is shown a configuration of a road surface friction coefficient determining device according to an embodiment. The road surface friction coefficient determination device of this embodiment is applied to a vehicle (not shown). Therefore, the vehicle detects a vehicle speed sensor 2 for detecting a vehicle speed, an operation angle of a steering wheel, that is, a steering angle of the vehicle. The steering wheel angle sensor 4 and the lateral G sensor 6 for detecting the actual lateral acceleration are appropriately arranged. The vehicle speed sensor 2 may be a wheel speed sensor provided on each wheel of the vehicle, and in this case, the vehicle speed V can be obtained from the wheel speed.

Sensor signals from these various sensors are sent to an electronic control unit (EC) arranged at an appropriate position on the vehicle.
U) 8 and the processing of the sensor signal described below is performed in this ECU 8. First, the calculation block 10 in the ECU calculates the calculated lateral acceleration Gyc based on the vehicle speed signal V from the vehicle speed sensor 2 and the steering wheel operation angle signal ha from the steering wheel angle sensor 4. This calculated lateral acceleration Gyc is supplied to the next differential processing block 12, and this differential processing block 1
When it is time-differentiated by 2, the differential value of the calculated lateral acceleration dGyc
Is output as. On the other hand, the calculated lateral acceleration Gyc is also supplied to the absolute value processing filter 14, and the absolute value | Gyc | of the calculated lateral acceleration is output from the absolute value processing filter 14. In the differential processing block 12, the actual lateral acceleration signal Gy from the lateral G sensor 6 is time-differentiated, and then the differential value dGy is output.

The differential values dGyc and dGy output from the differential processing block 12 are supplied to the next averaging processing filter 16. The averaging filter 16 takes a moving average of the differential value dGyc of the calculated lateral acceleration and the differential value dGy of the actual lateral acceleration within a predetermined time. In addition, by performing moving averaging processing on the differential value dGy of the actual lateral acceleration,
For example, even if the differential value dGy of the actual lateral acceleration fluctuates due to a slight lateral vibration of the running vehicle, the variation of the differential value dGy due to the fluctuation is smoothed in the subsequent signal processing. Regarding the differential value dGyc of the calculated lateral acceleration, it is not necessary to consider the variation due to the vehicle body vibration, but the same averaging processing filter 16 is used to match the output phase characteristic with the differential value dGy of the moving averaged actual lateral acceleration. To filter.

The differential values dGyc and dGy which have passed through the averaging processing filter 16 and have been moving averaged are both added and subtracted by the addition / subtraction unit 1.
8 are supplied. The addition / subtraction unit 18 takes the deviation between the differential value dGyc of the calculated lateral acceleration and the differential value dGy of the actual lateral acceleration, and the addition / subtraction unit 18 outputs the deviation dGye between them. On the other hand, the moving averaged differential value dGy of the actual lateral acceleration is also directly supplied to the next averaging processing filter 20. Further, as shown in the figure, the deviation dGye output from the addition / subtraction unit 18 is also supplied to the averaging processing filter 20. Both the deviation dGye and the differential value dGy of the actual lateral acceleration are averaged by the averaging filter 2
It is supplied to the operation block 22 via 0. That is,
The outputs of the deviation dGye and the differential value dGy are adjusted by the averaging processing filter 20 to the average value for each predetermined time, and then the next calculation block 22 calculates the effective value per unit time from each average value. It is calculated as (rms value). Therefore, from the calculation block 22, the effective value d of the deviation between the differential value of the calculated lateral acceleration and the differential value of the actual lateral acceleration.
The effective value dGyr of the differential value of Gyer and the actual lateral acceleration is output.

The effective value dGyer of these deviations and the effective value dGyr of the differential value of the actual lateral acceleration are supplied to the next normalization processing block 24 via the main signal processing route shown in FIG. On the other hand, the effective value dG of the differential value of the actual lateral acceleration
yr is also branched and supplied to a signal route for forming a determination prohibition signal in the OR circuit 30 described later. Further, the absolute value | Gyc | of the calculated lateral acceleration output from the absolute value filter 14 is supplied to the determination block 26.

Before explaining the normalization processing in the next normalization processing block 24 and the judgment processing in the judgment block 26,
The determination of the road surface friction coefficient based on the output phase difference between the calculated lateral acceleration and the actual lateral acceleration will be described with reference to FIGS. 2 and 3. FIG. 2 shows the time course of the calculated lateral acceleration output (Gyc) and the actual lateral acceleration output (Gy) during vehicle steering. During such steering, the actual lateral acceleration follows the steering with time. The property, that is, the phase delay of the actual lateral acceleration output with respect to the calculated lateral acceleration output shows a characteristic peculiar to the road surface μ from the relationship with the tire frictional characteristics. That is, on a road surface with a relatively high coefficient of friction (high μ road),
In contrast to the output of the calculated lateral acceleration Gyc shown by the one-dot chain line in the figure, the output of the actual lateral acceleration Gyc shown by the solid line is a predetermined time T.
There is a phase delay of ₁ and this delay is the output phase difference. On the other hand, a road surface with a low friction coefficient (low μ road)
In this case, as shown by the broken line in the figure, the output of the actual lateral acceleration Gy has a phase delay of a predetermined time T _{2 which} is later than the predetermined time T ₁ , and is higher during steering on a low μ road. The output phase difference between the calculated lateral acceleration Gyc and the actual lateral acceleration Gy tends to be larger than that on the μ road. Therefore, while the inventors of the present invention pay attention to the fact that the road surface friction coefficient can be determined based on the output phase difference between the calculated lateral acceleration and the actual lateral acceleration,
For the determination, the differential value dG of the actual lateral acceleration is set as a value that is not affected by the initial displacement of the sensor output value due to the sensor drift of the lateral G sensor 6 or the lateral inclination of the road surface (bank road).
I have verified that using y is valid. Further, since the differential value dGy of the actual lateral acceleration is used in the determination of the road surface μ, the differential value dGyc of the calculated lateral acceleration is also taken, and these differential values dGyc and d
The deviation dGye from Gy is obtained, and the magnitude of the above-mentioned output phase difference is quantitatively obtained from the effective value of this deviation dGye.

The curve in FIG. 3 shows the differential value of the calculated lateral acceleration Gyc and the actual lateral acceleration Gy in FIG. 2, and the solid line is shown in FIG. 3 with respect to the differential value dGyc of the calculated lateral acceleration indicated by the alternate long and short dash line. The differential value dGy of the actual lateral acceleration on the high μ road indicated by and the differential value dGy of the actual lateral acceleration on the low μ road indicated by the broken line
Shows the same tendency of phase delay as in FIG.
Then, the deviation dGy between these differential values dGyc and dGy
As for e, the deviation dGye ₂ at the time of steering on the low μ road at a certain time t ₀ is larger than the deviation dGye ₁ on the high μ road, and the effective values of such deviations dGye ₁ and dGye ₂ are shown in FIG. Then, the output of the effective value dGyer of the deviation on the high μ road shown by superimposing the solid line in FIG. 3 and the effective value dGyer of the deviation on the low μ road shown by the broken line are obtained.

As is apparent from FIG. 3, the effective value dGyer of the deviation between the differential value dGyc of the calculated lateral acceleration and the differential value dGy of the actual lateral acceleration is higher on the high μ road during steering on the low μ road. It is understood that it is quantitatively greater than. Therefore, in the signal processing shown in FIG. 1, the calculation block 22 quantitatively calculates the effective value dGyer of the deviation. However, in an actual vehicle, a quick steering wheel operation is performed. Therefore, in a situation where the rate of change (differential value) of lateral acceleration is larger than normal, the larger the rate of change, the calculated lateral acceleration and actual lateral acceleration. The output phase difference with acceleration tends to appear larger. In this case, the effective value dGyer of the above-mentioned deviation also becomes large quantitatively.
Therefore, in determining the road surface μ from such a quantitative comparison of the effective value dGyer of the deviation, it is necessary to remove the influence of the steering speed, that is, the change rate of the lateral acceleration on the effective value of the deviation. There is.

Therefore, in the next normalization processing block 24, the effective value dGyer of the deviation is normalized by dividing it by the rate of change of the actual lateral acceleration Gy, that is, the effective value dGyr of the differential value of the actual lateral acceleration. The normalized effective value Re of the deviation is supplied to the decision block 26. Decision block 2
In 6, the road surface μ is determined from the μ determination map prepared in advance based on the absolute value | Gyc | of the calculated lateral acceleration and the normalized effective value Re of the deviation.

Specifically, FIG. 4 shows an example of the μ decision map, and what can be simply understood from the horizontal axis of this map is that the normalized effective value Re of the deviation, that is, as described above. A region where the output phase difference between the calculated lateral acceleration and the actual lateral acceleration is relatively small indicates that the road surface friction coefficient is “high μ” in general, while the output phase difference (effective value of deviation Re Areas where) is relatively large generally indicate that the road surface friction coefficient is “low μ”.

Further, in this μ determination map, the vertical axis takes into consideration the influence of the absolute value of the calculated lateral acceleration | Gyc |, that is, the magnitude of the actual steering angle. Therefore, the above-mentioned “high μ” is taken into consideration. The region is set to an inverted trapezoidal region in which the effective value of the normalized deviation is smaller than the predetermined value Re ₁ even though the steering angle is large enough that the absolute value of the calculated lateral acceleration becomes the predetermined value Gyc ₁ or more. ing. Further, the above-mentioned “low μ” region is set as a region in which the effective value of the normalized deviation is larger than the predetermined value Re ₂ , but the absolute value of the calculated lateral acceleration is the predetermined value Gyc.
_When it is larger than _1, the region is gradually narrowed with the increase of the absolute value of the calculated lateral acceleration and the effective value of the normalized deviation. Note that the normalized effective value of the deviation is the predetermined value R
In the case of e _{1 to} Re ₂ , it is not clear which of the high μ road and the low μ road the output phase difference characteristic between the calculated lateral acceleration and the actual lateral acceleration is, so this portion on the map is not clear. Has no “high μ” or “low μ” region set. A map in which the absolute value of the calculated lateral acceleration is smaller than the predetermined value Gyc ₂ and the absolute value of the calculated lateral acceleration is smaller than the predetermined value Gyc ₁ and the normalized effective value of the deviation is smaller than the predetermined value Re ₂ Also in the upper part, the area of “high μ” or “low μ” is not set.

At the decision block 26, the absolute value | Gyc | of the input calculated lateral acceleration and the effective value Re of the normalized deviation Re are entered.
When the point on the map characterized by is in the “high μ” or “low μ” region, the μ determination result indicating the region is output. On the other hand, when the point on the map is not in either the region of “high μ” or “low μ”, “unknown” is output as the μ determination result.

The switch 28 is switched from the position shown in the figure when the judgment prohibiting signal is not outputted from the OR circuit 30, and allows the μ judgment result outputted from the judgment block 26 to pass therethrough. Here, the determination prohibition signal is the OR circuit 30.
It is supplied to the switch 28 on condition that the logical sum of the above is true, and the switch 28 is switched to the illustrated switching position. In this case, the switch 28 outputs “unknown” as the μ determination result instead of passing the μ determination result from the determination block 26.

Two kinds of steering condition signals are supplied to the OR circuit 30. Of these, the first steering condition signal has an effective value dGyr of the differential value of the actual lateral acceleration from the upper limit value α. Is also large or is smaller than the lower limit value β, it is turned on. Further, the second steering condition signal is turned on when the absolute value | Gy | of the actual lateral acceleration input from the lateral G sensor 6 is larger than the upper limit value γ. When at least one of the first and second steering status signals is turned on, the OR circuit 30 supplies the determination prohibition signal to the switch 28. More specifically, since the steering wheel operation is extremely quick, the rate of change of the actual lateral acceleration is extremely large (dG
yr> α), or the rate of change in actual lateral acceleration is extremely small (dGyr <β) because the steering operation is very slow (dGyr <β), or the level of actual lateral acceleration is excessive due to a sharp turn or a small radius turn ( If | Gyc |> γ), erroneous determination may be caused in any case, so the determination is prohibited.

As described above, the magnitude of the output phase difference between the calculated lateral acceleration and the actual lateral acceleration is quantitatively reflected in the normalized effective value Re of the deviation. Therefore, for example, when the vehicle is traveling on a high μ road such as an asphalt pavement, the output phase difference between the calculated lateral acceleration and the actual lateral acceleration is relatively small, and the effective value Re of the normalized deviation is quantitative. Will be a small value. In this case, the road surface friction coefficient is determined as “high μ” based on the μ determination map of FIG. On the other hand, when the vehicle travels on a low μ road such as a snowy road, the output phase difference between the calculated lateral acceleration and the actual lateral acceleration is relatively large. In this case, since the normalized effective value Re of the deviation is a large value quantitatively, the road surface friction coefficient is determined as “low μ” based on the μ determination map.

Further, the magnitude of the output phase difference between the calculated lateral acceleration and the actual lateral acceleration is quantified from the effective value of the deviation between these differential values, so that the vehicle leans laterally, for example. Even if the vehicle is traveling on a road surface (bank road) and the detected actual lateral acceleration Gy is given an initial displacement (sin θ of gravity acceleration) according to the road surface inclination angle θ,
The road friction coefficient can be accurately determined regardless of such initial displacement.

The present invention is based on, for example, the ECU shown in FIG.
In the inside signal processing circuit, a block element for obtaining a transfer function is not required. Therefore, the program capacity and the processing time for performing the FFT processing can be omitted, and the load on the ECU can be significantly reduced. Further, since the ECU to be used can be downsized, it is very suitable as a friction coefficient determination device to be mounted on an automobile.

In the above-mentioned embodiment, the judgment block 26 only judges the road surface μ, but the output μ judgment result is displayed on the display device installed in the instrument panel of the vehicle, and the μ judgment is made. Of course, the result can be used for slip control of a vehicle anti-skid brake system or traction control system.

Further, the μ judgment result is supplied to another road surface condition judging device, for example, a device for judging the road surface condition as “dry road”, “wet road” or “snow road” by using an optical sensor. It is also possible to grasp the information such as the road surface friction coefficient and the road surface state with higher accuracy in cooperation with another discriminating device.

[0027]

As described above, according to the road surface friction coefficient determining device of the first aspect, the road surface friction coefficient can always be accurately determined regardless of the inclination of the road surface and the sensor drift.

[Brief description of drawings]

FIG. 1 is a block diagram showing a circuit configuration of a road surface friction coefficient determination device according to an embodiment.

FIG. 2 is a graph for explaining an output phase difference between a calculated lateral acceleration and an actual lateral acceleration.

FIG. 3 is a graph for explaining an effective value of a deviation obtained from the differential value of the lateral acceleration output of FIG.

FIG. 4 is a graph showing a μ determination map.

[Explanation of symbols]

2 Vehicle speed sensor (vehicle speed detection means) 4 Steering wheel angle sensor (steering angle detection means) 6 Lateral G sensor (actual lateral acceleration detection means) 8 ECU

─────────────────────────────────────────────────── ─── Continuation of the front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) B60R 16/02 B60T 8/58 G01N 19/02

Claims (4)

(57) [Claims] 1. A vehicle speed detecting means for detecting a vehicle speed, a steering angle detecting means for detecting a steering angle, and a calculated lateral acceleration to be generated in a vehicle body based on the detected vehicle speed and steering angle, and outputs the calculated lateral acceleration. Calculating means, actual lateral acceleration detecting means for detecting and outputting the actual lateral acceleration generated in the vehicle body, and obtaining an output phase difference from the deviation between the differential value of the calculated lateral acceleration and the differential value of the actual lateral acceleration, A road surface friction coefficient determination device, comprising: a determination unit that determines a road surface friction coefficient based on an output phase difference. 2. The determination means is adapted to detect the calculated lateral acceleration.
By the effective value of the deviation between the minute value and the differential value of the actual lateral acceleration
Obtain the output phase difference, and based on this output phase difference the road surface friction
The road surface friction according to claim 1, wherein a coefficient is determined.
Rub coefficient determination device. 3. A normalization process is applied to the output phase difference.
The road surface friction coefficient according to claim 2, characterized in that
Judgment device. 4. The determination means is a derivative of the actual lateral acceleration.
When the effective value of the value is larger than the upper limit value, or
When the effective value of the differential value of acceleration is smaller than the lower limit value,
4. The method according to claim 2, wherein the judgment is prohibited.
Road friction coefficient determination device.