JP2000213981A - Vehicle mass calculation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To remove the influence of a resistance caused by a grade when calculating a vehicle mass from a driving force by an engine, a traveling resistance and an acceleration of the vehicle. SOLUTION: A true driving force F of a vehicle is calculated by drawing a traveling resistance from a vehicle driving force by an engine at a true driving force calculation part 26. An acceleration α in a forward/backward direction of the vehicle is calculated by an acceleration sensor 30. A relationship of the driving force F, the acceleration α, the vehicle mass M and a road grade Θis represented by a formula: α=F/M-gsinΘ. Since the grade Θ only has a low frequency component which is not timely and rapidly varied, an influence of the grade Θ can be removed from the true driving force F and the acceleration α by processing with high pass filters 28, 32 of a predetermined cut off frequency. The vehicle mass can be determined by the formula based on the true driving force F and the acceleration α after filter-processing without receiving the influence of the grade Θ.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for calculating a vehicle mass used for determining a shift timing of an automatic transmission.

[0002]

2. Description of the Related Art At present, the shift control of an automatic transmission determines a shift timing and a shift speed based on not only the operation amount of an accelerator pedal and a vehicle speed but also various factors such as a gradient of a running road and a mass of a vehicle. The aim is to achieve more efficient control that is suited to the passenger's sensitivity.

[0003] Conventionally, automatic transmissions have prevailed mainly in so-called private passenger cars, but are now also prevalent in commercial vehicles such as taxis and route buses. In the case of a private car, the number of occupants is relatively small, so that the mass of the vehicle does not change much. Therefore, it is not necessary to treat the vehicle mass as a variable in the control of the transmission. However, in the case of a route bus or the like, the mass of the vehicle greatly changes depending on the number of passengers. Therefore, it is preferable to treat the mass of the vehicle as a variable instead of a fixed value even in the shift control. In recent years, as for passenger cars, vehicles with a relatively large number of so-called minivan-type passengers have become widespread,
In these vehicles, it is necessary to treat the vehicle mass as a variable. There is also a demand for more accurate shift control in a conventional so-called four-door sedan type passenger car.

As a method of measuring the vehicle mass, a method of measuring a stroke amount of a suspension in a static state of the vehicle and a method of measuring with a load sensor provided on the suspension are conceivable.
The measurement is performed only when the vehicle is stationary.

[0005] As a method of measuring the vehicle mass while the vehicle is running, a method based on the relationship between acceleration and driving force is known. That is, it is a method of determining that the vehicle mass is large if the acceleration obtained when the predetermined driving force is generated is small, and determining that the vehicle mass is small if the acceleration is large.

In Japanese Patent Application Laid-Open No. 6-147304, the acceleration of a vehicle when an accelerator pedal is depressed,
2. Description of the Related Art There is disclosed a technique for calculating a vehicle mass using a neural network that inputs a time-series signal indicating a vehicle speed and a throttle valve opening degree. Japanese Patent Application Laid-Open No. Hei 6-201523 discloses that, while traveling on a road having a constant gradient and a change in the throttle valve opening is small, the acceleration, speed, and throttle valve opening of the vehicle are detected and the throttle valve opening is detected. The vehicle mass is measured by comparing two different states. According to these methods, the vehicle mass can be obtained while the vehicle is running.

[0007]

However, Japanese Patent Application Laid-Open No.
In Japanese Patent No. 47304, no consideration is given to the gradient of a road, and accurate calculation of vehicle mass cannot be performed unless the ground is flat. Also, Japanese Patent Application Laid-Open No. 6-201523
In the publication, accurate calculation of the vehicle mass cannot be performed unless the gradients of the two states to be compared are equal. As described above, according to the techniques disclosed in the above two publications, there is a problem that the gradient of the road must be accurately detected in order to accurately determine the vehicle mass.

An object of the present invention is to calculate a vehicle mass in a dynamic state of a vehicle without being affected by a gradient of a road.

[0009]

In order to solve the above-mentioned problems, a vehicle mass calculating apparatus according to the present invention calculates acceleration in the front-rear direction of a vehicle and obtains an acceleration signal. A driving force calculating unit that obtains a driving force signal from a force, a first signal processing unit that obtains a processed acceleration signal by removing an influence of a road gradient from the acceleration signal, and a driving signal calculating unit that removes an effect of a road gradient from the driving force signal. A second signal processing means for obtaining a processing driving force signal; and a vehicle mass calculating means for calculating a vehicle mass based on the processing acceleration signal and the processing driving force signal.

The relationship between the force acting on the vehicle and the acceleration of the vehicle is expressed by the equation of motion.

## EQU00001 ## (force acting on the vehicle) = (mass of the vehicle) .times. (Acceleration of the vehicle). If the acting force and the acceleration are known, the vehicle mass can be calculated. The forces acting on the vehicle include driving force of the prime mover, frictional resistance such as rolling resistance, air resistance, and resistance due to road gradient. The force other than the resistance due to the gradient can be obtained in advance if conditions such as the running speed and the throttle valve opening are determined. On the other hand, the resistance due to the gradient is not a characteristic of the vehicle body because it is a gradient of a road on which the vehicle is traveling at that time, and therefore cannot be obtained in advance.
However, when the vehicle is traveling on a road with a constant gradient (including a flat road), the resistance due to the gradient becomes a constant, and has only a DC component in a time series. Therefore, if this component is removed, an equation of motion that is not affected by the gradient can be obtained. In addition, even if the gradient is not constant, if a frequency band in which the change in the gradient gives resistance to the gradient can be specified, the frequency band can be removed and an equation of motion that is not affected by the gradient can be obtained. Furthermore, the frictional resistance such as the rolling resistance, the air resistance, and the like also have a low frequency of change, and the influence can be removed as in the case of the road gradient.

Therefore, the first and second signal processing means can be a filter for removing a frequency band in which the influence of the gradient appears. Furthermore, as described above, the effect of the constant gradient appears as a DC component, and is considered to be a low-frequency component even when the gradient changes, so that a high-pass filter that removes the low-frequency component is preferable. .

[0012]

Embodiments of the present invention (hereinafter, referred to as embodiments) will be described below. The equation of motion of the vehicle in the front-rear direction is as follows, where M is the vehicle mass, α is the acceleration, F is the force acting on the vehicle, excluding the road gradient force, 道路 is the road gradient, and g is the gravitational acceleration.

## EQU2 ## M × α = F−MgsinΘ (1) The force F is obtained by subtracting the running resistance from the driving force of the prime mover, and is hereinafter referred to as a net driving force. If the relationship between the driving force of the prime mover, the amount of operation required by the driver such as the operation amount of the accelerator pedal and the throttle valve opening, the rotation speed of the prime mover, the total gear ratio, etc. is obtained in advance, the operation in a certain prime mover The driving force can be calculated. The running resistance is the sum of the frictional resistance that is not affected by the vehicle speed, such as rolling resistance, and the air resistance that is approximately proportional to the square of the vehicle speed. The running resistance at speed can also be calculated. Therefore, the net driving force F of the vehicle at a certain point during traveling can be calculated. The acceleration α of the vehicle can be obtained as a differential value of the speed sensor or as an output of the acceleration sensor.

The vehicle mass M is a value that changes with respect to the initial mass m in consideration of changes in the number of passengers, the amount of loaded luggage, and the like. If the vehicle mass variation parameter is θ, the vehicle mass M is

## EQU3 ## M = θm (2) From equation (2), equation (1) is

Α = F / (θm) −g sinΘ (3) If the vehicle is traveling on a road with a constant slope (平坦 = 0 on a flat road), in equation (3), the term (−g sin Θ) relating to the slope resistance becomes a constant and affects only the DC component of the acceleration α. Therefore, by using a signal obtained by removing the DC component from the signal representing the acceleration α and the signal representing the net driving force F, a motion equation that is not affected by a constant gradient can be obtained.

However, although the road gradient Θ is not constant and generally changes, the change is relatively gradual and affects only the low-frequency component of the acceleration α.
Therefore, for each signal representing the acceleration α and the net driving force F, even if the gradient changes by removing not only the DC component but also a signal below a certain frequency,
An equation of motion that is not affected by the gradient can be obtained.
Assuming that the acceleration from which a signal below a predetermined frequency is removed is α and the net driving force is F , equation (3)

Α = F / (θm) −g sin Θ (4) Here, the term (−g sin Θ ) is considered to be a term caused by a measurement error of the gradient resistance or acceleration α and the net driving force F that cannot be completely removed. There is a known method for obtaining a true value from a value including an error, and the vehicle mass variation parameter θ or the vehicle mass M can be obtained using the known method.

Next, it will be examined to what frequency the effect of the gradient appears. As described above, if the gradient is constant, only the DC component has an effect on the acceleration α. Therefore, only the frequency characteristics during the period when the gradient changes from one value to another value may be considered. That is, if a signal below a certain frequency is removed, the DC component is naturally removed, and the influence of the gradient at a constant gradient is also eliminated. There is a road structure decree as a standard of road design for impact mitigation in a portion where the road gradient changes. The fact that impact mitigation is not required means that the slope change is considered to be small, so it may be considered that the slope change indicated in the road structure order is the largest example of the slope change, that is, the case where a higher frequency is affected. it can. Therefore, by setting a cutoff frequency that can remove the gradient change indicated in this road structure order, the influence of the gradient change on other roads can also be removed.

FIG. 1 shows the specification of a portion where the gradient changes according to the Road Structure Ordinance. As shown in FIG. 2, the road changes from the gradient theta _A gradient theta _B, length length of arc AB of the interference portion is defined. The required longitudinal curve length L shown in FIG.

(Equation 6) It is represented by A frequency analysis of a function Θ (t) showing a time change of a gradient と き when passing through a road having a required vertical curve length of 4.4 m /% at 80 km / h, which is twice as fast as the design speed, is shown in FIG. The frequency distribution shown by the solid line is obtained.
The distribution indicated by the dashed line in FIG. 3 is for the case where the passing speed is lower. As described above, even when the vehicle passes through a portion where the gradient of the road is changing at twice the design speed, the function Θ (t) has only a component having a frequency of approximately 1 Hz or less. Therefore, if the acceleration signal α and the net driving force signal F are processed using a high-pass filter capable of removing this frequency band, the influence of the gradient can be removed. As shown in FIG. 4, even if the frequency analysis is performed on the function Θ (t) of the gradient change when passing through an actual road (a road having a gradient as shown in the right-side diagram), only a component of approximately 1 Hz or less is obtained. It turns out that it has. That is, the frequency component of the change in the gradient is only 2 Hz at most. On the other hand, the change in the driving force has a larger component than the change in the gradient even in a region of 2 Hz or more. FIG. 5 is a diagram illustrating an example of a frequency analysis of a change in the throttle valve opening substantially corresponding to a change in the driving force. (B) shows the result of frequency analysis of the throttle valve opening in the running state shown in (a). Therefore, in the acceleration α and the net driving force F 2 processed by the high-pass filter having the cutoff frequency of 2 Hz, that is, in the high-frequency components of the acceleration and the net driving force, the vehicle mass M (= θm) is calculated using Expression (4).
Can be calculated. The value (2 Hz) of the cutoff frequency can be set to an appropriate value by a driving experiment or the like, and is not limited to 2 Hz.

Next, an example of a method of calculating the vehicle mass fluctuation parameter θ will be described. Regarding equation (4), the term (−g
Assuming that sin と ) is the residual e, equation (4) becomes

Α (k) = F (k) / (θm) + e (k) (5) Here, k indicates the number of times of sampling.

By calculating the vehicle mass variation parameter θ from equation (5), the vehicle mass M can be obtained. In the equation (5), if the residual e (k) is ignored, the parameter θ can be obtained immediately. However, if the residual e (k) cannot be ignored, the parameter θ of the parameter An estimate can be made. Input u used for estimation
To

## EQU8 ## If u = F / m (6), the estimated gain kn becomes

Kn (k + 1) = Pn (k) u (k + 1) / {λ + u (k + 1) Pn (k) u (k + 1)} (7) Here, Pn is an estimated gain parameter, and an initial value is selected as a sufficiently large integer, for example, 1000.

Pn (k + 1) = {1−kn (k + 1) u (k + 1)} Pn (k) / λ (8) Λ is a forgetting factor, for example, 0.99. From the above, the vehicle mass fluctuation parameter θ
Is

(11) θ (k + 1) = θ (k) + kn (k + 1) {y (k + 1) −u (k + 1) θ (k)} (9) Here, y (k + 1) = α (k). Therefore, the vehicle mass M estimated in the sampling period (k + 1) is represented as θ (k + 1) m.

FIG. 6 is a block diagram showing the configuration of the vehicle mass measuring apparatus according to the present embodiment. The engine driving force calculation unit 10 reads an engine output corresponding to these detection values from the engine output storage unit 16 from the detection values of the throttle valve opening sensor 12 and the engine rotation speed sensor 14. Further, a drive wheel output is calculated based on the gear ratio based on the detected value of the gear position sensor 18, and a driving force of the engine by the engine on the drive wheel is calculated based on the tire radius. The engine output corresponding to the throttle valve opening and the engine speed is stored in the engine output storage unit 16 in advance. Further, it is preferable that data considering not only steady values but also transient characteristics is stored. On the other hand, the running resistance calculation unit 20 reads the running resistance corresponding to the detected value from the running speed storage unit 24 from the detected value of the vehicle speed sensor 22, and calculates the running resistance. Running resistance storage unit 24
May store the running resistance as a map for the vehicle speed, or may be stored as a function indicating the running resistance curve. The net driving force calculation unit 26 calculates a force (net driving force) acting on the vehicle excluding the gradient resistance by subtracting the running resistance from the driving force of the vehicle by the engine.
The processed net driving force signal F is obtained by passing the time series signal F of the net driving force through the high-pass filter 28.

On the other hand, the time-series signal α output from the acceleration sensor 30 is also passed through the high-pass filter 32 to obtain a processed acceleration signal α . The acceleration can also be obtained by differentiating the output value of the vehicle speed sensor 22. Conversely, the vehicle speed can also be obtained by integrating the output value of the acceleration sensor 30.

The vehicle mass can be calculated by the vehicle mass calculator 34 based on the processed net driving force signal F and the processed acceleration signal α . As described above, when calculating the vehicle mass by the sequential least squares method, the fluctuation parameter estimation unit 36 estimates the fluctuation parameter θ of the vehicle mass, and calculates the vehicle mass M based on this and the initial value m of the vehicle mass. Is done.

FIGS. 7 and 8 show examples of calculating the vehicle mass according to this embodiment of the vehicle. FIG. 8 shows the calculation result of the vehicle mass when the vehicle travels on the road with the changing gradient shown in FIG. 7 at the throttle opening shown in FIG. FIG.
From this, it can be seen that the vehicle mass approaches the true value even if the gradient changes.

Note that each calculation unit shown in FIG.
It is a CPU (Central Processing Unit) that operates according to a predetermined program, and each storage unit is a well-known storage unit such as a ROM (Read Only Memory). In the embodiment shown in FIG. 6, the prime mover is described as an engine. However, the prime mover may have any form such as an electric motor and a combination of an engine and an electric motor. Furthermore, in the present embodiment, the signal of the net driving force obtained by subtracting the running resistance from the driving force of the prime mover is input to the high-pass filter, but if the frequency of the changing running resistance is equal to or less than the frequency of the changing road gradient, The signal of the driving force of the prime mover may be input to the high-pass filter.

[Brief description of the drawings]

FIG. 1 is a diagram showing specifications of a gradient change required for cushioning a road.

FIG. 2 is a view showing a longitudinal curve of a portion where a road gradient shown in FIG. 1 changes.

FIG. 3 is a function を showing a time change of a gradient when the vehicle passes through a changing portion of the road gradient shown in FIG. 1 at a predetermined speed.
It is a figure showing the result of having performed frequency analysis of (t).

FIG. 4 is a diagram showing a result of frequency analysis of a function Θ (t) indicating a time change of a gradient when the vehicle passes through an illustrated road gradient.

FIG. 5 is a diagram showing a frequency analysis result of a time change of a throttle valve opening (driving force).

FIG. 6 is a block diagram illustrating a configuration of a vehicle mass calculation device according to the present embodiment.

FIG. 7 is a diagram showing simulation conditions.

FIG. 8 is a diagram showing a result of a simulation of the present embodiment performed under the conditions shown in FIG. 7;

[Explanation of symbols]

26 Net driving force calculator, 28, 32 High pass filter, 30 acceleration sensor, 34 Vehicle mass calculator.

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Masataka Osawa 41-cho, Yokomichi, Nagakute-machi, Aichi-gun, Aichi Prefecture Inside Toyota Central Research Institute, Inc. (72) Inventor Masuji Oshima Nagakute-machi, Aichi-gun, Aichi No. 41, Changchun Yokomichi 1 Toyota Central Research Laboratory Co., Ltd.

Claims (2)

[Claims] 1. An acceleration calculating means for calculating an acceleration signal in a longitudinal direction of a vehicle to obtain an acceleration signal, a driving force calculating means for obtaining a driving force signal from a driving force of a motor, and removing an influence of a road gradient from the acceleration signal. A first signal processing means for obtaining a processing acceleration signal; a second signal processing means for removing an influence of a road gradient from the driving force signal to obtain a processing driving signal; and the processing acceleration signal and the processing drive A vehicle mass calculating unit configured to calculate a vehicle mass based on the force signal. 2. The vehicle mass calculating device according to claim 1, wherein the first signal processing means is a first high-pass filter that removes a band below a predetermined frequency, and wherein the second signal processing is performed. The means is a second high-pass filter that removes a band below the predetermined frequency.