JP2002365085A - Estimation apparatus for actual speed - Google Patents

Abstract

PROBLEM TO BE SOLVED: To exclude the influence of a gravitational acceleration and a yaw rate which are contained in the output of an acceleration sensor, and to precisely estimate an actual speed including a transient region in an apparatus which estimates the actual speed of a vehicle by integrating detected vehicle accelerations. SOLUTION: A speed on a vehicle coordinate axis is found on the basis of output information on a GPS in which a magnetic direction sensor is built. An estimated speed on the coordinate axis is found on the basis of a front acceleration sensor, a rear acceleration sensor and a transverse acceleration sensor as well as a yaw rate sensor. Mean values of both speeds are found, outputs of the respective acceleration sensors are corrected on the basis of the difference between both mean values, and a transverse-direction actual speed, a front-direction actual speed and a rear-direction actual speed are estimated and computed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an actual vehicle speed estimating apparatus, and more particularly to an apparatus for estimating an actual vehicle speed of a vehicle by integrating detected vehicle acceleration.

[0002]

2. Description of the Related Art Such an actual vehicle speed estimating apparatus is disclosed in Japanese Patent Application Laid-Open Nos.
No. 27329 has already been proposed. Of these,
No. -107155 obtains each wheel speed based on the actual vehicle speed in the front-rear and lateral directions of the vehicle and the yaw rate, and calculates the slip ratio of the four-wheel drive vehicle based on the difference between the actual wheel speed and the actual wheel speed. In order to determine the vehicle speed, the longitudinal and lateral accelerations measured on the vehicle are integrated.

In Japanese Patent Application Laid-Open No. 8-127329, it is determined whether the vehicle is traveling straight or turning. When the vehicle is traveling straight, the longitudinal acceleration is integrated to determine the actual vehicle speed in the longitudinal direction. The value divided by the yaw rate is determined as the actual vehicle speed in the front-rear direction.

[0004]

The problems of the above prior art will be described below. Assuming that the vehicle is moving while rotating at the yaw rate (angular velocity) r on the XY plane as shown in FIG. 6, the position vector of the center of gravity position O of the vehicle is If so, its speed Can be expressed as the following equation.

[0005]

(Equation 1)

The following equation is obtained by differentiating both sides of the equation (1).

[0007]

(Equation 2)

[0008] Change for Δt seconds If Δt is sufficiently small, the following equation is obtained from FIG.

[0009]

(Equation 3)

[0010]

(Equation 4)

Therefore, Is differentiated into the following equation.

[0012]

(Equation 5)

[0013]

(Equation 6)

By substituting these equations (5) and (6) into equation (2) and deforming, the acceleration vector at the center of gravity position O is represented by the following equation.

[0015]

(Equation 7)

If the longitudinal and lateral accelerations of the vehicle are set as a _x and a _y , respectively, their magnitudes are given by the following equations (7). , The following equations are obtained.

[0017]

(Equation 8)

[0018]

[Equation 9]

This indicates that the acceleration measured on the vehicle is affected not only by pure speed changes in the longitudinal and lateral directions of the vehicle but also by the yaw rate components vr and ur as errors. That is, JP-A-6-107155
As shown in the figure, when the vehicle speed is obtained by integrating the values detected by the longitudinal and lateral acceleration sensors on the vehicle, the above error due to the yaw rate is included when the vehicle turns,
The actual vehicle speed cannot be estimated accurately.

On the other hand, Japanese Patent Application Laid-Open No. 8-127329 basically assumes that a vehicle is turning at a constant vehicle speed. The actual vehicle speed in the front-rear direction of the vehicle at the time of turning is obtained by the following expression by transforming the above expression (9).

[0021]

[Equation 10]

Since this equation does not take into account the transient characteristics, the state in which the vehicle can be regarded as a steady state If the vehicle abnormally deviates from, for example, if the vehicle suddenly accelerates or decelerates during a turn, an error occurs. In any of the above cases, there is a process of integrating the acceleration signal obtained from the acceleration sensor, but the acceleration sensor not only senses the acceleration generated when the vehicle turns or accelerates or decelerates, but also senses the gravitational acceleration. I will.

That is, normally, the acceleration sensor is set so that the effect of the gravitational acceleration is canceled and the output becomes "0" when the vehicle is stopped on the horizontal plane.
When the vehicle 10 travels on a downhill as shown in FIG. 8, even if the vehicle is traveling at a constant speed without acceleration / deceleration due to resistance such as wind, and the acceleration sensor should output “0”,
The acceleration sensor calculates the gravitational acceleration in the vehicle acceleration detection axis direction.
detecting a force component g ₁ of g, so that occurs the output signal corresponding to the magnitude.

That is, the "0" point of the acceleration is offset. If the speed is calculated by performing an integration process based on this signal output, the estimated actual vehicle speed value gradually increases or decreases contrary to the actual situation. Apart from such estimating means, it is conceivable to directly detect the actual vehicle speed by a laser ground speed sensor, a radio wave ground speed sensor, or the like, but it is said that a vehicle equipped with these would be too expensive (million yen). There's a problem.

In addition, GPs whose vehicle mounting rate is increasing in recent years
Although the actual vehicle speed can be obtained also by S (Global Positioning System), estimating the actual vehicle speed only with this GPS is inferior in accuracy and low in sampling rate (1 Hz
It is not suitable for vehicle control.

Accordingly, the present invention is an apparatus for estimating the actual vehicle speed of a vehicle by integrating the detected vehicle acceleration, eliminating the effects of gravitational acceleration and yaw rate on the output of the acceleration sensor and accurately including the transient region. The purpose is to estimate the speed.

[0027]

In order to achieve the above object, an actual vehicle speed estimating apparatus according to the present invention includes a longitudinal acceleration sensor, a lateral acceleration sensor, a yaw rate sensor, a GPS incorporating a magnetic direction sensor, and The vehicle speed on the vehicle coordinate axis is obtained from the output information of the GPS, the estimated vehicle speed on the vehicle coordinate axis is obtained from each sensor, the average value of both vehicle speeds is obtained, and the longitudinal acceleration and the lateral acceleration are corrected based on the difference between both average values. Computing means for calculating the actual vehicle speed in the lateral direction and the actual vehicle speed in the front-rear direction.

That is, the present invention, in short, performs correction using the difference between the estimated average value of the actual vehicle speed and the average value of the actual vehicle speed calculated based on the output information of the GPS. Thus, the accuracy and response are good, and the actual vehicle speed can be estimated relatively inexpensively.

The average value may be calculated at the timing of the GPS output information, or a moving average value may be used as the average value. Hereinafter, the principle of the present invention will be described. V _{xn + 1} , v
_{yn + 1,} and the value of the longitudinal / lateral acceleration detected by the longitudinal acceleration sensor and the lateral acceleration sensor and the yaw rate of the vehicle detected by the yaw rate sensor at a time before Δt from this time, respectively, are a _xn , a _yn , r _n and the estimated front / rear and side actual vehicle speeds are v _xn and v _yn ,
Equations (8) and (9) above can each be written as:

[0030]

[Equation 11]

[0031]

(Equation 12)

Accordingly, the actual vehicle speed at a certain time can be obtained by the following equation obtained by modifying equations (11) and (12).

[0033]

(Equation 13)

[0034]

[Equation 14]

On the other hand, GPS measures the time required for a radio wave emitted from a satellite to reach a receiving point (vehicle, if mounted), and determines the current position of the receiving point based on the distance from the orbit. It is a system that determines the position. Generally, commercially available GPS outputs latitude / longitude information calculated from this time.

Accordingly, the latitude / longitude information output from the GPS at the time point m is L _Am and L _Om [rad], and the radius of the earth is R
Assuming d, the ground coordinates X _m and Y _m (coordinates with respect to the X and Y axes of the earth) of the vehicle are given by the following equations.

[0037]

(Equation 15)

[0038]

(Equation 16)

Although the above example shows an example of calculating the ground coordinates from the latitude / longitude information, it is not always necessary to use the above equations (15) and (16) in order to obtain the ground coordinate output. May be calculated from Assuming that the output of the ground coordinates before k · Δt before this point is X _m-1 and Y _m-1 , the actual ground speeds V _Xm and V _Ym obtained using the output information from the GPS are expressed by the following equations. Can be.

[0040]

[Equation 17]

[0041]

(Equation 18)

Note that as well as There is also a GPS which can directly output the data, and thus the GPS may be directly obtained using such a GPS. These actual vehicle speed values are converted into coordinates in the front-rear and lateral directions of the vehicle. Generally, the azimuth angle can be detected by the GPS mounted on the vehicle.

The "azimuth angle" mentioned here is not the traveling direction of the vehicle but the direction of the vehicle, that is, the direction of the nose of the vehicle. (At the time of turning, the traveling direction and the direction of the vehicle are different.) Generally, the GPS of the navigation system mounted on the vehicle does not have a function to detect the direction of the vehicle, so to detect the direction of the vehicle, It must be a GPS with a built-in magnetic direction sensor.

If this azimuth is represented by ψ, a matrix for converting the absolute coordinates of the ground surface into coordinates on the vehicle is given by the following equation.

[0045]

(Equation 19)

The speed in the vehicle coordinate system calculated based on the output information of the GPS is And put Then, the following equation is obtained.

[0047]

(Equation 20)

Here, the in-vehicle yaw rate sensor,
The vehicle speed in the on-vehicle coordinate system estimated from the lateral acceleration sensor

In addition, since the measurement rate of the sensor is usually higher than that of the sensor, it is calculated in accordance with the output sampling timing of the GPS in order to adjust the error with the GPS.

Let us consider obtaining each moving average. The calculated value at a certain point in time m when the GPS output information is detected

Assuming that the number of samplings to be averaged is p, the average value of each is obtained by the following equation.

[0052]

[Equation 21]

[0053]

[Equation 22]

Estimated vehicle speed Sampling time Δt, vehicle speed by GPS

Assuming that the sampling time is k · Δt,
The time required for the averaging is k · p · Δt. During this time When Deviation between Has occurred. This deviation It is considered that the factor is caused by the “0” point offset of the vehicle-mounted acceleration sensor. Therefore, the zero offset amount to be corrected is calculated by the following equation.

[0056]

[Equation 23]

Therefore, the corrected actual vehicle speed estimated value is expressed by the following equation.

[0058]

[Equation 24]

[0059]

[Equation 25]

In this way, the average error between the vehicle speed detected by the GPS and the vehicle speed estimated from the output of each sensor is determined to periodically correct the acceleration detected by the sensor. It is possible to eliminate the effects of the included gravitational acceleration and yaw rate.

[0061]

FIG. 1 shows an embodiment of an actual vehicle speed estimating apparatus according to the present invention. In this embodiment, an ignition switch (SW) 1, a longitudinal acceleration sensor 2, and a lateral acceleration sensor 3 are shown. And a yaw rate sensor 4, a GPS 5, and a calculation means 6.

The arithmetic means 6 includes a low-pass filter (LPF) 61 which receives detection signals from the sensors 2 to 4 and extracts only low-frequency components, and converts an analog output signal of the low-pass filter 61 into a digital signal. A / D conversion unit 62
Estimate the actual vehicle speed by inputting each digital signal output from the / D conversion unit 62, the output signal of the ignition switch 1, the latitude information L _A , the longitude information L _O , and the direction information 出力 output from the GPS 5 The actual vehicle speed calculating unit 63 converts the digital signal output from the actual vehicle speed estimating calculating unit 63 into an analog signal and converts the actual vehicle speed v _{x2 in} the front-rear direction (X-axis direction) and the lateral direction of the vehicle.
The D / A converter 64 outputs the actual vehicle speed v _y2 (in the Y-axis direction).

The actual vehicle speed estimation calculation by the actual vehicle speed estimation calculation unit 63 will be described in detail below with reference to the flowcharts shown in FIGS. FIG. 2 shows a main routine of a program executed in the actual vehicle speed estimation calculating section 63. First, it is determined whether or not the ignition switch is turned on from the output of the ignition switch 1, and the ignition switch is turned on. If not, set the SW flag to OFF (step
S5) Exit this routine.If the ignition switch is ON, it is determined whether the SW flag is OFF.
(Step S2) Initialization is executed only when the SW flag is OFF (Step S3). This is to clear the numerical values remaining at the time of the previous processing, such as when the vehicle starts.

In this initialization step, the estimated actual vehicle speed v _{x1 in the} front-rear direction (corresponding to v _xn in equations (11) and (12)) and the estimated actual vehicle speed v _{y1 in the} lateral direction (corresponding to v _yn ) are _calculated. Set to “0” for both, GPS
The coordinate values X ₁ and Y ₁ detected by 5 are both set to “0”. In addition, the longitudinal acceleration correction amount a _xoff and the lateral acceleration correction amount a _yoff are both set to “0”, and the sampling parameter N ₁ and the moving average parameter M ₁ are both set to “0”.

Further, the speed parameters v _GTTx and v _GTTy used in the GPS5 calculation process described later are both set to “0”,
The speed parameters v _ETTx and v _ETTy for _estimation are both set to “0”. Thereafter, the actual vehicle speed calculation routine is executed.
(Step S4).

FIG. 3 shows the actual vehicle speed calculation routine shown in FIG.
9 shows details of (Step S4). In this routine, first, the low-pass filter 61 and the A / D converter 6
The longitudinal acceleration a _x , the lateral acceleration a _y and the yaw rate r from the sensors 2 to 4 which are input via the input 2 are input (step S11).

[0067] Then, the parameter N ₁ "1" is incremented by, and N ₂ (step S12). Then, a decision is made as to whether the parameter N ₂ is equal to or greater than the constant k (step S13). This is because the sampling rate of sensors 2 to 4 is higher than that of GPS5.
This is to adjust using k.

Therefore, when N ₂ = k, the GPS
Read latitude information L _A , longitude information L _O and azimuth information か ら from 5
(Step S14). Then, the first actual vehicle speed estimation calculation is executed as shown in the figure (step S15). These equations correspond to the above equations (24) and (25).

However, at present, the parameters a _xoff ,
Since v _x1 , a _yoff , and v _y1 are all initially set to “0”, the actual actual vehicle speed estimation calculation is not performed, but formally the first estimated actual vehicle speed v _x2 and v _y2 will be determined. Thereafter, it is checked whether a re-parameter N ₂ is k or more (step S16), and proceeds to the average vehicle speed calculating routine only when at least N ₂ = k (step S17).

After executing the average vehicle speed calculation routine,
The parameter N ₂ is returned to "0" with (step S18), and with the newly obtained this time estimated actual vehicle speed in step S15 v _x2, v _y2 respectively set to the estimated vehicle speed v _x1, v _y1 previous, current parameter N ₂ is also changed to a reference value to be used for subsequent calculation by setting the previous parameters N ₁ of (step S19).

FIG. 4 shows an average vehicle speed calculation routine shown in FIG.
9 shows details of (Step S17). In this routine, first, the parameters M ₁ for the average is incremented by "1", and the parameter M ₂ (step S
twenty one).

[0072] Then, the latitude information L _A and longitude information L _O from GPS5 read in step S14 in the FIG. 3 X, Y-coordinate values X _2,
It converted into Y ₂ (step S22). This corresponds to the above equations (15) and (16). Then, using the present coordinate values X ₂ , Y ₂ and the previous coordinate values X ₁ , Y ₁ , the vehicle is converted into actual vehicle ground speeds V _X , V _Y as viewed from the GPS 5 (step S23). ).
This corresponds to the above equations (17) and (18). However, X _1, as the first time shown in step S3 in FIG. 2 Y
₁ is both "0".

Further, the vehicle speeds V _X and V _Y obtained in this manner are obtained.
(Absolute coordinate system on the ground) (Vehicle coordinate system) (step S24). This is the above equation
They correspond to (19) and (20). Vehicle speed obtained in this way

The forward / backward component v _Gx and the lateral component v _Gy are added to the previous values v _GTTx and v _GTTy , respectively, to obtain the current accumulated values v _GTx and v _GTy (step S25). However, for the first time,
Since both v _GTTx and v _GTTy are “0” as shown in step S3 in FIG. 2, the current vehicle speeds v _GTx and v _GTy are v _Gx and v
Equivalent to _Gy .

In step S25, the estimated actual vehicle speeds v _x2 , v _x calculated in step S15 in FIG.
Values v _ETx , v _{ETy obtained} by adding _y2 to the previous values v _ETTx , v _ETTy
Ask for. Also in this case, since v _ETTx and v _ETTy are “0” for the first time, the current values v _ETx and v _{ET y} are equal to the estimated actual vehicle speeds v _x2 and v _y2 .

Thereafter, the parameter M obtained in step S21
Judge whether ₂ is equal to the parameter p for averaging
(Step S26) The operations of steps S27 to S29 are executed only when M ₂ = p or more, and otherwise, these steps are skipped. That is, in step S27, the average values v _GAx , v _GAy , v _EAx , and v _EAy obtained by dividing the v _GTx , v _GTy , v _ETx , and v _ETy obtained in step S25 by the sampling number p, respectively.
Ask for. These equations correspond to the above equations (21) and (22).

Thereafter, an acceleration correction amount calculation routine is executed (step S28). The details of the acceleration correction amount calculation routine are shown in FIG. 5, and the time required for averaging the value obtained by subtracting v _GA from v _EA obtained in step S27 for the front-rear direction and the lateral direction of the vehicle is k · p · The current acceleration correction amounts a _xoffc and a _yoffc are obtained by subtracting the value obtained by dividing by Δt from the previous acceleration correction amount parameter a _xoff . This is in accordance with the above equation (23).

However, first, in step S3 in FIG.
As shown in the above, both the parameters a _xoff and a _yoff are “0”
Since it is, only the values shown in the above equation (23) is calculated as an acceleration correction amount a _Yoffc in acceleration correction amount a _Xoffc and laterally in the longitudinal direction (step S41).

The current correction obtained in this manner is
Quantity a_xoffc, A_yoffcIs the previous correction amount a_{x off}, A_yoffSet to
The lever routine ends, and the process proceeds to step S29 in FIG.
Step S42). In step S29 in FIG.
The sample for acceleration correction amount calculation is averaged by S28.
Since the processing has already been completed at the point of time p,
Data M _Two, V_GTX, V_GTy, V_ETx, v_ETyReset to “0”
Cut.

Further, the current vehicle speed values v _GTX , v _GTy , v _ETx , v _ETy obtained in step S25 are _compared with the previous vehicle speeds v _GTTX , v
_GTTy , v _ETTx , and v _ETTy are set (step S30), and the current coordinate values X ₂ and Y ₂ are set to the previous values X ₁ and Y ₁ , and the parameter M ₂ is set to the previous value M ₁ Then, the routine ends, and the process returns to step S18 in FIG. 3 (step S31).

In the above embodiment, the simple average of the vehicle speed
Values, but instead a moving average is calculated,
It is also possible to speed up the acceleration correction cycle.
This will be described below with reference to FIGS. 9 and 10. First, substitute for Figure 2
In the main routine of FIG. 9, step S3 is replaced.
In step S3 ', the parameter M₁
Is not used and v_GTTX, V _GTTy, V_ETTx, v_ETTyOf
Instead, v_{GTX1,2,… p}, v_{GTy1,2,… p}, V_{ETx1,2,… p}, v
_{ETy 1,2,… p}Are initialized to “0” by using.

In the vehicle speed moving average calculation routine of FIG. 10 instead of FIG. 4, steps S32 to S34 are executed instead of steps S25 to S27 and S30. That is, in steps S32 and S33 instead of steps S25, S26, and S30, the oldest data is discarded using four sets of p memory stuffs (not shown) prepared in the arithmetic unit 63, and the latest data is discarded. I'm taking it in. Therefore, steps S21, S29 and S30 become unnecessary. Then, instead of step S27, step S34
Is used to obtain p moving average values.

[0083]

As described above, according to the actual vehicle speed estimating apparatus according to the present invention, a GPS incorporating a magnetic azimuth sensor is provided.
From the output information, the vehicle speed on the vehicle coordinate axis is obtained, the estimated vehicle speed on the coordinate axis is obtained from each longitudinal / lateral acceleration sensor and the yaw rate sensor, the average value of both vehicle speeds is obtained, and the output of each acceleration sensor is obtained based on the difference between both average values. Is corrected to estimate and calculate the actual vehicle speed in the lateral direction and the actual vehicle speed in the front-rear direction, so that the influence of the gravitational acceleration and the yaw rate by the acceleration sensor can be removed, and it is possible to cope with a transient state. It is possible to realize a device that is responsive and relatively inexpensive.

[Brief description of the drawings]

FIG. 1 is a block diagram showing one embodiment of an actual vehicle speed estimation device according to the present invention.

FIG. 2 is a flowchart illustrating a main routine of a program executed by the actual vehicle speed estimation device according to the present invention.

FIG. 3 is a flowchart showing details of an actual vehicle speed calculation routine shown in FIG. 2;

FIG. 4 is a flowchart illustrating details of a vehicle speed average calculation routine illustrated in FIG. 3;

FIG. 5 is a flowchart showing details of an acceleration correction amount calculation routine shown in FIG. 4;

FIG. 6 is a diagram illustrating a state in which the vehicle is moving on the XY plane while rotating at a yaw rate r.

FIG. 7 is a graph for explaining a slight change of the vehicle in FIG. 6;

FIG. 8 is a diagram for explaining the effect of gravitational acceleration when obtaining a vehicle speed.

FIG. 9 is a flowchart illustrating a main routine of a program when a moving average value is used as an average vehicle speed instead of FIG. 2;

FIG. 10 is a flowchart showing details when performing a vehicle speed moving average calculation instead of FIG. 4;

[Explanation of symbols]

 1 Ignition switch (SW) 2 Longitudinal acceleration sensor 3 Lateral acceleration sensor 4 Yaw rate sensor 5 GPS 6 Calculation means 61 Low-pass filter (LPF) 62 A / D conversion section 63 Actual vehicle speed estimation calculation section 64 D / A conversion section Same in the figure Symbols indicate the same or corresponding parts.

Continued on the front page F term (reference) 2F029 AA02 AB07 3D044 AA01 AA31 AB01 AC00 AC26 AC28 AC55 AD14 AE01 AE04 AE18 3G093 BA23 BA27 CB10 DB00 DB05 DB18 FA02 FA11

Claims (3)

[Claims] 1. A longitudinal acceleration sensor, a lateral acceleration sensor,
GP with built-in yaw rate sensor and magnetic direction sensor
S and the vehicle speed on the vehicle coordinate axis from the output information of the GPS,
The estimated vehicle speed on the vehicle coordinate axis is obtained from the output of each sensor, the average value of the two vehicle speeds is obtained, and the longitudinal acceleration and the lateral acceleration are corrected based on the difference between the average values to obtain the actual lateral vehicle speed and the actual longitudinal vehicle speed. An actual vehicle speed estimating device, comprising: calculating means for calculating the vehicle speed. 2. The actual vehicle speed estimating device according to claim 1, wherein said calculating means calculates the average value at the timing of the GPS output information. 3. The actual vehicle speed estimation device according to claim 1, wherein the average value is a moving average value.