JP4180784B2 - Brake operation related quantity sensor zero correction device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an apparatus for correcting a zero point of a sensor that detects an amount related to a brake operation such as a brake stroke sensor.
[0002]
[Prior art]
For example, in an electric vehicle (EV), a hybrid electric vehicle (HEV), a regenerative brake device, and an electronically controlled brake device that electronically controls a normal mechanical brake, the regenerative brake and the mechanical brake are electronically controlled based on the amount of brake operation. It is necessary to accurately grasp the amount of brake operation.
[0003]
There is a brake position sensor (BPS) as means for detecting the brake operation amount. The brake position sensor outputs a signal (for example, a voltage value or a current value) corresponding to the brake depression amount.
[0004]
[Problems to be solved by the invention]
By the way, the above brake position sensor outputs a signal corresponding to the position of the brake pedal. However, the position when the brake pedal is not operated (fully closed position, zero point) due to an assembly error of the brake pedal and product variations of the BPS body. The BPS output value varies, and there is a problem that control based on the BPS output value cannot be performed accurately.
[0005]
As a device for correcting the zero point of a sensor that detects an amount related to a brake operation, for example, Japanese Patent Laid-Open No. 8-268243 discloses a zero point correcting device for a brake sensor. For example, in the embodiment of FIG. 1 in the publication, there is disclosed a technique for correcting a zero point of a brake sensor using a sensor output when a stop switch for detecting presence / absence of a brake operation detects non-operation of a brake. . Further, in the embodiment of FIG. 15 in the publication, there is disclosed a technique for correcting the zero point by assuming that the brake is not operated when the sensor output is in a predetermined region.
[0006]
However, in the embodiment of FIG. 1 of the above publication, since there is a variation in the operating range of the stop switch, the zero point correction is easily affected by this variation, and the brake fully open state cannot be reliably detected, and high accuracy is expected. There is a problem that you can not. Further, in the embodiment of FIG. 15 of the above publication, since the brake output is always considered as the brake non-operating state if the sensor output is within a predetermined region, the brake fully open state cannot be reliably detected, and high accuracy is expected. Can not. In view of this point, in the above publication, the correction value is regulated within a predetermined range and the abnormal correction value is excluded so as to ensure the accuracy. There is room for improvement.
[0007]
The present invention has been devised in view of such problems, and it is possible to correct the zero point of a sensor for detecting an amount related to a brake operation with high accuracy, and to correct the zero point of a brake operation-related amount sensor. An object is to provide an apparatus.
[0008]
[Means for Solving the Problems]
In order to achieve the above goal, the zero correction device for a brake operation related amount sensor according to the present invention corresponds to a brake operation amount of a vehicle by a brake operation related amount sensor (for example, a brake stroke sensor including a hydraulic pressure sensor). The state quantity is detected, and the state where the detection output of the sensor is within a predetermined range corresponding to the vicinity of the brake opening end and the variation of the detection output of the sensor is within the predetermined range is continued for a predetermined time or longer by the zero learning means. Based on the detection output of the sensor under the condition, the zero-corresponding output of the sensor is learned, and the output of the sensor is corrected based on the zero-corresponding output by the sensor output correcting means. Item 1).
[0009]
As a result, if the detection output of the brake operation-related quantity sensor is within a predetermined range corresponding to the vicinity of the brake release end, it can be considered that the brake is almost released, and in this state the fluctuation of the sensor output is within the predetermined range. If this state continues for a predetermined time or more, it can be determined that the brake is completely released. And since the sensor output under such conditions can be judged as the output when the brake is fully released, if the sensor zero-point corresponding output is learned based on this, the zero point can be learned with high accuracy, and the sensor output can be corrected with high accuracy. it can.
[0010]
In addition, since the zero point is corrected based on the output of the sensor itself without using a brake switch, one of the conditions is that the fluctuation of the sensor output is within a predetermined range without being affected by variations in the brake switch. As a result, mislearning due to noisy signals is not caused.
[0011]
The state in which the variation in the detection output of the sensor is within a predetermined range is a state in which the detection output of the sensor is in a minute range preset with respect to a value for determining the zero point of the sensor. It is preferable that it is present (claim 2).
The zero point learning means preferably is configured to learn a zero point corresponding output on the basis of the detection output of most brake open end closer under the condition (claim 3).
As described above, the zero point can be learned with higher accuracy by learning the zero point corresponding output based on the sensor output closest to the brake opening end in a state where it can be specified that the brake is completely released. .
[0012]
The zero learning means updates the learning value to the sensor output when the sensor output used for the learning is on the brake release side from the stored learning value, and to the brake operation side of the learning value if the learning operation is on the brake operation side. It is preferable to limit the renewal amount within a predetermined amount (claim 4 ).
In this way, while learning to the brake release side is performed quickly, the amount of brake operation based on the zero point calculated after learning can be quickly increased by limiting the learning to the brake operation side. A sudden decline is prevented. In other words, the zero point of the brake operation related amount sensor is quickly learned and corrected for the side where the brake is strengthened and carefully learned and corrected for the side where the brake is weakened.
[0013]
Further, the brake operation detecting means for detecting the presence or absence of the brake operation and the learning by the zero learning means is not completed, the brake is in the non-operating state, and the detection output of the sensor corresponds to the vicinity of the brake open end. when within a predetermined range, it is preferable to further comprising an initial learning means for learning the zero point above in accordance with a detection output of the sensor (claim 5).
[0014]
This makes it possible to correct the zeros by the initial learning means even when the zero learning means is not easily executed because a stable output cannot be obtained immediately, preventing learning delays immediately after initialization. can do.
[0015]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
1 and 2 show a zero correction apparatus for a brake operation related quantity sensor according to an embodiment of the present invention. FIG. 1 is a schematic block diagram thereof, and FIG. 2 is a flowchart for explaining the zero correction.
[0016]
[Device configuration]
As shown in FIG. 1, this zero point correction device corrects the zero point (fully closed opening) of a brake stroke sensor (BPS) 1 as a brake operation related quantity sensor, both of which are electronic control units (ECUs) 10. The initial learning means 11, the zero point learning means 12, and the sensor output correction means 13 are provided. In BPS1, the brake operation amount (depression amount of the brake pedal 2) is represented by the magnitude of the voltage value, and a larger voltage value is output as the brake operation amount (brake pedal depression amount) is larger.
[0017]
The zero point learning (also referred to as fully closed opening learning) of the BPS1 by the ECU 10 is performed by learning the value of the A / D conversion value of the BPS opening when the BPS1 is fully closed in the zero point learning means 12. It has become. The learning value obtained in this way is backed up by the battery 4 being connected, but is initialized when the battery 4 is not backed up due to the battery 4 being removed, etc., and learning at the time of BPS fully closed determination Until this is done, initial learning means 11 performs initial learning.
[0018]
[Initial learning]
The initial value of BPS fully closed opening learning used at the time of initialization described above is the average value of the zero point detection output of BPS1 or the brake pedal depression decrease by a predetermined amount in consideration of the product variation of BPS1. Set to the value shifted to the side. For example, assuming that the detected output of BPS1 is a voltage and the detected output of BPS1 increases when the brake pedal is depressed, and the average value of the zero point detected output of _BPS1 is V _0AVE , The initial value of learning of the _opening degree is set to the average value V _0AVE [V] or a value (V _{0AVE −α1} ) [V] on the _depressing side by a minute amount α1 from the average value V _0AVE .
[0019]
The initial learning means 11 performs initial learning after initialization until learning at the time of BPS full-close determination is performed, and the following conditions are provided as initial learning conditions.
(1) The engine is not in the engine stall or start mode. This is intended to perform stable learning by eliminating the time of voltage fluctuation.
(2) The battery voltage Vb is higher than the reference value. This also eliminates the case where a sufficient voltage cannot be obtained and tries to perform stable learning.
[0020]
(3) The brake switch (brake operation detecting means for detecting whether or not the brake is operated) 3 is off, that is, the brake is not operated. This is because the on / off information of the brake switch 3 has a large error and low reliability, but if the brake switch 3 is off, it is certain that at least the brake is not operated, and learning at the BPS full-close determination is performed. This is because it is important information at the time of simple initial learning.
[0021]
{Circle around (4)} After the initialization, until the learning for the BPS fully closed determination is performed. This is because the initial learning is not necessary if the learning at the BPS full-close determination is performed.
(5) The A / D conversion value θAD of the BPS opening value is within a predetermined range in the vicinity of the BPS fully closed opening degree (the following expression is satisfied).
BPS full-closed opening learning lower limit value ≦ θAD ≦ BPS full-closed opening learning upper limit value This condition is provided to prevent erroneous learning due to a noise-like signal.
[0022]
The initial learning means 11 performs initial learning at a predetermined cycle when all the above conditions (1) to (5) are satisfied.
In the initial learning by the initial learning means 11, the BPS fully closed opening learning fully closed position θIDL is predetermined based on the A / D conversion value (this is referred to as the BPS opening A / D converted value) θAD of the detection signal of BPS1. Set (update) in the cycle. The sensor output correction means 13 corrects the zero point (fully closed opening) based on the BPS fully closed opening learning fully closed position θIDL during initial learning.
[0023]
In other words, the initial learning means 11 compares the BPS opening A / D conversion value θAD with the BPS fully closed opening learning fully closed position θIDL in a predetermined cycle, and according to the magnitude relationship between them, as follows: BPS fully closed opening learning fully closed position θIDL is updated.
(1) θIDL <θAD θIDL = θIDL
That is, the previous θIDL is held.
(2) When θIDL ≧ θAD θIDL = θIDL−α ≧ θAD
That is, the previous θIDL is decreased by a predetermined initial learning update width holding width α. α is a minute amount (α1> α).
[0024]
As described above, in the initial learning, the BPS fully closed opening learning fully closed position θIDL is learned and updated only on the decrease side (brake release side) in consideration of safety related to vehicle control in which this initial learning is reflected. It is. In other words, the original fully closed learning (zero learning), which will be described later, accurately grasps the fully closed opening (zero point), but the initial learning is a rough learning performed based on the on / off information of the brake 3. Therefore, the accuracy is not necessarily high.
[0025]
That is, the on / off signal of the brake switch 3 that operates based on the mechanical operation is not always accurate, so it is desirable to perform learning update carefully. If the fully closed position (zero point) is corrected to the open side, the brake signal will be grasped higher, which is a safe side. This is allowed, and if the fully closed position (zero point) is corrected to the closed side, the brake signal is lowered. This is forbidden because it is not safe to grasp.
[0026]
[Zero learning]
Next, the zero point learning by the zero point learning means 12 will be described. The following conditions are provided as learning conditions for zero point learning (fully closed opening degree learning).
(1) The engine is not in the engine stall or start mode. This is intended to perform stable learning by eliminating the time of voltage fluctuation.
(2) The battery voltage Vb is higher than the reference value. This also eliminates the case where a sufficient voltage cannot be obtained and tries to perform stable learning.
(3) The A / D conversion value θAD of the BPS opening value is within a predetermined range in the vicinity of the BPS fully closed opening degree (the following equation is satisfied).
[0027]
BPS full-closed opening learning lower limit value ≦ θAD ≦ BPS full-closed opening learning upper limit value This condition is provided to prevent erroneous learning due to a noise-like signal.
(4) θC ≦ θAD <θC + β was maintained for a predetermined time Ts.
Note that θC is a BPS opening for determining whether BPS is fully closed (BPS full closing determination BPS opening). The BPS full-closed determination BPS opening θC is set to θC = θAD when the BPS full-closed determination condition is not satisfied, and otherwise the BPS full-closed determination BPS opening θC is not updated. Β is a minute amount (here, α> β). Also, Ts is about several seconds (or less). This condition is based on the condition that the BPS opening A / D conversion value θ AD is stable for a predetermined period, but is provided to ensure the reliability of learning.
[0028]
In the initial learning by the zero point learning means 12, zero point learning (fully closed opening degree learning) is performed when each of the above conditions (1) to (4) is satisfied.
In zero learning by the zero learning means 12, the BPS fully closed opening learning fully closed position θIDL is set (updated) in accordance with the BPS fully closed determination BPS opening θC set based on the BPS opening A / D conversion value θAD. ) The sensor output correcting means 13 corrects the zero point (fully closed opening) based on this BPS fully closed opening learning fully closed position θIDL during zero learning.
[0029]
In other words, the zero point learning means 12 compares the BPS fully closed determination BPS opening θC when the zero point learning condition is satisfied with the BPS fully closed opening learning fully closed position θIDL, and according to the magnitude relationship between them as follows. BPS fully closed opening learning fully closed position θIDL is updated.
(1) When θIDL ≧ θC θIDL = θC
The update of θIDL = θC is updated every time Ts used for the zero point learning condition if the zero point learning condition is continuously satisfied.
(2) When θIDL <θC θIDL = min (θIDL + γ, θC)
That is, in the case of θIDL <θC, the smaller one of θIDL + γ and θC is updated to the BPS fully closed opening learning fully closed position θIDL. This update is performed only once every time the condition that θAD is equal to or smaller than the fully closed opening learning upper limit (θAD ≦ full closing opening learning upper limit) is satisfied. Note that γ is set to a minute amount similar to β.
[0030]
[BPS learning correction]
Next, the BPS learning correction by the sensor output correcting means 13 will be described. Based on the BPS fully closed opening learning fully closed position θIDL obtained by the above learning, the sensor output (A / D conversion value θAD) is corrected to zero (fully closed opening) to obtain BPS opening θBPS with learning correction.
θBPS = θAD−θBPSL
ΘBPSL is a BPS opening fully closed position learning value that satisfies the following conditions, and is backed up by a battery.
θBPSL = θIDL Fully closed reference position
BPS learning value lower limit ≤ θBPSL ≤ BPS learning value upper limit [0031]
[Device operation]
Since the zero correction device for a brake operation related quantity sensor as one embodiment of the present invention is configured as described above, for example, as shown in the flowchart of FIG. Correction) Note that (1) the engine is not in the engine start or start mode, and (2) the condition that the battery voltage Vb is equal to or higher than the reference value is generally satisfied, these determination steps are omitted in FIG. Actually, both conditions (1) and (2) are determined as learning conditions.
[0032]
That is, first, it is determined whether or not there is battery backup data (step S10), and if there is no battery backup data, initialization (including initial value setting of BPS fully closed opening learning) is performed (step S20). Next, the A / D conversion value θad of the BPS1 sensor output is read (step S30), and it is determined whether or not the fully closed learning has been completed (step S40). Here, if the fully closed learning has not been completed. The initial learning condition {circle around (4)} is satisfied until the learning is performed at the time of the BPS full-close determination after initialization, and the initial learning condition is determined.
[0033]
Here, it is assumed that (1) the engine is not in the engine start or start mode, and (2) the battery voltage Vb is equal to or higher than the reference value. (4) BPS fully closed after initialization (3) Determination of whether the brake switch 3 is OFF (step S50), and (5) A / D conversion value of the BPS opening value. It is determined whether θAD satisfies “BPS fully closed opening learning lower limit value ≦ θAD ≦ BPS fully closed opening learning upper limit value” (step S60).
[0034]
Here, if the conditions (3) and (5) are satisfied in step S50, the initial learning condition is satisfied, and the initial learning in steps S70 to S90 is performed. That is, it is determined whether or not θIDL <θAD (step S70). If θIDL <θAD, θIDL = θIDL is set (step S80). If θIDL ≧ θAD, θIDL = θIDL−α is set (step S90).
On the other hand, if the learning at the time of full closure has been completed (step S40), if the brake 3 is not turned off, the learning condition for zero learning (3) The A / D conversion value θAD of the BPS opening value is “BPS full-closed opening / closing”. It is determined whether or not “degree learning lower limit value ≦ θAD ≦ BPS fully closed opening learning upper limit value” is satisfied (step S100). Here, when the learning condition (3) for zero learning is satisfied and when the initial learning is performed (the learning condition (3) for zero learning is satisfied in step S60), (4) θC ≦ θAD It is determined whether <θC + β is maintained for a predetermined time Ts (step S110).
[0035]
Here, if the learning condition (4) for zero learning is satisfied, the learning condition (1) (2) for zero learning is satisfied as a premise, and the learning condition (3) for zero learning is satisfied ( In step S100), all of the learning conditions for zero learning are satisfied, so the zero learning in steps S120 to S140 is performed.
That is, the BPS fully closed determination BPS opening θC and the BPS fully closed opening learning fully closed position θIDL when the zero learning condition is satisfied are compared, and the BPS fully closed opening learning is performed as follows according to the magnitude relationship between them. The fully closed position θIDL is updated. The BPS fully closed determination BPS opening θC is updated to θC = θAD when the BPS fully closed determination condition is not satisfied (step S170).
[0036]
That is, θIDL <θC is determined (step S120). If θIDL <θC is not satisfied (in the case of θIDL ≧ θC), θIDL = θC is set (step S140). If it is established, it is updated every time Ts used for the zero learning condition.
On the other hand, if θIDL <θC, θIDL = min (θIDL + γ, θC). That is, in the case of θIDL <θC, the smaller one of θIDL + γ and θC is updated to the BPS fully closed opening learning fully closed position θIDL. This update is performed only once every time the condition that θAD is equal to or smaller than the fully closed opening learning upper limit (θAD ≦ full closing opening learning upper limit) is satisfied.
[0037]
Thus, based on the BPS fully closed opening learning fully closed position θIDL obtained by learning, a BPS opening fully closed position learning value θBPSL (= θIDL one fully closed reference position) is calculated (step S150), A BPS opening θBPS with learning correction (= θAD−θBPSL) is calculated (step S160).
Therefore, according to the zero point correction device for the brake operation related quantity sensor, the zero point of BPS1 can be learned with high accuracy and the output of BPS1 can be corrected with high accuracy, so that control related to brake operation can be performed with high accuracy. It becomes like this.
[0038]
In the zero learning that is the main learning, the brake switch is not used and the zero is corrected based on the output of the BPS 1 itself, so that it is not affected by variations in the brake switch.
In addition, since one of the conditions is that the fluctuation of the sensor output is within a predetermined range (zero learning condition (3)), there is no mislearning due to a noisy signal and high accuracy and reliability are achieved with a simple configuration. Correction can be realized.
[0039]
In addition, since the zero point corresponding output is learned based on the detection output closest to the brake opening end (minimum output voltage value), the zero point can be learned with higher accuracy, and control related to brake operation can be performed with higher accuracy. Can be implemented.
In particular, when the output value (A / D conversion value) θC from BPS1 is on the brake release side (θIDL ≧ θC) from the stored learned value θIDL, the learned value θIDL is updated to the sensor output value θC, and the sensor output value θC is In the case of the brake operation side (θIDL <θC) from the learning value θIDL, the update amount of the learning value to the brake operation side is limited to a predetermined amount (a minute amount) γ. While the learning correction is promptly performed, the learning correction is carefully performed on the weakening side of the brake, and a learning effect in which the safety for the brake operation is emphasized is obtained.
[0040]
In addition, the initial learning means 11 can correct the zero by learning even in a state where the stable output cannot be obtained immediately (the zero learning condition (4) is not satisfied) and the learning by the zero learning means is not easily executed. Thus, it is possible to prevent a learning delay immediately after initialization, and to improve the control accuracy of the brake operation immediately after initialization.
[0041]
[Others]
The present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the spirit of the present invention.
For example, as a brake operation related quantity sensor, in addition to the brake stroke sensor, a quantity related to other brake operations, such as a fluid pressure sensor (hydraulic pressure sensor) for detecting fluid pressure (hydraulic pressure) such as brake fluid pressure, is detected. Widely applicable to sensors. The brake operation detecting means is not limited to a brake switch as long as it can detect whether or not the brake is operated, but an increase in cost can be suppressed by using an existing brake switch.
[0042]
【The invention's effect】
As described above, according to the zero correction device of the brake operation related quantity sensor of the present invention, the zero point of the brake operation related quantity sensor can be learned with high accuracy and the output of the sensor can be corrected with high accuracy. Such control can be performed with high accuracy.
[0043]
In addition, since the zero point is corrected based on the output of the sensor itself without using the brake switch, it is not affected by variations in the brake switch, and the fluctuation of the sensor output is within a predetermined range. Since it is one, there is an advantage that high-accuracy and high-reliability correction can be realized with a simple configuration without causing erroneous learning due to a noisy signal.
[0044]
If the zero point learning means is configured to learn the zero point corresponding output based on the detection output closest to the brake opening end under the above-described state, the zero point can be learned with higher accuracy, and control relating to brake operation can be performed. It becomes possible to carry out with higher accuracy (Claim 3 ).
When the sensor output used for the learning is on the brake release side from the stored learning value, the learning value is updated to the sensor output by the zero point learning means, and to the brake operation side of the learning value if the learning operation is on the brake operation side. If the renewal amount of the brake is limited to a predetermined amount, the zero point of the brake operation related amount sensor is quickly learned and corrected when the brake is strengthened, and carefully learned and corrected when the brake is weakened. Thus, it is possible to control the brake operation with emphasis on safety (claim 4 ).
[0045]
Further, the brake operation detecting means for detecting the presence or absence of the brake operation and the learning by the zero learning means is not completed, the brake is in the non-operating state, and the detection output of the sensor corresponds to the vicinity of the brake open end. When it is within the predetermined range, by further comprising an initial learning means for learning the zero point according to the detection output of the sensor, a stable state cannot be obtained immediately and learning by the zero point learning means Even in a situation where it is difficult to execute, the zero learning can be corrected by the initial learning means, a learning delay immediately after the initialization can be prevented, and the control accuracy for the brake operation can be improved immediately after the initialization. (Claim 5 ).
[Brief description of the drawings]
FIG. 1 is a schematic block diagram showing a configuration of a zero point correction device for a brake operation related amount sensor according to an embodiment of the present invention.
FIG. 2 is a flowchart illustrating zero point correction by a zero point correction device for a brake operation related amount sensor according to an embodiment of the present invention.
[Explanation of symbols]
1 Brake stroke sensor (BPS)
2 Brake pedal 3 Brake switch (Brake operation detection means)
11 Initial learning means 12 Zero learning means 13 Sensor output correction means

Claims (5)

A brake operation related quantity sensor for detecting a state quantity corresponding to the brake operation quantity of the vehicle;
Based on the detection output of the sensor in a state where the detection output of the sensor is within a predetermined range corresponding to the vicinity of the brake opening end and the variation of the detection output of the sensor is within the predetermined range for a predetermined time or more. , Zero learning means for learning the zero corresponding output of the sensor,
Sensor output correction means for correcting the output of the sensor on the basis of the output corresponding to the zero point, a zero point correction device for a brake operation related quantity sensor. The state in which the variation in the detection output of the sensor is within a predetermined range is a state in which the detection output of the sensor is in a minute range preset with respect to a value for determining the zero point of the sensor. is there
The zero point correction device for a brake operation related quantity sensor according to claim 1, wherein: 3. The brake according to claim 1, wherein the zero point learning means is configured to learn the zero point corresponding output based on a detection output closest to the brake opening end of the sensor under the state. Zero correction device for operation related quantity sensor. The zero-point learning means updates the learning value to the sensor output when the sensor output used for the learning is on the brake release side with respect to the stored learning value that has already been stored, and on the brake operation side with respect to the stored learning value. 4. The zero correction apparatus for a brake operation related amount sensor according to claim 1, wherein the amount of update of the learned value to the brake operation side is limited to a predetermined amount. Brake operation detecting means for detecting presence or absence of brake operation;
When the learning by the zero point learning means is not completed and the brake is in a non-operating state and the detection output of the sensor is within a predetermined range corresponding to the vicinity of the brake opening end, the detection output of the sensor is The zero correction device for a brake operation related quantity sensor according to any one of claims 1 to 4 , further comprising initial learning means for learning the zero in response.

Images