JP2007237978A - Brake controller - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a brake controller capable of accomplishing a good brake feeling. <P>SOLUTION: On the basis of the feedforward control law and the feedback control law, the brake controller corrects the target deceleration GxRef calculated on the basis of the pedal stroke ST and the master cylinder pressure PMC. Because in particular, an efficacy measurement correction term learnt on the basis of the brake pad condition is used to the feedforward term KFF, this involves less influence such as secular deterioration etc. of constituent elements of the brake device during the feedback control. Accordingly hunting of the control amount can be precluded or suppressed. Also the feedback control is executed continuously while the feedforward term KFF is taken into account, so that the followup performance of the control can be held. As a result, even if there is turbulence resulting from a secular deterioration etc. of any constituent element(s) of the brake device, it is possible to hold a good brake feeling while the stability of the control system is well maintained. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a brake control device that controls braking force applied to wheels provided in a vehicle.

  2. Description of the Related Art Conventionally, a brake control device that calculates a target deceleration based on a hydraulic pressure in a master cylinder and a pedal stroke of a brake pedal and controls a braking force so that the target deceleration is applied to the vehicle is known. (For example, refer to Patent Document 1).

Such a brake control device sends the brake fluid pressurized according to the operation amount of the brake pedal toward each wheel cylinder. In this brake control, the deviation between the required deceleration based on the brake pedal operation amount and the actual deceleration (hereinafter referred to as “actual deceleration”) is calculated sequentially, and the braking control amount is determined so as to eliminate the deviation. The feedback control is executed.
JP-A-62-18359

  However, components such as brake pads that brake each wheel originally have variations among individuals, and each component also deteriorates over time. For this reason, if it is attempted to compensate for the disturbance to the braking effect due to such variation among individuals or deterioration over time by only feedback control, the control system becomes unstable. That is, if the feedback gain is increased so as to improve the followability of the control, control hunting is likely to occur, resulting in a problem that the brake feeling is lowered.

  The present invention has been made in view of such problems, and realizes a good brake feeling while maintaining the stability of the control system even when there is a disturbance due to aging deterioration of the components of the brake device. An object of the present invention is to provide a brake control device that can be used.

  In order to solve the above problems, a brake control device according to an aspect of the present invention controls a braking force applied to a wheel by controlling a hydraulic pressure supplied to a wheel cylinder. This brake control device is based on target deceleration calculation means for calculating a target deceleration of a vehicle based on operation of a brake operation member, and target deceleration calculated by the target deceleration calculation means based on at least a braking effectiveness state. The target deceleration correction means for correcting using both the feedforward term learned in this manner and the feedback term based on the deviation between the target deceleration taking into account the feedforward term and the actual deceleration, and the target deceleration correction means Based on the target deceleration corrected by the above, the target hydraulic pressure calculation means for calculating the target hydraulic pressure to be supplied to the wheel cylinder, and the target hydraulic pressure calculated by the target hydraulic pressure calculation means is supplied to the wheel cylinder. Hydraulic control means for controlling the hydraulic pressure.

  According to this aspect, the target deceleration calculated based on the operation amount of the brake operation member is corrected based on the feedforward control law and the feedback control law, and the target hydraulic pressure that realizes the corrected target deceleration is obtained. Calculated. The hydraulic pressure supplied to the wheel cylinder is controlled to be the target hydraulic pressure. Here, since the feedforward term learned based on at least the effective state of the brake is used for the calculation of the feedback term, the correction amount by the feedback control can be reduced. That is, the influence of aging deterioration of the components of the brake device in feedback control is reduced. Therefore, control hunting can be prevented or suppressed. In addition, since feedback control is continuously executed while taking the feedforward term into account, it is possible to maintain control followability. As a result, the stability of the control system can be maintained and a good brake feeling can be maintained even when there is a disturbance due to aging deterioration of the components of the brake device.

  The target deceleration correction means may calculate a correction term that compensates for a change in the effectiveness of the brake based on the state of the components of the brake device including at least the brake pad as the feedforward term. The “components of the brake device” referred to here may include not only the brake pads but also peripheral components of the brake pads, and the like, which change the state of the brake due to some factor, for example.

  According to this aspect, for example, when the braking effectiveness of any wheel decreases due to the progress of aging of any of the components of the brake device, a correction term that compensates for the decrease in the effectiveness is calculated. Learned as a feedforward term. For this reason, the control amount is raised beforehand by the feedforward term, and the feedback gain of the feedback control can be relatively reduced to maintain its stability.

  More specifically, the target deceleration correction means may sequentially update the correction term learned based on the state of the components of the brake device and the correction term based on the weight of the vehicle as the feedforward term. Good. That is, changes in the state of the brake pads and the like due to variations among individual brake pads and aging, etc., affect the effectiveness of the brakes over the long term. On the other hand, changes in the number of passengers and load weight of the vehicle temporarily change the inertial force when the vehicle travels, and thus have a short-term effect on the effectiveness of the brake. Here, the influence on the effectiveness of both brakes is taken into consideration, and learning is performed as a correction term that constitutes a feedforward term. By updating each correction term sequentially, a more appropriate feedforward term can be used.

  However, as will be described later in the embodiment, the braking effectiveness described above is dependent on the vehicle speed, and depending on the magnitude of the fluctuation of the deceleration, it may not be able to be measured properly due to the influence of the transient characteristics.

  Therefore, the target deceleration correction means may update the feed-forward term by performing learning based on the braking effectiveness state only when a preset measurement effective condition is satisfied. The “measurement effective condition” here means a condition for appropriately measuring the braking effectiveness state, and may include the above-described conditions based on the magnitude of fluctuations in vehicle speed and deceleration. It is also possible to set based on other determination indicators that hinder proper measurement of.

  In addition, in order for the feedforward term to function effectively, long-term learning may be required to some extent. For example, when a vehicle is shipped from a factory, there is no deterioration over time even if there are variations among individual brake pads. For this reason, there is a case where the followability to the control of the feedforward term learned based on the effective state of the brake cannot be obtained satisfactorily. In such a case, it is preferable to increase the feedback gain in order to ensure control followability.

  Therefore, the target deceleration correction means may increase the feedback gain by a predetermined amount until learning based on the braking effectiveness state is performed a predetermined number of times, and decrease the feedback gain when the learning exceeds the predetermined number of times.

  According to the brake control device of the present invention, it is possible to achieve a good brake feeling while maintaining the stability of the control system.

  The best mode for carrying out the present invention will be described below in detail with reference to the drawings.

[First Embodiment]
FIG. 1 is a system diagram showing a brake control device according to a first embodiment of the present invention. The brake control device 10 shown in the figure constitutes an electronically controlled brake system for a vehicle, and independently controls the four-wheel brakes of the vehicle based on the amount of operation of the brake pedal 12 as a brake operation member by the driver. And optimal control.

  The brake pedal 12 is connected to a master cylinder 14 that sends out brake fluid as a working fluid in response to a depression operation by a driver. The brake pedal 12 is provided with a stroke sensor 46 for detecting the depression stroke. One output port of the master cylinder 14 is connected to a stroke simulator 24 that creates a reaction force according to the operating force of the brake pedal 12 by the driver. A simulator cut valve 23 is provided in the middle of the flow path connecting the master cylinder 14 and the stroke simulator 24. The simulator cut valve 23 is a normally closed electromagnetic on-off valve that is closed when not energized and is switched to an open state when an operation of the brake pedal 12 by the driver is detected. The master cylinder 14 is connected to a reservoir tank 26 for storing brake fluid.

  A brake hydraulic pressure control pipe 16 for the right front wheel is connected to one output port of the master cylinder 14, and the brake hydraulic pressure control pipe 16 applies a braking force to the right front wheel (not shown). It is connected to the cylinder 20FR. A brake hydraulic pressure control pipe 18 for the left front wheel is connected to the other output port of the master cylinder 14, and the brake hydraulic pressure control pipe 18 is for the left front wheel that applies a braking force to the left front wheel (not shown). Connected to the wheel cylinder 20FL. A right electromagnetic on-off valve 22FR is provided in the middle of the brake hydraulic control pipe 16 for the right front wheel, and a left electromagnetic on-off valve 22FL is provided in the middle of the brake hydraulic control pipe 18 for the left front wheel. The right solenoid on-off valve 22FR and the left solenoid on-off valve 22FL are both normally open solenoid valves that are open when not energized and switched to closed when energized.

  A right master pressure sensor 48FR for detecting the master cylinder pressure on the right front wheel side is provided in the middle of the brake hydraulic control pipe 16 for the right front wheel. A left master pressure sensor 48FL for measuring the master cylinder pressure on the left front wheel side is provided. In the brake control apparatus 10, when the brake pedal 12 is depressed by the driver, the stroke operation amount is detected by the stroke sensor 46. The master detected by the right master pressure sensor 48FR and the left master pressure sensor 48FL is detected. The depressing operation force (depressing force) of the brake pedal 12 can also be obtained from the cylinder pressure. As described above, it is preferable from the viewpoint of fail-safe that the master cylinder pressure is monitored by the two pressure sensors 48FR and 48FL on the assumption of the failure of the stroke sensor 46. Hereinafter, the right master pressure sensor 48FR and the left master pressure sensor 48FL are collectively referred to as a master cylinder pressure sensor 48 as appropriate.

  On the other hand, one end of a hydraulic supply / discharge pipe 28 is connected to the reservoir tank 26, and a suction port of an oil pump 34 driven by a motor 32 is connected to the other end of the hydraulic supply / discharge pipe 28. . The discharge port of the oil pump 34 is connected to a high pressure pipe 30, and an accumulator 50 and a relief valve 53 are connected to the high pressure pipe 30. In the present embodiment, a reciprocating pump having two or more pistons (not shown) that are reciprocally moved by the motor 32 is employed as the oil pump 34. Further, as the accumulator 50, an accumulator 50 that converts the pressure energy of the brake fluid into the pressure energy of an enclosed gas such as nitrogen is stored.

  The accumulator 50 stores the brake fluid that has been pressurized to, for example, about 14 to 22 MPa by the oil pump 34. Further, the valve outlet of the relief valve 53 is connected to the hydraulic supply / discharge pipe 28. When the pressure of the brake fluid in the accumulator 50 is abnormally increased to about 25 MPa, for example, the relief valve 53 is opened and the high pressure brake is opened. The fluid is returned to the hydraulic supply / discharge pipe 28. Further, the high-pressure pipe 30 is provided with an accumulator pressure sensor 51 that detects the outlet pressure of the accumulator 50, that is, the pressure of the brake fluid in the accumulator 50.

  The high pressure pipe 30 is connected to the right front wheel wheel cylinder 20FR, the left front wheel wheel cylinder 20FL, the right rear wheel wheel cylinder 20RR, and the left rear wheel through the pressure increasing valves 40FR, 40FL, 40RR, 40RL. It is connected to the cylinder 20RL. Hereinafter, the wheel cylinders 20FR to 20RL will be collectively referred to as “wheel cylinders 20”, and the pressure increase valves 40FR to 40RL will be appropriately collectively referred to as “pressure increase valves 40”. Each of the pressure increasing valves 40 is a normally closed electromagnetic flow control valve (linear valve) that is closed when not energized and is used to increase the pressure of the wheel cylinder 20 as necessary. A disc brake unit is provided for each wheel of the vehicle (not shown), and each disc brake unit generates a braking force by pressing the brake pad against the disc by the action of the wheel cylinder 20.

  Further, the wheel cylinder 20FR for the right front wheel and the wheel cylinder 20FL for the left front wheel are connected to the hydraulic supply / discharge pipe 28 via the pressure reducing valve 42FR or 42FL, respectively. The pressure reducing valves 42FR and 42FL are normally closed electromagnetic flow control valves (linear valves) used for pressure reduction of the wheel cylinders 20FR and 20FL as necessary. On the other hand, the wheel cylinder 20RR for the right rear wheel and the wheel cylinder 20RL for the left rear wheel are connected to the hydraulic supply / discharge pipe 28 via a pressure reducing valve 42RR or 42RL which is a normally open electromagnetic flow control valve. . Hereinafter, the pressure reducing valves 42FR to 42RL are collectively referred to as “pressure reducing valve 42” as appropriate.

  Wheel cylinders for detecting the wheel cylinder pressure, which is the pressure of the brake fluid acting on the corresponding wheel cylinder 20, in the vicinity of the wheel cylinders 20FR to 20RL for the right front wheel, the left front wheel, the right rear wheel, and the left rear wheel Pressure sensors 44FR, 44FL, 44RR and 44RL are provided. Hereinafter, the wheel cylinder pressure sensors 44FR to 44RL are collectively referred to as “WC pressure sensor 44” as appropriate.

  The right electromagnetic on-off valve 22FR and the left electromagnetic on-off valve 22FL, the pressure increasing valves 40FR to 40RL, the pressure reducing valves 42FR to 42RL, the oil pump 34, the accumulator 50, and the like constitute the hydraulic actuator 80 of the brake control device 10. The hydraulic actuator 80 is controlled by an electronic control unit (hereinafter referred to as “ECU”) 200.

  The ECU 200 is a nonvolatile memory such as a CPU that executes various arithmetic processes, a ROM that stores various control programs, a RAM that is used as a work area for data storage and program execution, and a backup RAM that can retain stored contents even when the engine is stopped. An A / D converter or the like for converting analog signals input from a memory, an input / output interface, various sensors, and the like into digital signals is provided.

  The ECU 200 is electrically connected to various actuators including the hydraulic on / off valves 22FR and 22FL, the simulator cut valve 23, the pressure increasing valves 40FR to 40RL, the pressure reducing valves 42FR to 42RL, and the like. The ECU 200 is electrically connected to various sensors / switches that output signals for use in control. That is, the ECU 200 receives signals indicating the wheel cylinder pressure in the wheel cylinders 20FR to 20RL from the WC pressure sensors 44FR to 44RL. Further, a signal indicating the pedal stroke of the brake pedal 12 is input from the stroke sensor 46 to the ECU 200, and signals indicating the master cylinder pressure are input from the right master pressure sensor 48FR and the left master pressure sensor 48FL, and the accumulator pressure sensor 51 is input. A signal indicating the accumulator pressure is input.

  In the brake control apparatus 10 configured as described above, the ECU 200 calculates the target deceleration of the vehicle from the pedal stroke indicating the depression amount of the brake pedal 12 and the master cylinder pressure, and each of the vehicle decelerations is calculated according to the calculated target deceleration. A target hydraulic pressure that is a target value of the wheel cylinder pressure of the wheel, that is, a target wheel cylinder pressure is obtained. Then, the ECU 200 controls the pressure increasing valve 40 and the pressure reducing valve 42 so that the wheel cylinder pressure of each wheel becomes the target wheel cylinder pressure.

  On the other hand, at this time, the electromagnetic on-off valves 22FR and 22FL are closed, and the simulator cut valve 23 is opened. Therefore, the brake fluid sent from the master cylinder 14 by the depression of the brake pedal 12 by the driver flows into the stroke simulator 24 through the simulator cut valve 23.

  When the accumulator pressure is less than the lower limit value of the preset control range, the oil pump 34 is driven by the ECU 200 to increase the accumulator pressure. When the accumulator pressure enters the control range, the drive of the oil pump 34 is stopped. Is done.

  Next, an outline of the brake control according to the present embodiment will be described.

  First, before the driver turns on the ignition switch, that is, before energization of each solenoid valve, each solenoid valve is in the state shown in FIG. 1 due to the biasing force of the built-in spring. At this time, the brake oil at atmospheric pressure reaches from the master cylinder 14 to the right front wheel and left front wheel cylinders 20FR and 20FL via the right and left electromagnetic on-off valves 22FR and 22FL, respectively. On the other hand, the right rear wheel and the left rear wheel cylinders 20RR and 20RL are also braked at the same atmospheric pressure as the hydraulic pressure in the reservoir tank 26 via the hydraulic supply / discharge pipe 28 and the normally open pressure reducing valves 42RR and 42RL. Oil has reached. At this time, all four wheel cylinder pressures are atmospheric pressure, and no braking force is generated. However, even when the driver does not energize, if the driver depresses the brake pedal 12, the braking force corresponding to the depression force directly acts on the wheel cylinders 20FR and 20FL of the right front wheel and the left front wheel, and the right front wheel and the left A braking force is generated on the front wheels.

  When the driver turns on the ignition switch, the motor 32 operates as necessary, and the accumulator pressure enters the control range. Thereafter, each solenoid valve is in the state shown in FIG. Subsequently, when the driver steps on the brake pedal 12, the master cylinder 14 is first pushed in, and the communication between the master cylinder 14 and the reservoir tank 26 is shut off. Also, the right and left electromagnetic on-off valves 22FR and 22FL are closed, the simulator cut valve 23 is opened, and the communication of atmospheric brake oil from the master cylinder 14 to the right front wheel and left front wheel cylinders 20FR and 20FL is cut off. The Further, the pressure reducing valves 42RR and 42RL for the right rear wheel and the left rear wheel are closed, and the four pressure increasing valves 40FR, 40FL, 40RR and 40RL are opened. The opening degree of each solenoid valve is controlled based on the target wheel cylinder pressure of each wheel calculated through various calculations described later.

FIG. 2 is a functional block diagram relating to brake control by the brake control device.
In the ECU 200, first, the target deceleration calculation unit 61 uses the known method to reduce the target based on the pedal stroke ST acquired via the stroke sensor 46 and the master cylinder pressure PMC acquired via the master cylinder pressure sensor 48. A speed GxRef is calculated.

  Then, the target deceleration correction unit 62 corrects the target deceleration GxRef in consideration of a predetermined disturbance using a feedforward control law and a feedback control law. That is, the correction coefficient calculation unit 63 calculates the correction coefficient K using the actual deceleration (hereinafter referred to as “actual deceleration”) G input from the G sensor 70 that detects the current acceleration of the vehicle. The correction coefficient K is calculated by the product of a feedforward term KFF and a feedback term KFB, which will be described in detail later. Then, the target deceleration correction calculation unit 64 calculates the target deceleration GxRefh corrected by the correction coefficient K.

  Then, the target hydraulic pressure calculation unit 65 calculates the target wheel cylinder pressure Pref of the wheel cylinder 20 of each wheel based on the corrected target deceleration GxRefh by a known method.

  Then, the hydraulic pressure control unit 66 applies a control current I to the pressure increasing valve 40 and the pressure reducing valve 42 so that the target wheel cylinder pressure Pref is realized, and these are controlled to be opened and closed, and an appropriate braking force is applied to each wheel. Is done.

  In this embodiment, the target deceleration calculation unit 61 corresponds to the target deceleration calculation unit, the target deceleration correction unit 62 corresponds to the target deceleration correction unit, and the target hydraulic pressure calculation unit 65 corresponds to the target hydraulic pressure calculation unit. The hydraulic control unit 66 corresponds to the hydraulic control means.

  Next, the flow of the brake control process of the present embodiment will be specifically described. FIG. 3 is a flowchart showing the brake control process. This process is executed by the ECU 200 at predetermined intervals during braking.

  When the process is started, the ECU 200 first reads the pedal stroke ST detected by the stroke sensor 46 in the ECU 200 (S10), and then the master cylinder pressure detected by the master cylinder pressure sensor 48. PMC is read (S12). Note that a measurement value of either of the two master cylinder pressure sensors 48 may be used as the measurement value, or an average value of the two measurement values may be used. Further, a smooth signal may be obtained by appropriately filtering a signal indicating the master cylinder pressure.

  Subsequently, the target deceleration calculation unit 61 calculates a target deceleration GxRef based on the pedal stroke ST and the master cylinder pressure PMC (S14). In the ECU 200, the relationship between the pedal stroke ST, the master cylinder pressure PMC, and the target deceleration GxRef is previously mapped and stored.

  Subsequently, the target deceleration correction unit 62 appropriately corrects the target deceleration GxRef to obtain a corrected target deceleration GxRefh (S16). This correction process will be described in detail later.

  Subsequently, the target hydraulic pressure calculation unit 65 calculates the target wheel cylinder pressure Pref of each wheel based on the corrected target deceleration GxRefh (S18). In the ECU 200, the relationship between the target deceleration GxRefh and the target wheel cylinder pressure Pref is previously mapped and stored.

  Then, the hydraulic control unit 66 calculates a control current I corresponding to the target wheel cylinder pressure Pref, and energizes the pressure increasing valve 40 or the pressure reducing valve 42 to execute appropriate hydraulic control (S20). As a result, the pressure increasing valve 40 and the pressure reducing valve 42 are controlled to open and close, and an appropriate braking force is applied to each wheel.

  When it is confirmed that the ignition switch is turned off when the above processing is completed (YES in S22), the target deceleration correction unit 62 performs the current effect correction coefficient Hosei for the next processing. (Described later) is stored in a non-volatile memory such as a backup RAM (S24). Details of the effect correction coefficient storage processing will be described later.

  FIG. 4 is a flowchart showing the target deceleration correction process.

  In the target deceleration correction process of S16 described above, the correction coefficient calculation unit 63 calculates a correction coefficient K expressed by the following equation (1).

K = KFF · KFB (1)
KFF: Feedforward term KFB: Feedback term That is, the correction coefficient computing unit 63 first computes the feedforward term KFF (S30). This feed-forward term KFF is a correction term corresponding to long-term changes in braking effectiveness due to variations among individual brake pads of each wheel and changes over time. The specific calculation procedure will be described in detail later. Describe.

  Subsequently, the correction coefficient calculation unit 63 calculates a feedback term KFB represented by the following formula (2) by PID control (S32).

KFB = 1 + Pgain · err + Igain · ∫err + d / dt (Dgain · err) (2)
Pgain: Proportional gain
Igain: integral gain
Dgain: differential gain Note that err is the deviation between the target deceleration taking into account the feedforward term and the actual deceleration, and is calculated by the following equation (3).

err = GxRef · KFF-G (3)
G: Actual Deceleration Subsequently, the correction coefficient calculation unit 63 calculates the correction coefficient K from the above equation (1) using the feedforward term KFF and the feedback term KFB calculated in this way (S34).

  Then, the target deceleration correction calculation unit 64 calculates the corrected target deceleration GxRefh by the following equation (4) (S36).

GxRefh = K · GxRef (4)
FIG. 5 is a flowchart showing feedforward term calculation processing.

  The feed forward term KFF of S30 described above is expressed by the following equation (5).

KFF = KFFA / KFFB / KFFC (5)
KFFA: Effective measurement correction term
KFFB: Fade correction term
KFFC: Weight learning correction term Here, the effectiveness measurement correction term KFFA is slow in the long term for the effectiveness of the brake, such as the variation between individual brake pads and their peripheral components, etc., and the deterioration over time. It is a correction term for the disturbances that affect it. The fade correction term KFFB is a correction term for a disturbance that affects the effectiveness of the brake in a short time such as a fade phenomenon of the brake pad. Furthermore, the weight learning correction term KFFC is a correction term for disturbances that have a short-term influence on the effectiveness of the brake due to inertial force when the vehicle travels, such as changes in the number of passengers and load weight of the vehicle.

  In this feedforward term calculation process, the correction coefficient calculation unit 63 first calculates an effect measurement value ErrRatetemp representing the effect of the brake by the following equation (6) in order to calculate the effect measurement correction term KFFA (S40).

ErrRatetemp = (realGx−realGx0) / nominalGx−1 (6)
Here, realGx corresponds to the actual deceleration. Specifically, this is the actual deceleration G acquired from the G sensor 70 minus the deceleration of the deceleration factor unrelated to the brake hydraulic control. In the form, it is calculated by the following equation (7).

realGx = G sensor value-Other system deceleration-Slope correction deceleration (7)
That is, realGx is calculated by subtracting other system deceleration such as deceleration due to engine braking from the G sensor value, and further subtracting the deceleration corresponding to the slope correction. Here, the deceleration corresponding to the slope correction means a deceleration generated by an inertial force that acts when the vehicle is traveling on a slope. For this realGx, for example, an average value for a predetermined time (for example, 0.2 seconds) when the vehicle speed passes through a predetermined value (for example, 30 km / h) in the deceleration process of the vehicle can be used. When the vehicle is a hybrid vehicle or the like, the deceleration due to the regenerative brake can be included in the other system deceleration.

  The other system deceleration may be performed when the steering angle of a steering wheel (not shown) is greater than or equal to a predetermined value or when the parking brake (not shown) is turned on during acceleration such as when the accelerator pedal is depressed by the driver. When the angular change is large, an appropriate value may not be calculated. For this reason, in such a case, the measurement of the effective measurement value is not performed.

  Further, realGx0 is realGx immediately before the driver depresses the brake pedal. Here, in order to increase the accuracy of the effective measurement value ErrRatetemp, measurement is performed based on the change amount, not the actual measurement value of the G sensor. Further, nominal Gx is a deceleration calculated (predicted) from the wheel cylinder pressure detected by the WC pressure sensor 44, and a predetermined time (for example, 0) when the vehicle speed passes through a predetermined value (for example, 30 km / h). .2 seconds) average value can be used.

The effectiveness measurement value ErrRatetemp described above is measured when a predetermined measurement effective condition is satisfied in order to improve measurement accuracy. That is, 1) the effect measurement for a predetermined time (0.2 seconds in the above example) is possible when the vehicle speed is a predetermined value (30 km / h in the above example), and 2) the fluctuation of realGx during the effect measurement is predetermined. The measurement effective condition may be equal to or less than the value (eg, 3 m / s ² / sec or less). Condition 1) measures the amount of change based on a certain speed because the braking effect is highly dependent on the vehicle speed. Condition 2) is set to reduce the influence of transient characteristics. That is, even if the vehicle speed is a predetermined value, for example, when the amount of operation of the brake pedal by the driver is very fast, a large amount of error is included. Therefore, when the fluctuation of realGx is large as described above, the effect measurement is not performed or ignored.

  Subsequently, the correction coefficient calculation unit 63 performs the filtering process of the effect measurement value ErrRatetemp calculated as described above based on the preset upper limit value and lower limit value of the effect measurement value (S42). This excludes a portion that is not possible as an effective state of the brake. When the effective measurement value ErrRatetemp exceeds the upper limit value or falls below the lower limit value, the current measurement value is ignored. This is to ensure the stability of the control by ignoring the large deviation of the effective measurement value ErrRatetemp.

  Subsequently, the correction coefficient calculation unit 63 performs an effective measurement value averaging process to improve measurement accuracy, and calculates an average value ErrRate of ErrRatetemp (S44). For example, the average value of ErrRatetemp stored from the past is calculated.

  Then, the correction coefficient calculation unit 63 calculates the effectiveness correction coefficient Hosei represented by the following formula (8) (S46).

Hosei = 1 / (1 + ErrRate) (8)
This is a correction coefficient for supplementing the effectiveness of the brake by adding the reduced amount when the effectiveness of the brake is reduced from the originally planned value.

  Subsequently, the correction coefficient calculator 63 performs a filtering process based on the preset one-trip fluctuation amount limit value for the effectiveness correction coefficient Hosei calculated in this way (S48). Here, one trip means the time between the ignition switch on and off, but since the change in the braking effectiveness in one trip is limited to some extent, if the braking effectiveness is excessive, An abnormality is considered to have occurred. Therefore, the difference between the average value (described later) of the effectiveness correction coefficient Hosei memorized when the ignition switch was previously turned off and the effectiveness correction coefficient Hosei calculated this time, that is, the fluctuation amount of the effectiveness correction coefficient Hosei is the one-trip fluctuation. If it is larger than the amount limit value, the effect correction coefficient Hosei of this time is ignored.

  Further, the correction coefficient calculation unit 63 performs the filtering process of the effectiveness correction coefficient Hosei calculated as described above based on the preset upper limit value and lower limit value of the effectiveness correction coefficient (S50). That is, if the effectiveness correction coefficient Hosei exceeds the upper limit value or falls below the lower limit value, the current measurement value is ignored. This is to ensure the stability of control by ignoring the large deviation of the effectiveness correction coefficient Hosei.

  As described above, the effect correction coefficient Hosei calculated without being ignored is set as the effect measurement correction term KFFA (S52).

  Subsequently, the correction coefficient calculation unit 63 calculates a fade correction term KFFB that eliminates the disturbance due to the fade phenomenon of the brake pads (S54). By incorporating the fade correction term KFFB, for example, when the temperature of the brake pad rises in the short term and the effectiveness of the brake decreases, the effectiveness of the brake can be compensated.

  That is, the correction coefficient calculation unit 63 estimates the temperature of the brake pad based on the detection value of a temperature sensor (not shown) installed around the brake pad. In the ECU 200, the relationship between the estimated temperature of the brake pad and the fade correction term KFFB is previously mapped and stored. The correction coefficient calculation unit 63 refers to the map based on the estimated temperature of the brake pad and sets a fade correction term KFFB.

  Subsequently, the correction coefficient calculation unit 63 calculates a weight learning correction term KFFC that eliminates disturbances caused by inertial force due to vehicle weight, such as changes in the number of passengers and load weight of the vehicle (S56). By incorporating the weight learning correction term KFFC, for example, when the braking effectiveness decreases during traveling on a downhill of the vehicle, the braking effectiveness can be compensated.

  That is, the correction coefficient calculation unit 63 estimates the current vehicle weight from the acceleration torque of the vehicle acquired from a torque sensor (not shown), the acceleration acquired from the G sensor 70, the vehicle height acquired from a vehicle height sensor (not shown), and the like. In the ECU 200, the relationship between the acceleration torque and acceleration and the vehicle weight, and the relationship between the vehicle weight and the weight learning correction term KFFC are previously mapped and stored. The correction coefficient calculation unit 63 obtains the vehicle weight with reference to the former map, and further sets the weight learning correction term KFFC with reference to the latter map.

  Then, the correction coefficient calculation unit 63 calculates the feedforward term KFF by substituting the effectiveness measurement correction term KFFA, the fade correction term KFFB, and the weight learning correction term KFFC set as described above into the above equation (5). (S58).

  FIG. 6 is a flowchart showing the correction coefficient storage process. This process is the process of S24 described above, and stores the “previous effect correction coefficient Hosei” used for the next process of S48. When it is confirmed that the ignition switch is turned off, the correction coefficient calculation unit 63 stores the current effect correction coefficient Hosei in a non-volatile memory such as a backup RAM for the convenience of the next processing.

  That is, the correction coefficient calculation unit 63 first estimates the temperature of the brake pad based on detection values of a temperature sensor or the like (not shown) installed in the vicinity thereof (S70).

  Then, the correction coefficient calculation unit 63 determines whether or not the temperature of the brake pad is equal to or lower than a preset storage permission temperature (S72). At this time, if the temperature of the brake pad exceeds the storage permission temperature (NO in S72), it is considered that the brake pad or its peripheral parts have the above, so the correction coefficient storage processing is interrupted. . That is, the last effective correction coefficient Hosei stored at present is used for the next processing of S48.

  On the other hand, if the temperature of the brake pad is equal to or lower than the storage permission temperature (YES in S72), the correction coefficient calculation unit 63 subsequently executes the effect correction coefficient storage value averaging process (S74). Here, in order to prevent the effect of the effectiveness correction coefficient Hosei calculated by the current arithmetic processing from becoming excessive, the annealing process is performed between the effect correction coefficient Hosei stored last time based on the following equation (9).

Hosei memory value = β · Hosei + (1-β) · Hosei previous memory value (9)
Hosei memory value: Effective correction coefficient memorized this time Hosei
Hosei last remembered value: remembered when the last ignition switch was turned off
Effectiveness correction coefficient Hosei
Here, β is a weighting coefficient, and can be set to 0.2 (that is, 20%), for example.

  The correction coefficient calculation unit 63 stores the Hosei storage value calculated by the above equation (9) in a non-volatile memory such as a backup RAM (S76).

  As described above, in the present embodiment, the target deceleration GxRef calculated based on the pedal stroke ST and the master cylinder pressure PMC is corrected based on the feedforward control law and the feedback control law. A target wheel cylinder pressure Pref that realizes GxRefh is calculated. In particular, since the effectiveness measurement correction term KFFA learned based on the state of the brake pad or the like is used for the feedforward term KFF, the influence of deterioration of the components of the brake device in the feedback control is reduced. Therefore, hunting of the control amount can be prevented or suppressed. Further, since the feedback control is continuously executed while taking the feed forward term KFF into consideration, the control followability can be maintained. As a result, the stability of the control system can be maintained and a good brake feeling can be maintained even when there is a disturbance due to aging deterioration of the components of the brake device.

  In addition, when the braking effectiveness of any of the wheels decreases due to, for example, deterioration of any of the components of the brake device, a correction term that compensates for the reduction in the effectiveness is calculated as a feedforward term KFF. To be learned. For this reason, the control amount is raised beforehand by the feedforward term KFF, and the feedback gain of the feedback control can be relatively reduced to maintain its stability.

[Second Embodiment]
In the first embodiment described above, long-term learning is required to some extent in order to make the particularly effective measurement correction term KFFA of the feedforward term KFF function effectively. On the other hand, there is no deterioration over time even if there are variations among individual brake pads etc. at the time of vehicle shipment from the factory. For this reason, at the beginning of shipment from the factory, there is a case where the followability to the control of the feedforward term KFF learned based on the effective state of the brake cannot be obtained. Therefore, in the present embodiment, the feedback gain is switched at a timing at which the effective measurement correction term KFFA can be effectively functioned. The present embodiment is substantially the same as the first embodiment except that the contents of the target deceleration correction process shown in FIG. 4 are slightly different, and therefore the same configuration and method will be described. Will be omitted.

  FIG. 7 is an explanatory diagram illustrating a control map used in the target deceleration correction process according to the second embodiment.

  That is, in the present embodiment, the control map shown in FIG. 7 is set in order to determine whether or not the effectiveness measurement correction term KFFA can function effectively. In the figure, the horizontal axis represents the KFFA reliability index, and the vertical axis represents the proportional gain Pgain of the feedback term KFB. Here, the KFFA reliability index is an index for determining the effectiveness of the effectiveness measurement correction term KFFA, and it can be determined that the feedforward control considering the effectiveness measurement correction term KFFA functions more effectively as the value increases. The KFFA reliability index is expressed by the following equation (10) and is sequentially stored in a backup RAM or the like.

KFFA reliability index = number of learning points + number of learning memory trips (10)
Here, “the number of learning points” is incremented by 1 each time the effective measurement value ErrRatetemp is satisfied in the process described in the first embodiment. Further, “the number of learning storage trips” is incremented by 1 each time the effective correction coefficient Hosei is learned and stored when the ignition switch is turned off in S62 described later.

  As shown in the figure, when the vehicle is shipped from the factory, the feedback gain (in this case, the proportional gain Pgain) is increased to ensure control followability, and when the KFFA reliability index exceeds a predetermined value (for example, 30), The feedback gain is lowered to a predetermined fixed value. An appropriate value of the proportional gain Pgain is obtained in advance through experiments or the like.

  FIG. 8 is a flowchart showing the target deceleration correction process.

  In the target deceleration correction process of the present embodiment, the correction coefficient calculation unit 63 first calculates the feedforward term KFF as in the first embodiment (S30).

  Subsequently, the correction coefficient calculation unit 63 reads the KFFA reliability index (S60), sets the proportional gain Pgain with reference to the control map (S62), and calculates the feedback term KFB based on the proportional gain Pgain. (S32).

  Then, the correction coefficient calculation unit 63 calculates the correction coefficient K from the above equation (1) using the feedforward term KFF and the feedback term KFB calculated in this way (S34), and according to the above equation (4). The corrected target deceleration GxRefh is calculated (S36).

  As described above, in this embodiment, the feedback gain is increased by a predetermined amount until learning based on the braking effectiveness state is normally performed a predetermined number of times, and when the learning exceeds the predetermined number of times, the feedback gain is increased. I tried to lower it. As a result, the feedforward term KFF can function effectively.

  The present invention is not limited to the above-described embodiments, and various modifications such as design changes can be added to the embodiments based on the knowledge of those skilled in the art. Embodiments to which is added can also be included in the scope of the present invention.

  For example, in each of the above embodiments, the target deceleration GxRef is calculated based on the pedal stroke ST and the master cylinder pressure PMC that change according to the depression amount of the brake pedal 12, and the feedforward control law and the target deceleration GxRef are calculated. Although an example in which correction based on the feedback control law is performed has been shown, the target deceleration GxRef may be calculated based on one of the pedal stroke ST and the master cylinder pressure PMC. Alternatively, a pressure sensor that detects a stroke simulator pressure applied to the stroke simulator 24 may be provided, and the target deceleration GxRef may be calculated based on the stroke simulator pressure.

1 is a system diagram showing a brake control device according to a first embodiment of the present invention. It is a functional block diagram concerning brake control by a brake control device. It is a flowchart showing a brake control process. It is a flowchart showing a target deceleration correction process. It is a flowchart showing a feedforward term calculation process. It is a flowchart showing a correction coefficient storage process. It is explanatory drawing showing the control map used for the target deceleration correction process which concerns on 2nd Embodiment. It is a flowchart showing a target deceleration correction process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Brake control apparatus, 12 Brake pedal, 14 Master cylinder, 20 Wheel cylinder, 24 Stroke simulator, 40 Pressure increase valve, 42 Pressure reduction valve, 44 WC pressure sensor, 48 Master cylinder pressure sensor, 61 Target deceleration calculation part, 62 Target reduction Speed correction unit, 63 correction coefficient calculation unit, 64 target deceleration correction calculation unit, 65 target hydraulic pressure calculation unit, 66 hydraulic control unit, 70 G sensor, 200 ECU

Claims (5)

In the brake control device that controls the braking force applied to the wheel by controlling the hydraulic pressure supplied to the wheel cylinder,
Target deceleration calculation means for calculating the target deceleration of the vehicle based on the operation of the brake operation member;
Feedback based on the deviation between the target deceleration calculated by the target deceleration calculation means based on at least the effective state of the brake and the target deceleration and the actual deceleration considering the feedforward term Target deceleration correction means for correcting using both of the term and
Target hydraulic pressure calculation means for calculating a target hydraulic pressure to be supplied to the wheel cylinder based on the target deceleration corrected by the target deceleration correction means;
Hydraulic control means for controlling the hydraulic pressure supplied to the wheel cylinder based on the target hydraulic pressure calculated by the target hydraulic pressure calculating means;
A brake control device comprising:   The target deceleration correction means calculates, as the feedforward term, a correction term that compensates for a change in the effectiveness of the brake based on a state of a component of a brake device including at least a brake pad. The brake control device described.   The target deceleration correction means sequentially updates, as the feedforward term, a correction term learned based on a state of the component and a correction term based on the weight of the vehicle. The brake control device described.   The target deceleration correction means updates the feed-forward term by performing learning based on the effectiveness state of the brake only when a preset measurement effective condition is satisfied. Brake control device.   The target deceleration correction means increases the feedback gain by a predetermined amount until learning based on the braking effectiveness state is performed a predetermined number of times, and decreases the feedback gain when the learning exceeds the predetermined number of times. The brake control device according to claim 4.

Images