METHOD FOR IMPROVING THE CONTROL BEHAVIOR OF AN ELECTRONIC MOTOR VEHICLE BRAKING SYSTEM

Abstract

The invention relates to a method for improving the control behavior of an electronic motor vehicle braking system which comprises at least a slip control function. Wheel dynamic information which is evaluated as a criterion for initiating a control intervention is used individually for each wheel and is compared with control thresholds for a pressure reduction phase, a pressure maintenance phase, and a pressure buildup phase for generating corresponding braking torques by means of a vehicle braking system. According to the invention, the expected acceleration change of a vehicle wheel is calculated from a pressure change at said wheel, said pressure change being caused by a control intervention; the actual acceleration change at the vehicle wheel, said acceleration change being caused by the pressure change, is determined from measured wheel speeds as wheel dynamic information; and the control behavior of the slip control is adapted when the actual acceleration change deviates from the expected acceleration change by a defined degree such that the deviation is minimized.

Description

The invention relates to a method for improving the control behavior of an electronic motor vehicle braking system according to the preamble of patent claim 1.

Systems for preventing the vehicle wheels from locking (anti-lock braking systems, ABS), for controlling traction slip of the driven wheels (traction-slip control system, ASR) are known and comprise slip control.

To prevent the wheels from locking when braking, the anti-lock braking system analyzes the turning performance of the individual wheels and determines a reference vehicle speed therefrom. In addition, the wheel slip is established and analyzed individually for each wheel. Threshold values of the wheel slip are generally predetermined for controlling the wheel slip, which define the use of the control.

In known methods for the braking control of vehicles, an attempt is made to prevent overbraking—that is to say an excessive slip between the wheels of the vehicle and the highway surface. The slip optimum is approx. 7 to 15% slip depending on the tire type. “Slip onset” is understood to mean exceeding this optimum, wherein the tire also loses its cornering stability characteristics.

A corresponding controller should therefore be able to achieve the set control target long-term. If a control deviation still persists after a certain time, the controller recontrols, for example by means of further pressure-decrease phases or pressure-increase phases, in order subsequently to reduce the control deviation. The temporal behavior is determined by the parameterization of the controller.

In the case of a well applied controller, large control deviations normally only occur if something “unexpected”, that is to say a fault, occurs and the motor vehicle braking system does not behave as expected owing to this fault.

It is the object of the invention to create an above-mentioned method, using which an imminent large control deviation is detected before the occurrence thereof, so that the occurrence of such a control deviation is prevented or at least the duration of the occurrence of such a control deviation is shortened by means of a timely control intervention.

This object is achieved by means of a method with the features of patent claim 1.

A method of this type for improving the control behavior of an electronic motor vehicle braking system, which comprises at least the function of a slip control, in which analyzed wheel dynamics information is used for each individual wheel as a criterion for initiating a control intervention and is compared with control thresholds for a pressure-decrease phase, a pressure-maintenance phase and a pressure-increase phase for generating corresponding braking moments by means of a vehicle braking system, is characterized according to the invention in that

• • the expected acceleration change at a vehicle wheel is calculated from a pressure change at this wheel effected by means of a control intervention,
  • the actual acceleration change effected by the pressure change at the vehicle wheel is determined from measured wheel speeds as wheel dynamics information, and
  • in the case of a deviation of the actual acceleration change from the expected acceleration change by a defined amount, the control behavior of the slip control is adjusted in such a manner that the deviation becomes minimal.

Using this method according to the invention, the quality of the slip control can be improved substantially, i.e. the depth of slip onsets is reduced, in particular the duration of the presence of a large control deviation is shortened substantially.

The relevant variables in this method according to the invention are on the one hand the pressure change effected by a control intervention, which leads to a corresponding change of the braking moment, and on the other hand the change in the wheel acceleration effected by this pressure change or by this braking-moment change, wherein the change of the wheel acceleration, that is to say—viewed mathematically—the derivative of the wheel acceleration or the second derivative of the speed is generally termed jerk and constitutes a differential variable which is calculated in the control algorithm from the difference of two adjoining acceleration values.

Starting from the pressure change effected by a control intervention, the acceleration change initiated thereby, that is to say the expected acceleration change, is calculated, a braking pressure model being used for this. Using this braking pressure model, first the braking pressure acting on the vehicle wheel is calculated and the expected acceleration change is calculated therefrom, further parameters, such as the highway coefficient of friction, coefficient of friction of the lining of the brake disk, etc. also being used for the latter. This expected acceleration change is then compared with the actually beginning acceleration change determined from the wheel speed as the second derivative thereof.

According to an advantageous embodiment of the invention, the defined amount is given by a tolerance band defined around the value of the expected acceleration change. If the value of the actual acceleration change lies outside this tolerance band, the deviation is too large, so that a control intervention becomes necessary.

An alternative embodiment of the invention provides that the defined amount is given by a value range, the values of which are compared with the quotient of the actual acceleration change and the expected acceleration change.

According to a further preferred development of the invention, the control behavior is adjusted by means of a change of the controller amplification factors and/or by means of a change of the controller frequency and/or by means of a change in the pulse height relating to the pressure request.

As the signals for calculating the expected acceleration change and the signals for determining the actual acceleration change must physically relate to the same point in time, it is provided according to a development that the pressure change signals used for determining the expected acceleration change and the wheel speed signals used for determining the actual acceleration change are temporally synchronized to one another.

The method according to the invention is explained in more detail in the following with reference to the attached FIG. 1. This FIG. 1 shows the curves occurring at a vehicle wheel for the speed v, the braking pressure p and the expected and actual acceleration change, which represents a jerk value j as a differential variable.

The following is based on an ABS control of a motor vehicle braking system using an ABS controller, using which the braking behavior of the vehicle wheels of a vehicle is controlled and an excessive slip is prevented in the process.

FIG. 1 shows a section from such an ABS control of a vehicle wheel, which is in an unstable phase, that is to say in a braking phase with slip onset.

Thus, at time t₁, a pressure-decrease phase is introduced by the ABS controller, with the consequence that the braking pressure p and thus also the wheel speed v decreases in accordance with the representation according to FIG. 1.

On the basis of the pressure change at the vehicle wheel, an expected acceleration change is calculated as jerk value j_exp from the gradient of the pressure p by means of a braking pressure model. As the pressure p for t<t₁ and t>t₃ is constant, the value zero results for the expected acceleration change j_exp. With the start of the pressure decrease, that is to say shortly after the time t₁, this calculated signal j_exp changes to a positive value, which remains constant due to the constant gradient of the pressure until the end of the pressure decrease at time t₃, in order to subsequently drop again to zero.

The actual acceleration change is determined as jerk value j_act and calculated from the wheel speed v as 2nd derivative. As long as the speed curve v is curved to the right, the actual jerk j_act shows negative values and becomes positive if the speed curve v transitions to a left curvature (cf. FIG. 1).

During the entire ABS control, the actual jerk value fact is compared with the expected jerk value j_exp.

Generally, these two variables j_exp and j_act only differ slightly within a predetermined amount, i.e. the ABS control behaves as expected.

However, if the situation illustrated in FIG. 1 by way of example, that the difference between these two values j_exp and j_act exceeds a defined amount at time t₂, occurs, this means that the acceleration change j_act that is set is substantially smaller than the expected acceleration change j_exp.

The defined amount is given by a tolerance band defined around the value of the expected acceleration change j_exp. A fault is assumed if the actual acceleration change j_act lies outside this tolerance band. Therefore, a large control deviation is imminent, which may lead to a slip onset which will last for a long time and can only be corrected over a plurality of control loops.

The cause of such a fault, that is to say the deviation of the actual acceleration change j_act from the expected acceleration change j_exp beyond the defined amount may for example lie in the wheel load decreasing, a decreasing highway coefficient of friction being present (due to a change in the highway surface) and/or the tire coefficient of friction decreasing because of the slip. The cause may also lie on the actuator side of the vehicle braking system, for example a smaller coefficient of friction of the lining may be present than was assumed during the calculation of the expected jerk value j_exp, as a result of which the braking-moment decrease is actually smaller than expected. The defined amount for detecting a fault may also be specified by a value range, the values of which are compared with the quotient of the actual acceleration change j_act and the expected acceleration change j_exp. If this quotient lies outside this value range, a fault is assumed.

If a deviation of the actual acceleration change j_act from the expected acceleration change j_exp by the defined amount detected in the above-mentioned manner is then present, the ABS controller intervenes and carries out a correction at time t₂, in that for example by means of a parameter switch, the pressure request is corrected such that the pressure decrease is continued in a timely manner and consequently at a time t₃, the actual jerk value j_act follows the expected jerk value j_exp within the defined amount, as is illustrated in FIG. 1. From this time t₃, the deviation between these two values j_exp and j_act is only slight and therefore no longer lies outside the defined amount.

Also illustrated in FIG. 1 is the situation where an early extension of the pressure-decrease phase is not carried out in spite of the deviation of the two variables j_exp and j_act lying outside the defined amount. The pressure decrease would then be ended at time t₂ (cf. FIG. 1, braking pressure curve p, illustrated dashed), with the consequence that the slip also increases further, that is to say the slip onset is extended (cf. FIG. 1, speed curve v, illustrated dashed). The ABS control would only react to this slip onset with a pressure decrease at a later point in time, that is to say after a decrease pause, with the consequence that the speed v would fall further after the time t₂.

However, using the method according to the invention, the pressure decrease is continued or even amplified because of the deviation between the expected jerk value j_exp and the actual jerk value j_act lying outside the defined amount.

In the case of a deviation of the two jerk values j_exp and j_act lying outside the defined amount, the control behavior of the ABS controller can also consist in changing the controller amplification factors, particularly the I part (also the other parts), reducing the pause time of the control, that is to say increasing the control frequency or changing the pulse height.

When determining or calculating the variables j_exp and j_act it is to be ensured that the associated signals with regards to the pressure change and the acceleration change are synchronized to one another, and that they do not physically relate to the same point in time, even if they are calculated simultaneously in the ABS controller.

As the mathematical derivation in the ABS controller is for the most part determined computationally by means of a differentiation between the respective signal variable at the current point in time and the signal variable at a preceding point in time, there is a time delay here which adds up over multiple derivations. The actually “oldest” signal is therefore relevant during the synchronization and older values of the physically newer signals must be made available by means of intermediate storage. If one of the signals used is filtered, the other signals used are correspondingly also to be filtered in the same way.

The above-described exemplary embodiment relates to the case of “pressure decrease”, in order to allow re-acceleration. Naturally, it is also valid for the case of “pressure increase”, in order to limit the re-acceleration.

The illustrated method is suitable not only for ABS control, if the vehicle wheel is in the unstable region of the p-slip curve, but also generally for a slip control in the unstable region of a vehicle wheel.

Claims

1. A method for improving the control behavior of an electronic motor vehicle braking system, comprising at least the function of a slip control, in which analyzed wheel dynamics information is used for each individual wheel as a criterion for initiating a control intervention and is compared with control thresholds for a pressure-decrease phase, a pressure-maintenance phase and a pressure-increase phase for generating corresponding braking moments by means of a vehicle braking system, characterized in that the expected acceleration change (jexp) at a vehicle wheel is calculated from a pressure change at this wheel effected by means of a control intervention,the actual acceleration change (jact) effected by the pressure change at the vehicle wheel is determined from measured wheel speeds as wheel dynamics information, andin the case of a deviation of the actual acceleration change (jact) from the expected acceleration change (jexp) by a defined amount, the control behavior of the slip control is adjusted in such a manner that the deviation becomes minimal.

2. The method as claimed in claim 1, characterized in that the defined amount is given by a tolerance band defined around the value of the expected acceleration change (jexp)

3. The method as claimed in claim 1, characterized in that the defined amount is given by a value range, the values of which are compared with the quotient of the actual acceleration change (jact) and the expected acceleration change (jexp).

4. The method as claimed in one of the preceding claims, characterized in that the control behavior is adjusted by means of a change of the controller amplification factors and/or by means of a change of the controller frequency and/or by means of a change in the pulse height relating to the pressure request.

5. The method as claimed in one of the preceding claims, characterized in that the pressure change signals used for determining the expected acceleration change (jexp) and the wheel speed signals used for determining the actual acceleration change (jact) are temporally synchronized to one another.