DE19638306A1 - Vehicle braking system with slip regulation - Google Patents

Abstract

The braking system allows a required braking slip to be provided at a given vehicle wheel, with detection of the wheel rotation rates (Nvr,Nvl,Nhr,Nhl), for calculation of the vehicle longitudinal velocity (Vref) and the momentary wheel slip. The required braking effect is calculated from the longitudinal velocity and the required slip and compared with the actual braking effect for regulation of the wheel braking.

Description

State of the art

The invention relates to a braking system with the features of Claim 1.

Previous slip regulators, such as those from the EP, B1.0 365 604 (corresponds to US 5,136,509) are known, use a proportional-differential control law to get one Set the desired target speed of a wheel. Of the The setpoint is derived from a reference variable that the Longitudinal vehicle speed corresponds, and one predetermined target brake slip of the wheel is formed. The Setting the target speed or The target slip occurs by adjusting Valves that lower, maintain or increase brake pressure can. The setting of the target speed or the target slip is in general associated with a considerable amount of space.

Furthermore, for example, from

• - Publication 1: Vadim I. Utkin: Sliding Modes in Control and Optimization, Springer Verlang, Berlin 1992 and  
• - Publication 2: Jean-Jaques E. Slotine and Weipeng Li: Applied Nonlinear Control, Prentice Hall International, England Woods Cliffs, NJ, 1991 structure-variable controller known.

The object of the present invention is a desired target brake slip with as little as possible Adjust the effort.

This object is achieved by the features of claim 1 solved.

Advantages of the invention

As mentioned, the invention relates to a braking system for Setting a predeterminable brake slip at least a wheel having braking means in a motor vehicle. Here, speed values are recorded that reflect the movements of the Vehicle wheels, preferably the wheel speeds, represent. Furthermore, a Longitudinal vehicle speed representative reference quantity determined depending on the recorded speed values. Dependent from the recorded speed values and the determined The reference value will then be the current slip at the Wheel representing slip size determined. Furthermore a target value for a braking effect is dependent on the determined reference size and the to be set Target slip determined. The setting of the predeterminable Brake slip then occurs by adjusting the Braking effect on the determined target value. This setting is preferably done by hiring or change a corresponding brake pressure.

The braking system according to the invention can be used with little Adjustment effort an adjustment of the desired brake slip  can be achieved. This results in a very calm Pressure course during the slip control and a large one Robustness against disturbances of the input variables.

In an advantageous embodiment of the invention provided that the instantaneous braking effect on the wheel representative actual size is determined. For setting the braking effect on the specific setpoint is then the certain actual size compared with the certain target value. In this embodiment, it can be provided that a Friction coefficient representing the friction value of the vehicle wheel is determined depending on the determined reference variable. The actual size then depends on the determined Friction size and the change over time of the determined Slip size determined.

In the case of the last-mentioned configuration, provision can be made be that to adjust the braking effect on the certain setpoint, the braking effect is changed, whereby such a change will only be made if the certain actual size of the certain value in predeterminable Way deviates. From the hysteresis of the actuators and from Interference with the input signals can result in a constant and switching the controller (chattering) surrender. This embodiment of the invention allows this Chattering effect can be at least reduced, with which the great robustness against disturbances of the input variables is achieved.

In another embodiment of the invention, that at least depending on the particular slip size and whose change over time is a switching value. For Setting the braking effect to the specific setpoint the braking effect is then dependent on the formed Switching value.  

Provision can also be made for the latter configuration be that to adjust the braking effect on the certain setpoint, the braking effect is changed, whereby a change is only made if the educated Switching value from a specifiable setpoint switching value in specifiable way deviates. This allows this Design of the chattering effect mentioned at least can be reduced.

The formation of the switching value can depend on one Comparison of the determined slip size with the predeterminable Brake slip happen.

Furthermore, it can be provided that the coefficient of friction of the Vehicle wheel representative friction size depending on the determined reference size is determined and the formation of the Switching value happens depending on the determined coefficient of friction.

For the mentioned formation of the coefficient of friction of the vehicle wheel Friction quantity representing the vehicle deceleration representative delay value depending on the determined reference size can be determined.

Further advantageous configurations are the See subclaims.

drawing

The Fig. 1 with the parts a and b is a block diagram of the invention. Fig. 2 shows with the parts a, b and c, the function or operation of a variable structure controller, while FIG. 3 shows an equivalent mechanical model of a single wheel for Derivation of the switching line equation disclosed. FIG. 4 shows the principle of a brake slip controller and FIG. 5 shows the division of the controller areas in comparison to the μ slip curve.

Embodiment

A structurally variable controller divides the entrance space a switching function s (x) in different areas and used in the individual areas depending on the sign of s (x) different area rules. the equation s (x) = 0 defines a button in the entrance room. The switching function is selected so that once The system reaches the button exactly on this Button the specified target value (target slip) reached.

The advantage of using a structurally variable controller on the brake slip control is that the choice of Switching line is done in such a way that no further effort is required. This results in a very quiet pressure curve during the slip control and great resilience to interference from the Input variables.

To calculate this "optimal" switching function, the Road surface friction used. Because this value is not direct can be measured, it is here from internal Brake slip controller sizes estimated.

Before the invention is based on an exemplary embodiment a brake slip controller is described in detail in the following, briefly on the principle known per se of a structurally variable controller.  

Principle of a structure-variable controller

As already mentioned, this makes a structurally variable controller characterized that he entered the entrance room with a Button s (x) = 0 divided into two areas. Depending on The sign of the switching function s (x) becomes another Area regulation law used.

Convention: The sign of s (x) is chosen such that u⁺ <u⁻. With a suitable choice of the area rule laws u⁺ and u⁻, the system remains on the button once the button has been reached, since the system trajectories on both sides of the button return to the button. This can be seen in FIG. 2a, which goes back to page 4 of the publication 1 mentioned at the beginning.

FIG. 2a shows the basic behavior of a structure-variable controller which uses different control laws u⁺ and u⁻ depending on the sign of a switching function s (x).

The manipulated variable u is not defined on the s (x) = 0 button. In practice, the controller's finite sampling time, the hysteresis of the actuators and interference in the input signals result in the controller constantly toggling back and forth ("chatter" , English chattering). Switching between u⁺ and u⁻ can mean that the system can no longer leave the switching line s (x) = 0. This situation is shown in FIG. 2b, which goes back to page 283 of publication 2 mentioned at the beginning. In the event of small deviations from the switching line, the controller guides the system back to the switching line.

The manipulated variable that the system would exactly control on the button is called the equivalent control u _eq . It results from the mean manipulated variable of the controls u⁺ and u⁻. This is shown in FIG. 2c, which goes back to page 284 of the publication 2 mentioned at the beginning and represents the manipulated variable which the system guides exactly along the switching line.

Application on brake slip control

A brake slip controller calculates a brake pressure difference Δp _R for each wheel from the wheel speeds, which in turn is converted via a correction element into a control time of the intake / exhaust valves for the brake pressure (see EP, B1,0 365 604 mentioned at the outset).

The switching function s (x) is now selected so that in the Movement along the button is not triggered got to:

Δp _{R, eq} = 0 (2)

FIG. 5 shows the known profile of the μ (λ) slip curve (profile of the braking effect u or the brake pressure depending on the brake slip λ). The position of the switching function in the stable region of the µ (λ) curve (up to the maximum of the µ (λ) slip curve) is determined by the "optimal" brake pressure u _s in the steady state of braking. With constant brake slip λ _s , there is a stationary brake pressure p _s .

The “optimal” switching line of the proposed regulator thus runs in the stable region of the μ (λ) curve shown in FIG. 5 on the line of constant “optimal” pressure u _s .

As will be explained in detail below, this "optimal" shift line is converted into the input variables slip λ and slip change from the mechanical replacement model of a single wheel shown in FIG. 3.

In the unstable area of the μ (λ) curve shown in FIG. 5, a proportional-differential control law is used, as in some cases as before (eg EP, B1.0 365 604), but in some cases the brake pressure is kept constant to return to the stable range at a higher pressure level. Fig. 5 shows the individual controller areas compared to the µ (λ) curve. In the stable region of the µ (λ) curve, the system thus moves to the stationary braking state on the switching line of the optimal braking pressure u _s . The chattering effect is avoided by an interpolation area (dark gray in Fig. 5) around the "optimal" brake pressure u _s .

Derivation of the switching equation

The equations of the shift line in the entrance space of the brake slip controller result from the modeling of a single wheel. The forces and moments that act on a single wheel are shown in FIG. 3. The brake pressure p _R generates a braking torque M _B = k _{M *} p _R on the wheel, with k _M as the pressure amplification factor. The normal force F _z on the wheel is constant when considering a single wheel and corresponds to the weight of the proportional vehicle mass m: F _z = m _* g. The frictional force results from the normal force, multiplied by the coefficient of friction µ. The coefficient of friction is strongly dependent on the brake slip

where V _{Fzg denotes} the longitudinal vehicle speed and Nrad the wheel rotation speed (wheel speed) of the wheel under consideration. The following also applies:

F _R = µ (λ) _* F _z

The differential equations for the vehicle speeds V _Fzg and the wheel _speed N _rad follow from the momentum and spin conservation laws of mechanics. The differential equation for the brake slip λ can be derived from this
Equation 3:

with the abbreviations

are. G denotes the acceleration due to gravity and u the current one Brake pressure.

If the target slip λ _{s is} set during braking, the brake pressure is constant p _{R, s} . The value of p _{R, s} or u _{R, s} results from the condition that the slip change becomes zero with constant slip λ _s :
Equation 4

If this "optimal", stationary brake pressure u _{R, s is} reached, no further pressure change is necessary. The "optimal" switching line of the structure-variable control approach in the stable region of the µ (λ) curve results from the differential equation of the slip:
Equation 5:

 Substituting equation 3c into equation 5, so you can see that the value

represents the deviation of the instantaneous actual brake pressure u from the target value u _s .

In the unstable area, the switching line according to FIG. 5 is selected. However, other functions can also be selected. A movement on the switching line only exists in the stable area, therefore the selection of the switching line in the unstable area is of little importance.

The switching line in the stable area has the following Characteristics:

• - There is a movement on the switching line to the rest position with target slip λ _s
• - For the equivalent control, the following applies: Δp _{R, eq} = 0.

The vehicle _speed V _Fzg and the current coefficient of friction µ (λ) are included in the calculation. Both quantities are not available as measured quantities and must therefore be estimated. The following:

Estimation of vehicle speed

Since all four wheels of a vehicle assume more or less high slip values during braking, the longitudinal vehicle _speed V _{Fzg is} not known during the control. However, the brake slip controller determines an estimated value V _ref , which is extrapolated with an estimated vehicle deceleration b _x . Both are updated at regular intervals (see EP, B1.0 365 604). The control law of the brake slip controller must use these estimates in order to calculate the relative slip values from the measurements of the wheel speeds N _rad .

Estimation of the coefficient of friction

During braking, the brake slip controller sets the specified target slip λ _s on each wheel. As is stated in the aforementioned EP, B1.0 365 604, this slip control is switched off from time to time and the wheel is braked in a targeted manner so that the reference variables V _ref and b _x can be determined anew. FIG. 4 illustrates this situation.

As can be seen from FIG. 4, the relative target slip _{s is} set from a given reference slip λ _Ref . The combination of λ _Ref and _s gives the absolute target slip λ _s . If the controller is switched off and the wheel is accelerated by the amount Δ, the new values of V _ref and b _{x are} determined again in this state. However, this means that the key points of this movement are known and can be converted into µ values. The estimated vehicle deceleration b _x indicates the µ value when the controller is switched off:

µ (λ _Ref ) = -b _x / g (7)

If you add the wheel acceleration during the transition from the slip control to the switched-off controller to this value, you get the µ value with absolute target slip λ _s . Since the µ (λ) curve is usually curved, there is no linear interpolation between these corner points, but a 3rd degree polynomial µ * is calculated, which is represented by the points µ * (0) = 0, µ * ( _see ) = µ (λ _s ) -µ (λ _ref ) and is determined by a horizontal tangent at the maximum at _s . In the unstable range, a decrease in the µ (λ) curve to a blocking coefficient of friction µ _Block = c _* with c = 0.8 is assumed. An alternative approach to approximation uses a simple sine curve in the stable region and keeps µ * constant in the unstable region.

in which

is. The value Δ corresponds to Predefinable parameter of the target re-acceleration of the Wheel when the brake pressure is reduced by Δp (variable).

The coefficient of friction thus results from the addition of the coefficient of friction at λ _Ref with the curved interpolated curve µ *

Regulatory law

The rule law is therefore:

In a narrow environment around the switching line in a stable area, interpolation is carried out between assembly and disassembly. In the switching function equations, the estimates V _ref and () are used instead of the unknown vehicle speed and friction values. The PD control law (Δp = f _{P, D} ) in the unstable range corresponds to the calculation described in EP, B1.0 365 604.

Specific embodiment

A concrete exemplary embodiment is to be illustrated below with reference to FIGS. 1a and 1b.

In FIG. 1a, the blocks 101 mit wheel speed sensors are used for this purpose, the index i representing the front (i = v) or rear (i = h) axis and the index j representing the right (j = r) or designated left (j = l) vehicle side.

These speed _{signals are} fed to block 102 , which estimates the longitudinal vehicle speed V _Fzg as V _{ref in a} manner known _{per se} (see above-mentioned heading "Estimating the Vehicle _Speed "). In addition, the vehicle _acceleration b _{x is} estimated in block 102 from the estimated vehicle _speed V _ref .

In block 103 , the slip values λÿ are formed from the detected speed values Nÿ according to equation 2a, the reference speed V _ref formed in block 102 being used as the vehicle _speed V _Fzg .

In the following, only the further processing of the signals for the front right wheel will be shown with reference to FIG. 1. The brake slip of the other wheels is carried out by further processing of the same type based on the slip values formed in block 103 .

The longitudinal acceleration b _x formed in the block is divided in block 107 by the gravitational acceleration g, which leads to the μ value μ (λ _Ref ) listed in equation 7 when the controller is switched off.

The slip value λvr is further processed in block 108 according to equation 8 to the estimated value μ * (λ _vr ), for which purpose the setpoint slip λ _s (block 106 ) and the value in block 108

(Block 111 ) is supplied, which can be predetermined as a vehicle parameter.

At node 109 , when the controller is switched off, the μ-value μ (λ _Ref ) is additively linked to the estimated value μ * (λ _vr ) according to equation 9, which leads to the estimated coefficient of friction (λ _vr ), which is fed to block 105 .

The target brake pressure u _s is formed in block 112 according to equation 4, for which purpose the vehicle parameter a and the target slip λ _{s are} supplied to this block. The value μ (λ _s ) required in equation 4 is obtained according to equations 8 and 9 by the additive combination of μ (λ _Ref ) and

acquired. The setpoint brake pressure u _s , like the slip value differentiated in block 104 , is supplied to block 105 .

In block 105 , the value s (λ _vr ; is formed from the input variables in accordance with equation 5 or 6 (depending on the value of the slip λvr) and is fed to block 113 ( FIG. 1b), which determines a brake pressure change Δp in accordance with the above-mentioned control law. This change in brake pressure .DELTA.p is converted in block 114 by means of an inverse pressure controller into control times t _{I / O} for the inlet and outlet valves 115 of the brake system.

Claims (10)

1. Brake system for setting a predeterminable brake slip (λ _s ) on at least one wheel having brake means in a motor vehicle, wherein

• Speed values (Nÿ) are recorded, which represent the movements of the vehicle wheels, preferably the wheel speeds,
• a reference variable (V _ref ) representing the longitudinal vehicle speed is determined as a function of the detected speed values,
• a slip variable (λ _ist ) representing the instantaneous slip on the wheel _{is determined as a} function of the detected speed values and the determined reference variable,
• a target value (u _s ) for a braking effect is determined as a function of the determined reference variable (V _ref ) and the target slip to be set (λ _s ),
• - The predeterminable brake slip (λ _s ) is set to the determined target value (u _s ) by setting (Δp) the braking effect.

2. A braking system according to claim 1, characterized in that the instantaneous braking effect on the wheel which represents an actual value _(u) is determined, and terminating (Ap) of the braking effect on the determined target value (u _s), the specific actual value _(u) with the determined target value (u _s ) is compared. 3. Brake system according to claim 2, characterized in that a coefficient of friction (λ) representing the coefficient of friction of the vehicle wheel is determined depending on the determined reference variable (V _ref ) and the actual variable (u _ist ) is dependent on the determined coefficient of friction and the change over time of the determined one Slip size (λ _ist ) is determined. 4. A braking system according to claim 1, characterized in that _(is λ) at least depending on the particular slip amount and the change over time of a switching value is formed, and terminating (Ap) of the braking effect on the determined target value (u _s), the braking effect depends on the formed switching value is made. 5. Braking system according to claim 1, characterized in that the setting of the braking effect by a change (Δp) of the brake pressure in the braking means ( 115 ). 6. Braking system according to claim 2, characterized in that to adjust (Δp) the braking effect to the specific setpoint (u _s ), the braking effect is changed, a change (Δp) is only made when the specific actual variable (u _is ) deviates from the determined value (u _s ) in a specifiable manner. 7. Braking system according to claim 4, characterized in that to adjust (Δp) the braking effect on the specific setpoint (u _s ), the braking effect is changed, a change (Δp) is only made when the switching value formed from a predetermined setpoint switching value (0) deviates in a predefinable manner. 8. Braking system according to claim 4, characterized in that a coefficient of friction (λ) representing the coefficient of friction of the vehicle wheel is determined depending on the determined reference variable (V _ref ) and the formation of the switching value is dependent on the determined coefficient of friction. 9. Brake system according to claim 3 or 8, characterized in that a deceleration value (b _x ) representing the vehicle deceleration is determined as a function of the determined reference quantity (V _ref ) to form the friction quantity (λ) representing the coefficient of friction of the vehicle wheel. 10. Brake system according to claim 4, characterized in that the formation of the switching value is dependent on a comparison of the specific slip size (λ _is ) with the predetermined brake slip (λ _s ).