GB2297619A

GB2297619A - Method of predicting vehicle standstill

Info

Publication number: GB2297619A
Application number: GB9601610A
Authority: GB
Inventors: Dagobert Masur
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 1995-02-02
Filing date: 1996-01-26
Publication date: 1996-08-07
Anticipated expiration: 2016-01-26
Also published as: DE19503270A1; GB9601610D0; GB2297619B; DE19503270C2

Description

- 1 METHOD OF PREDICTING VEHICLE STANDSTILL 2297619 The present invention

relates to a method of determining a time point when a vehicle becomes stationary.

It is necessary to determine the time point when a vehicle becomes stationary for cases where information on this time point is to be further evaluated. For example, information on the stationary state of the vehicle is required for use in implementation of an anti-roll function. After the vehicle has attained a stationary state, an anti-roll device serves to apply a parking brake, without further interventation of the driver, until the vehicle is started again. For reasons of safety and driving comfort, a build up of braking pressure at wheels at which braking is to be applied should not take place until the vehicle is stationary. The smallest, reliably measurable wheel speed is the threshold speed, so that the time when the vehicle becomes stationary cannot be measured precisely.

When controlling an internal combustion engine it is known, according to EP 0 490 088 A2, to identify crankshaft angle position by means of an absolute value angle sensor which comprises a wheel which is mounted on the crankshaft and bears on its surface a code which is scanned by a sensor. The electrical signals acquired during this process are fed to an engine stationary-state detection circuit, in an input stage of which the input signal is electronically differentiated and fed to a counter, which in the absence of an input signal, generates an output signal corresponding to the engine state. Such a procedure could not provide an exact result for the actual time when the vehicle itself becomes stationary.

There is therefore a need for a method for determining a time point when a vehicle becomes stationary, which method may allow for the requirements of safety and driving comfort and requires only minimal outlay.

According to a first aspect of the present invention there is provided a method of determining a time point when a vehicle becomes stationary, comprising the steps of determining a threshold speed value for at least one wheel of the vehicle during a braking phase, subsequently detecting at least one further speed value for the at least one wheel at a plurality of successive time points until attainment of a measurable minimum speed, and estimating the time point for the stationary state of the vehicle by forming at least one speed gradient between such speed values and by extrapolation thereof is to the zero speed.

According to a second aspect of the invention there is provided a method of operating a vehicle with an anti-roll control, comprising the steps of performing the method according to the first aspect of the invention and causing the parking brake of the vehicle to be automatically applied at the estimated time point and maintained until the vehicle is restarted.

It is therefore provided that a threshold speed value is determined for at least one wheel during a braking phase, that, subsequently, at least one further speed value for the wheel or wheels is identified at a plurality of successive times until a measurable minimum speed is reached, and that the time far the stationary state of the vehicle is estimated by forming one or more s peed gradients between the identified speed values and by extrapolating them to the zero speed. By forming the speed gradient or gradients between the identified speed values and by extrapolating them to the zero speed, a time for when the vehicle becomes stationary is obtained; this time satisfactorily represents the actual time - which cannot be measured precisely - when the vehicle becomes stationary. If this time is used for, for example, the time of application of a vehicle parking brake, jerky braking is avoided and the parking brake is applied sufficiently quickly.

The accuracy of the estimated value of the time when the vehicle becomes stationary can be improved by extrapolating the average speed gradient between the threshold speed value and the last speed value identified and/or by extrapolating the instantaneous speed gradient between the next-to-last speed value identified and the last speed value identified.

The result of the estimation of the time when the vehicle becomes stationary becomes even more reliable if there is provision that at least the last two instantaneous speed gradients are compared with the two corresponding average speed gradients and, if these two instantaneous speed gradients either exceed or drop below the corresponding average speed gradients by a fixed minimum amount, the instantaneous speed gradient or, otherwise, the average speed gradient is used to estimate the time. Preferably, the average speed gradients and the instantaneous speed gradients are identified at at least three wheels of the vehicle and the maximum of the estimated stat i onary- state times of the wheels is selected as the estimated time when the vehicle becomes stationary.

The estimated result can be optimised by using the instantaneous speed values instead of the average speed values when the at least last two instantaneous speed gradients exceed or drop below the corresponding average speed gradients by the fixed minimum amount at at least three wheels.

By virtue of these measures, random influences which may affect the estimated value are virtually completely eliminated.

The method is advantageously used in a motor vehicle with an anti-roll function, the parking brake of the vehicle being automatically applied at the estimated time and maintained until the vehicle is restarted. Safety and driving comfort during braking are ensured in that early, jerky activation of the parking brake system is ruled out, but the process of applying the parking brake does not take place too late.

is Preferably, a control unit of an anti-lock brake system/traction control system is triggered at the estimated time and the vehicle driving wheels have the necessary braking pressure applied to them via valves of the system, the anti-roll function preferably being triggered by the driver activating the brake and when the engine is disengaged or idling. By virtue of these measures, the method of predicting vehicle standstill is integrated to an optimum degree into an anti-lock brake system/traction control system in order to implement the anti-roll function.

Examples of present invention will now be more particularly described with reference to the accompanying drawing, the single figure of which is a diagram showing the speed profile v of a wheel, plotted against time t and a number of average speed gradients and instantaneous speed gradients.

Referring now to the drawing, there are shown speed gradients which are formed by detecting speed values vo.) vi v N-2t v N-11 v N = v min for a wheel speed curve a at measurement times tos t 1, t N-21 t N-1 and t N 1, wherein v 0 is a threshold value and v min is last measurable wheel speed. The curve a can be obtained from,, for example, an actual speed curve a' by smoothing, using a suitable filter method if the actual speed curve a' is not sufficiently smooth for identifying the speed gradients.

The formation of speed gradients is initiated as soon as the speed of a wheel drops below a defined threshold speed value. For each wheel the average speed gradients m m and m between the 1 - N-1 N defined threshold value v 0 and the actual speed value v 1, v N-1 and v N are determined and extrapolated as far as the zero speed in order to obtain a first reference value for a time t S1 - t SN-1 and t SN when the vehicle becomes stationary. This reference value is updated for each wheel up to the last measurable wheel speed. On failure of a rotational speed sensor to provide a speed signal, the last detected time t SN serves as a provisional time for when the associated wheel becomes stationary. On failure of the last sensor signal, the maximum of the calculated stationary-state times for the wheels is used as an estimated value for the expected time when the vehicle becomes stationary.

A "fine estimation", which allows for change in the speed gradient and is based on the formation of an instantaneous speed gradient P,... IJN-11 PNI runs in parallel with this procedure. This procedure also starts at the time t 0 when the speed drops below - 6 the threshold speed value v 0 In contrast with the procedure described above, the reference speed and reference time for the formation of gradients do not remain constant but are always replaced by the actual wheel speed and associated times after a time period, which is to be fixed, has expired. Otherwise, the described procedure is maintained for the calculation of the new stationarystate times tSF1 tSFN- 1 and t SW Instead of the average speed gradient, this procedure supplies the instantaneous speed gradient in approximated form.

If the n last instantaneous speed gradients ON' PN-i at at least three wheels either drop below or exceed the average speed gradients MN1) m N-1 by a fixed minimum amount, the instantaneous speed gradients are used instead of the average speed gradients for estimating the time t SFN Otherwise, the average speed gradients are utilised to estimate the time t SN when the vehicle becomes stationary.

The described gradient method permits the time when the vehicle becomes stationary, which cannot be detected by technical measuring means, to be predicted.

The estimated time t,, or tSFN when the vehicle becomes stationary can be used as a basis for automatic signal issue or control. An important application is, for example, the implementation of an anti- roll function for motor vehicles, which function is monitored and triggered by means of a control unit for an anti-lock brake system/traction control system. After the vehicle has been braked to a stationary state, an antiroll device serves to apply a parking brake, without the intervention of the driver, and maintained until the vehicle is started again. For this purpose, the driving wheels of the vehicle have the necessary braking pressure applied to them via the valves of the traction control or anti-lock brake system. At the same time, for reasons of safety and driving comfort, the build up of pressure should not take place until the vehicle is stationary, for which the estimated time when the vehicle becomes stationary, identified according to the method described above, is advantageous.

Claims

1. A method of determining a time point when a vehicle becomes stationary, comprising the steps of determining a threshold speed value for at least one wheel of the vehicle during a braking phase, subsequently detecting at least one further speed value for the at least one wheel at a plurality of successive time points until attainment of a measurable minimum speed, and estimating the time point for the stationary state of the vehicle by forming at least one speed gradient between such speed values and by extrapolation thereof to the zero speed.

2. A method as claimed in claim 1, wherein the step of estimating comprises extrapolating the average speed gradient between the threshold speed value and the last detected speed value andlor extrapolating the instantaneous speed gradient between the is penultimate and the last detected speed values.

3. A method as claimed in claim 1 or 2, wherein at least the last two instantaneous speed gradients are compared with the two corresponding average speed gradients and if said two instantaneous speed gradientss exceed or fall below the corresponding average speed gradients by a fixed minimum amount the instantaneous speed gradient and, otherwise, the average speed gradient is used to estimate the time point for the stationary state.

4. A method as claimed in claim 2 or claim 3, wherein the average speed gradients and the instantaneous speed gradients are formed from speed values detected at at least three wheels of the vehicle and the maximum of the stationary-state time points estimated from the values applicable to the individual wheels is selected as the estimated value for the expected time point when the vehicle becomes stationary.

5. A method as claimed in claim 3 when appended to claim 4, wherein the instantaneous speed values are used instead of the average speed values when said two instantaneous speed gradients exceed or fall below the corresponding average speed gradients by the fixed minimum amount at said at least three wheels.

6. A method as claimed in claim 2 and substantially as hereinbefore described with reference to the accompanying drawing.

7. A method of operating a vehicle with an anti-roll control, comprising the steps of performing the method as claimed in any one of the preceding claims and causing the parking brake of the vehicle to be automatically applied at the estimated time point and maintained until the vehicle is restarted.

8. Method as claimed in claim 7, wherein a control unit of an anti- lock brake system/traction control system of the vehicle is triggered at the estimated time point and a braking pressure is applied to the driving wheels of the vehicle by way of valves of the system.

-

9. Method as claimed in claim 7 or claim 8, wherein the anti-roll function is triggered by the vehicle driver actuating the vehicle brake and when the vehicle engine is disengaged from the vehicle driving wheels or idling.