DE19954697C2 - Method for correcting the digital or digitized signal of a displacement or angle sensor - Google Patents

Description

The invention relates to a method for correcting the digital or di gitalized signals (actual value) of a displacement or angle sensor deviation of the ACTUAL Value of its TARGET value.

Displacement or angle sensors are used a lot in industrial measurement technology used several times. Among other things, angle sensors are also used in the event of force vehicles installed to determine the steering angle. The steering angle or the steering angle in motor vehicles is required to with to be able to apply this value to a vehicle dynamics control system NEN. Such a driving dynamics control system receives in addition to the above Steering angle values further measurement data, such as the wheel speed or the Dre hung the motor vehicle about its vertical axis. Are needed for one is the absolute steering angle and the other is the steering wheel speed so that these values could be recorded together with the others Data evaluated by the vehicle dynamics control system and for control of actuators, for example the brakes and / or engine management can be implemented. As steering angle sensors are at for example, optoelectronic or electrically operating.  

The requirements for the accuracy of determining the required Steering angle with regard to the resolving power of the steering angle sors are not inconsiderable. These requirements can be found below different ways are encountered, it being within this Description does not matter in what way the gefor resolution is realized. Every path or angle shot mer has its own characteristic curve, which is usually idealized by the th signal curve deviates. This applies particularly to such or angle sensors, which are provided to work electrically. Such trained transducers are designed as potentiometers, at which one or more grinders are used as scanning elements. The Deviation of the characteristic of such a transducer can in the case of a Angle sensor such a measure of the idealized signal curve have, which is larger than that originally for such an angle runners designed resolving power so that this again level sert can be. So far, such deviations have not been corrected been carried out, at least not when using such a recording mers as a steering angle sensor in a motor vehicle, in which as a rule only a very low computing and storage capacity is available.

DE 42 20 885 describes a method for evaluating the output Signals from an angle sensor to determine the steering angle position known, in which to record the entire possible range of rotation a rough signal is generated on the steering column. To increase the This previously known method has two resolving powers fine signals repeating over the entire rotation range needed. As a result, problems with the evaluation of the fine signals are problematic of repeating itself over the entire rotation range of the resulting maxima and minima.

In order to also evaluate the area of the maxima and minima, the Ge Subject of this document provided that the two fine signals run offset to each other and that always that fine signal in the further evaluation arrives, which is the better, with regard to its evaluator The result is less problematic and therefore not in that linear range of the maxima and minima is located. In the case of the According to the previously known method, an admissibility range is defined in the fine signal that comes into the subsequent evaluation  must be located. If a determined fine signal is not in the inlet range, the other reaches due to the angular misalignment in the admiss fine range in the evaluation. The further Of The content of this document only relates to procedures steps, in what way a selection in the further Fine signals are evaluated.

The method described in DE 42 20 885 A1 is an angle sensor according to DE 40 18 187 A1, based on the inductively operating is seen. In contrast to such angle transducers, which according to Art of a potentiometer are designed to work, is the sensor characteristics dependent errors less. A correction of the fine signal with regard on the angle sensor's own characteristic is the subject of DE 42 20 885 A1 not provided.

A correction of a determined steering angle is from DE 41 30 142 A1 known. A correction is made to the subject of this document the error that occurs during assembly with regard to a not agreement of the straight position of the wheels and that actually angle detected in this wheel position, which ideally amounts to 0 °. This The determined correction angle value is stored so that everyone can follow it determined steering angle value corrected with this correction angle value that can. A correction of the signal determined by the angle sensor depending on the characteristic curve of the sensor, the Ge also was not the subject of this document.

Another method for operating an optical steering angle sensor is known from DE 195 06 019 A1. In this document is a comparison routine described, according to which a current code word precedes compared with regard to a match. In However, this document is a correction procedure for correcting the measured values not dependent on the transducer characteristic wrote.

The post-published DE 198 31 960 A1 discloses a displacement sensor, and a method for measuring a change in path. That in this Thursday The method described in the document proposes to use digital scanning Rough determination of the path change to use and for fine determination  to scan an incremental track analogously to the change in path. Furthermore, fenbart, the offset between rough and fine determination in the calc The output signal must be taken into account and by a dynami characteristic control always work the scanning units in their Ensure the time of work.

The invention lies on the basis of this prior art discussed hence the task of preparing a method mentioned at the beginning with which a sufficient correction of the actual value regarding deviation of its transducer characteristics from its TARGET Value is possible, this correction method in particular only one should require little computing and storage effort and thus above al lem is suitable for use in motor vehicles.

According to the invention, this object is achieved by a method for correction the digital or digitized signal (actual value) of a path or Angle sensor dependent on the sensor characteristics Ab deviation of the ACTUAL value from its TARGET value, in which in ei In the first step, the signal curve of the at least one, the recording mer assigned sensing element in several, each by a signal dependent offset marked sectors is divided and at which, when operating the transducer, each actual value depending on its Part of the maximum signal assigned to a sector and around it Sector characteristic offset value is corrected, the sector be drawn assignment of an actual value by means of a threshold value comparison ches is carried out, and then the corrected actual Value for determining the position of the sensing element within the measuring area realm of the transducer is evaluated.  

In the correction method according to the invention, there is a first Step a sectorization of the waveform. Every sector extends over a certain section of the entire measuring range of the recording mers. The sector formation is signal dependent, with each sector being one only offset value is assigned to correct an actual value. at a sectorization of the signal curve of a steering angle sensor, at for example in eight sectors, extending over a total of 360 ° with the provision of a single scanning element, for example a loop he only needs a memory requirement for these eight offset values. at Providing two scanning elements, one becomes twice as large Required memory requirements.

The offset value characterizing each sector is determined Signal dependent, so that in a sector with this offset value to kor rigorous signal curve a sufficient approximation to the ideali sied waveform within this sector. Such Sector formation is carried out when the transducer is started up for the first time led, in which the characteristic curve is also recorded. The determined sparkling wine Ren and the corresponding offset values are for the operation of the open in a memory, for example in an EPROM or a EEPROM stored.

When the transducer is in operation, each ascertained actual value is assigned to a sector depending on its share of the maximum signal in the claimed method. The actual value is the digital output signal of the sensing element in incremental sensors or the digitized output signal of an analog sensing element, the output signal of which has been digitized. The maximum signal depends on the system used and is 256 digits for an 8-bit system, for example. After the actual value has been assigned to a sector, this is corrected by the offset value characterizing this sector. Then the corrected and thus approximated to the TARGET value with sufficient accuracy actual value for determining the position of the scanning member within the measuring range of the sensor is evaluated. In this way, when using such a sensor as a steering angle sensor in a motor vehicle, the steering angle is determined where, in the case of a characteristic-dependent deviation of the actual value, the dimensional accuracy of the steering angle sensor is not influenced or is influenced only to a minor extent.

The sector-specific assignment of an actual value takes place by means of a or several threshold value comparisons, each sector being marked by an An partial interval is characterized on the maximum signal. When using several, for example two scanning elements, for example grinders, the actual value is determined by considering the shares on the maximum signal of both sensing elements.

Plausibility checks can be carried out when performing the inventive method correct correction method are used, with such Plausibility check of both the maximum possible offset value to the cor rectification or when capturing undefined results Values can be specified.

Further refinements and advantages of the invention are part of it further subclaims and the following description of a Embodiment with reference to the accompanying figures.

Show it:

Fig. 1: A schematic flow chart for describing a correction method,

. FIG. 2 is a diagram illustrating the waveform of two the receiver of Figure 1 associated with Abtastorganen ei ner sector division of the waveform,

Fig. 3: The corrected angular error of the waveform of the Transd mer of Figure 1.

FIG. 4: The signal curve of the transducers of FIG. 1 without a correction which is dependent on the characteristic curve and

Fig. 5: A diagram of another angle sensor with a curve showing the uncorrected and the corrected angle error curve.

An angle sensor 1 of a steering angle sensor is designed to work electrically and comprises a potentiometer 2 extending over 360 °, on which two wipers 3 , 4 are arranged with an offset of 90 °. The output signals of the transducer 1 act on an A / D converter 5 , which converts the analog output signals of the transducer 1 into digital signals. In the illustrated embodiment, an 8-bit converter is used. The A / D converter 5 acts on the output side of a weighting element 6 , by means of which the signal profiles of the wipers 3 , 4 are weighted as a function of the characteristic. Furthermore, the A / D converter 5 acts on an assignment element 7 , which is used both for dividing the signal curves in sectors and for assigning detected actual values to a sector. Depending on the design of the signal curves, an offset value is assigned to each sector for correction. This offset value is stored in a memory 8 , for example an EPROM. By dividing the signal curve into individual sectors, each of which is assigned a single offset value for correction, a sufficient approximation of an ACTUAL value assigned to this sector is possible after the correction has been carried out on an idealized signal curve, without a corresponding one large storage capacity would be necessary. The same applies to the computing effort.

This first sectorization of the signal profiles, as described above, carried out during the initialization of the transducer. Such sectori However, it can also be set at a later time, for example in regular time intervals are carried out, for example wise changes in the characteristic due to aging or temperature or to take into account the signal curves.

When the transducer 1 is in operation, the determined actual values of the grinders 3 , 4 are each assigned to a sector by means of a threshold value comparison method. Depending on the proportion of the actual value of a grinder 3 or 4 , the value determined by both grinders 3 , 4 is assigned to one of the sectors intended for the transducer. This sector assignment of the two grinders 3 , 4 takes place according to the following conditions:

• - S1 <64 (A / D value for ¼Vcc) and S2 <= 128 (A / D value for ½Vcc): sector 8
• - S1 <64 (A / D value for ¼Vcc) and S2 <128 (A / D value for ½Vcc): sector 1
• - S1 <128 (A / D value for ½Vcc) and S2 <= 128 (A / D value for ½Vcc): sector 7
• - S1 <128 (A / D value for ½Vcc) and S2 <128 (A / D value for ½Vcc): sector 2
• - S1 <192 (A / D value for ¾Vcc) and S2 <= 128 (A / D value for ½Vcc): sector 6
• - S1 <192 (A / D value for ¾Vcc) and S2 <128 (A / D value for ½Vcc): sector 3
• - S1 <256 (A / D value for Vcc) and S2 <= 128 (A / D value for ½Vcc): sector 5
• - S1 <256 (A / D value for Vcc) and S2 <128 (A / D value for ½Vcc): sector 4

In the exemplary embodiment described, the signal curve is divided into eight sectors. After an ACTUAL value has been assigned to a sector in the assignment element 7 , the ACTUAL value is corrected in a correction stage 9 by the offset value assigned to this sector. Not shown in FIG. 1 is a plausibility check with a view to the correction to be carried out, which precedes the actual correction. The corrected actual value is formed in a measured value formation device 10 as a function of the weighting of the actual value in the signal curve and then output at an interface 11 for further processing, for example for applying a driving dynamics control system. The ACTUAL value corrected on the output side in the measured value generating device 10 is matched to the theoretical TARGET value with sufficient accuracy.

Fig. 2 is a diagram showing the waveform of the grinder 3, 4 and the division of about 360 ° extending path into eight sectors. The sectors are marked with different offset values, so that each determined ACTUAL value can be corrected by the offset value assigned to a sector. A diagram with the corrected error, which is an angular error in the exemplary embodiment shown, is shown in FIG. 3. From the diagram it is clear that the ideal "0" marked angle error curve is clearly approximated, in particular with regard to the signal curve if this is shown uncorrected, as is shown for comparison in FIG. 4.

Fig. 5 shows the signal curve of an uncorrected sensor with respect to its characteristic and a signal curve with the signal curve corrected according to the method described above. This further example also makes it clear that the error deviation due to the characteristic curve of the transducer is significantly reduced by the corrective measures.

Depending on the intended purpose, a Sectorization carried out in a different number of sectors be, the number of sectors influences the claim to the Has accuracy of correction.  

Compilation of the reference symbols

1

Angle sensors

2

potentiometer

3

grinder

4

grinder

5

A / D converter

6

weighting element

7

association section

8th

Storage

9

correction level

10

Meßwertbildungseinrichtung

11

interface

Claims (5)

1. Procedure for correcting the digital or digitized Signa les (actual value) of a displacement or angle sensor with respect to the Sensor characteristic curve-dependent deviation of the actual value from its TARGET value, in which the signal ver run of the at least one sensor assigned to the pickup go into several offset values, each dependent on a signal marked sectors is divided and in the case of operation of the transducer each actual value depending on its share in Maximum signal assigned to a sector and around that sector characteristic offset value is corrected, the sector bezo assignment of an actual value by means of a threshold value ver the same is carried out, and in which the corri gated actual value for determining the position of the sensing element inside half of the measuring range of the transducer is evaluated. 2. The method according to claim 1, characterized in that for Determining the signal dependent sectors depending on the waveform of the sensing element is weighted. 3. The method according to claim 1 or 2, characterized in that a signal acquisition with several, expediently offset mutually arranged scanning organs and this is an actual Value is made up of signal components of each scanning element. 4. The method according to any one of claims 1 to 3, characterized indicates that a maximum offset value is specified. 5. The method according to any one of claims 1 to 4, characterized records that before the step of performing the correction Plausibility check with regard to the intended correction measure is carried out.