DEVICE AND METHOD FOR DETERMINING POSITION, LENGTH OR ANGLE

Description

[0001] The present invention relates to a device for determining position, length, or angle, and to a corresponding method. For prior art, reference is made to DE 10 2016 101 965 A1 and WO 2017 / 043 249 A1.

[0002] Devices for determining position, length, or angle, also known as encoders, and the corresponding methods can be used in a wide variety of technical fields. For example, such devices can be used in machine tools, enabling the measurement of a tool's position or angle relative to the workpiece. Other applications include rotary angle sensors, for example, for motor feedback systems.

[0003] Conventionally, such devices comprise a first and a second part, which are movable relative to each other. A code with a plurality of code segments of the first and second type is arranged on the first part. A readout device for detecting at least a part of the code is attached to the second part, the readout device comprising several sensors, each configured to detect the individual code segments and outputting a corresponding output signal. Furthermore, such devices comprise a measurement signal generation unit configured to generate at least one corresponding measurement signal from the respective output signals, and a position determination unit configured to determine a relative position between the first and second parts, taking into account, and in particular based on, the at least one measurement signal.

[0004] Numerous factors can negatively impact the accuracy of position determination. These range from systematic errors or variations, such as differences in the precise positioning of individual sensors and / or their sensitivity, to randomly occurring individual errors, such as bit errors in the analog-to-digital converters, and even to aging-related variations, for example, a decrease in sensor sensitivity over their lifespan. Such factors can be corrected separately and / or collectively in a variety of ways.

[0005] One object of the present invention is to present devices and methods which correct at least a selection of these influences.

[0006] This problem is solved by devices and methods according to the independent claims. Further developments of these are described in the dependent claims.

[0007] The device according to the invention is characterized in that it further comprises a correction unit configured to effect, in particular directly or indirectly, an influence on the amplitude of the at least one measurement signal before it is used by the position determination unit to determine the relative position. The correction unit is further configured to effect a predefined, in particular constant and / or uniform, influence on the measurement signals during a calibration run, and to generate a calibration signal from the sensor output signals during the calibration run, to compare the course of the generated calibration signal with a predetermined reference course, and to generate a correction table and / or a correction function from the result of the comparison.Finally, the correction unit is designed to influence the course of at least one measurement signal beyond the calibration run, in accordance with the generated correction table and / or correction function.

[0008] According to the invention, a calibration run is performed in which the influence on the measurement signals is known, and this influence preferably remains constant or unchanged throughout the entire calibration run. Deviations from an ideal curve can then be detected in the calibration signal thus generated. These deviations can then be compensated for in subsequent operation (i.e., after the calibration run) using the correction table and / or the correction function.

[0009] During the calibration run, for example, at least one complete rotation can be performed with a rotary encoder. With a linear encoder, the entire travel range can be traversed at least once. With periodic coding, it is also possible to traverse only a portion of all periods, in particular just a single period. Generally, every relative position between the first and second parts can therefore be assumed at least once during the calibration run.

[0010] The shape of the measurement signal can be influenced in various ways, both directly and indirectly. For example, in an optical encoder with photodiodes as sensors, an optically effective coding system, and associated illumination, the shape of the measurement signal can be indirectly influenced by varying the illuminance. This results, for instance, in a variation of the amplitude of the measured values ​​at different times.

[0011] Accordingly, the calibration run can be carried out, for example, with constant illuminance by means of lighting controlled by the correction unit.

[0012] Regarding the measurement signal generation unit, it should be noted that it can also be designed as an analog-to-digital converter and thus only digitize analog output signals. In such a configuration, the measurement signals can be influenced, for example, directly by applying a base signal to the output signals before their digitization, or by varying the threshold values ​​used by the analog-to-digital converter to digitize the output signals.

[0013] The predefined influence during the calibration run can also take the form of a complete absence of influence. The main point here is that the calibration run is performed under the most clearly defined and specific conditions possible. This is necessary to actually generate a suitable correction function and / or correction table from the output signals of the calibration run.

[0014] The calibration signal can, for example, be a specific combination of analog or digital output signals from different sensors. Alternatively, the calibration signal can simply be the analog or digital output signal of a single sensor, particularly in the form of a reference sensor.

[0015] The reference curve mentioned shows a curve that is assumed to be optimal or a desired curve, which results, for example, from a simulation of the device while neglecting any undesirable influences. It therefore shows what curve the calibration signal should ideally exhibit.

[0016] The correction table and / or correction function ultimately generated serves, figuratively speaking, as the basis for generating a correction signal. This signal, output by the correction unit, influences the measurement signals in a specific way. As in the examples above, this influence can affect the measurement signals both indirectly and directly. With optical sensors, an indirect influence on the measurement signals can be achieved through variations in illuminance, while no further influence on the measurement signal occurs, for example, through variations in threshold values ​​in an analog-to-digital converter. A direct influence on the measurement signals can be achieved through variations in the threshold values ​​of an analog-to-digital converter and allows for the elimination of indirect influence on the measurement signals, particularly through variations in illuminance. This variant is particularly useful when two encodings are arranged side-by-side.Coding tracks, specifically an incremental track and an absolute track, are relevant when both are illuminated simultaneously. A variance in the combined illuminance, based on a correction table and / or function derived from the measurement signals of one track, could lead to undesirable variations in the measurement signals of the other track. The effect indicated by the correction table and / or function serves to compensate for unwanted influences, which are indicated by deviations in the calibration signal from the reference signal, and thus correct the measurement signals. These corrected measurement signals can then be used by the position determination unit for coarse and / or fine determination of the relative position, enabling a more accurate measurement result.

[0017] The special feature of this type of correction is that it directly influences the measurement signals and does not involve any correction within or after the position detection unit. This makes the correction according to the invention relatively easy to implement and even retrofittable to devices from the prior art.

[0018] In particular, the influence on the measurement signal displayed or defined by the correction table and / or function depends on previously determined information about the relative position between the two parts. In other words, the correction to be made preferably depends on an expected or estimated relative position at the time of the correction.

[0019] It should also be noted that the calibration run can be performed not only before commissioning the device. During the calibration run, the applied correction of the measurement signals is deactivated, and the illumination is kept constant. After a calibration run of, in particular, one period in length, the system switches back to the regulated "normal operation," at which point the corrected correction table and / or correction function is then applied.

[0020] Preferably, the calibration signal is the sum of two squared output signals or the square root of such a sum, and the reference curve is a constant.

[0021] A constant reference waveform for a corresponding calibration signal can be achieved, for example, if both output signals have a sinusoidal waveform and a phase shift of 90° to each other. The so-called vector length indicated by this calibration signal is a relatively easy-to-obtain and evaluate basis for generating a corresponding correction table and / or function. The calibration signal can also be the sum of the magnitudes of corresponding output signals. Finally, the calibration signal can also be proportional to the sum of two squared output signals, to the square root of such a sum, and / or to the sum of the magnitudes of corresponding output signals.

[0022] Preferably, the calibration signal corresponds to a control signal which is intended to serve as the basis for effecting the intended influence on the at least one measurement signal beyond the calibration run.

[0023] Specifically, the calibration signal is not a freely chosen combination of output signals, but rather a specific combination of output signals from special sensors, which also directly serves as the basis for the intended manipulation. This significantly simplifies the creation and application of the correction table and / or function. An example of this is the vector length mentioned earlier.

[0024] Preferably, the correction unit is designed to generate the correction table and / or correction function in such a way that the resulting influence causes a corresponding calibration signal to exhibit the specified reference curve.

[0025] This makes the correction table and / or function directly suitable for correcting undesirable influences during the calibration run, indicated by deviations of the calibration signal from the reference curve. Influences already present during the calibration run can then be reliably corrected for subsequent operation of the device. Consequently, these influences, which were already present at the time of calibration, can be largely disregarded when identifying and correcting new influences, such as aging effects, during subsequent operation of the device, and do not need to be laboriously separated from the newly occurring influences. Undesirable waveforms of the measurement signals in subsequent operation, i.e., beyond the calibration run, thus always indicate newly occurring influences caused by an actual change in certain properties of the device or its environment.This facilitates the analysis and identification, and if necessary, the correction of newly emerging influences.

[0026] Preferably, the correction unit is configured to generate the correction table and / or correction function such that, when applied to the output signals, particularly directly or indirectly during digitization, it yields at least one adapted measurement signal. The position determination unit is configured to use the at least one adapted measurement signal to determine the relative position beyond the calibration process.

[0027] As mentioned above, the output signals can be directly influenced by applying a base signal. Analog relative amplification or attenuation of the output signals is also possible. This influence can affect all output signals equally or be sensor-specific. Indirect influence on the output signals is achieved, in particular, by varying the threshold values ​​used to digitize the output signals. By then using the adapted measurement signals to determine the relative position, the position determination unit can operate in the conventional manner. Specifically, the position determination unit does not need any knowledge of the correction table and / or function to generate corrected position information.

[0028] Preferably, the position determination unit is designed to use a readjustment table and / or function when determining the relative position, which differs from the correction table and / or correction function generated by the correction unit.

[0029] In other words, there is a recalibration table and / or function, distinct from the previously described correction table and / or function, which does not serve to influence the measurement signals themselves, but rather to modify the evaluation of the measurement signals by the position determination unit. Such recalibration tables and / or functions, which are known per se, serve to correct influences that are not indicated or correctable by the correction table and / or function described above. Additional consideration of the aforementioned recalibration table and / or function enables an even more precise determination of the relative position.

[0030] Preferably, the two parts are only movable translationally or rotationally relative to each other.

[0031] Such relative movements are particularly easy and reliable to evaluate.

[0032] Preferably, the sensors are optical sensors, for example in the form of photodiodes; capacitive sensors, for example in the form of capacitors and voltmeters; inductive sensors, for example in the form of magnetic coils and ammeters; or magnetic sensors, for example in the form of Hall sensors. The encoding is designed accordingly, in particular at least optically, for example in the form of a strip with black and white sections of different reflectivity or in the form of sections with different transmissivity; capacitively, for example in the form of a series of capacitors with different charge levels; or magnetically, for example in the form of a series of permanent magnets.

[0033] The different types of sensors and coding offer different advantages and disadvantages, which allow for a particularly targeted alignment of the device with the respective area of ​​application.

[0034] Preferably, the coding is a periodic coding, in particular an incremental coding.

[0035] In other words, it is a code with a pattern of corresponding code segments that repeats at predetermined intervals. Specifically, two different types of code segments can be alternately linked to form what is known as incremental coding. Signals from such devices are particularly easy to analyze and manipulate as desired.

[0036] Preferably, the correction unit is designed to generate the correction table and / or correction function in such a way that it maps and / or compensates for variations in the relative positioning of the different sensors and / or in the sensitivity of the different sensors.

[0037] The two examples mentioned are systematic influences which, according to the invention, are particularly easy to identify and compensate for.

[0038] A method according to the invention for determining the relative position between two parts movable relative to each other using the device described above comprises the following steps: performing a calibration run in the form of a relative movement of the two parts relative to each other while simultaneously outputting corresponding output signals from the sensors; generating a calibration signal from the sensor output signals; comparing the shape of the generated calibration signal with a predetermined reference shape; generating a correction table and / or a correction function from the result of the comparison; generating at least one measurement signal influenced according to the correction table and / or correction function from the output signals beyond the calibration run;and the determination of the relative position between the two parts, taking into account, in particular on the basis of, at least one influenced measurement signal.

[0039] This method allows for the correction of detected negative influences directly in the measurement signals, even before they are evaluated to determine the relative position. It is therefore particularly reliable and easy to implement.

[0040] Preferably, the calibration run includes a plurality of changes in the direction of the relative movement of the two parts relative to each other.

[0041] This makes it easier to identify systematic errors or deviations and to generate a corresponding correction table and / or function.

[0042] Preferably, the calibration run is performed under constant boundary conditions, in particular at constant temperature and / or lighting.

[0043] This minimizes avoidable variations and influences, preventing them from being addressed by the generated correction table and / or function. This allows the focus to remain on correcting systematic and / or structural errors or deviations.

[0044] Preferably, both a correction function and a correction table are generated. The correction table contains a plurality of values ​​which, when inserted into the correction function, specify it locally.

[0045] This makes it possible to define a correction function that exhibits a changing curve depending on the values ​​entered from the correction table. This allows for a particularly comprehensive specification of the desired impairment of the measurement signals.

[0046] Preferably, a correction table is generated which contains at least, and in particular exactly, one correction value for each identifiable relative position between the two parts.

[0047] This enables a very specific, comprehensive correction of the measurement signals and thus a particularly accurate determination of the relative position.

[0048] The invention is described below by way of example only, with reference to the drawings. It shows: Fig. 1 schematically shows the basic structure of a device according to the prior art; Fig. 2 schematically shows a first modification according to the invention of the device shown in Fig. 1 the setup shown; Fig. 2B schematically the ideal course of two exemplary output signals and a reference course for an exemplary calibration signal generated therefrom; Fig. 2C schematically the actual course of two exemplary output signals and an exemplary calibration signal generated therefrom; Fig. 3 schematically a second modification according to the invention of the in Fig. 1 the structure shown; Fig. 4 schematically a third modification according to the invention of the structure shown in Fig. 1 shown setup;

[0049] Fig. 1 Figure 10 schematically shows the basic structure of devices 10 for determining position or length. The device 10 shown serves as an encoder, for example in a motor feedback system (not shown).

[0050] The device 10 comprises a first part 14 to which an encoding 12 in the form of an incremental encoding 12 is attached. The device 10 further comprises a second part 18 to which a readout device 16 is attached. The encoding 12 and the readout device 16 are attached to the first part 14 and to the second part 18 such that they move relative to each other together with the two parts 14 and 18. In the present example, the first part 14 and the second part 18—and thus the encoding 12 and the readout device 16—are purely translationally movable relative to each other (see the double arrow B). Purely rotational relative movements can also be readily implemented by a person skilled in the art.

[0051] The code 12 is formed by a multitude of consecutive code sections 22-0 to 22-9 of the first type (shown in white) and second type (shown in black). In Fig. 1 Only ten such code sections 22-0 to 22-9 are shown. Coding 12 can include further code sections to the left and / or right of the depicted code sections 22-0 to 22-9. This would allow a determination of the relative position between the two parts 14 and 18 over a larger area than is possible with the ten depicted code sections 22-0 to 22-9.

[0052] The readout device 16 comprises eight sensors 20-1 to 20-8, for example in the form of photodiodes, wherein the sensors 20-1 to 20-8 are arranged side by side along the coding 12. The sensors 20-1 to 20-8 are aligned with the coding 12 and designed to detect the different code sections 22-0 to 22-9 of the coding 12.

[0053] A light source 40 is provided to ensure uniform illumination of the code sections 22-0 to 22-9. This light source illuminates at least those code sections 22-0 to 22-9 that lie within the detection range of the sensors 20-1 to 20-8 (the area between the two dashed-dotted arrows). Each sensor 20-1 to 20-8 receives a specific amount of light reflected (or transmitted) by the respective code sections 22-0 to 22-9, depending on the type of code sections 22-0 to 22-9 within its detection range. The sensors 20-1 to 20-8 output a corresponding signal, for example, a voltage or current value.

[0054] To facilitate the evaluation of the measured values ​​from sensors 20-1 to 20-8, the code sections 22-0 to 22-9 are spatially identical to each other. It is assumed here that the detection range of each sensor 20-1 to 20-8 is half as wide as the individual code sections 22-0 to 22-9 are long. The sensors 20-1 to 20-8 are aligned with the coding 12 such that they form a continuous detection range (see the area between the dashed-dotted arrows in Figure 12). Fig. 1 ) map onto the coding 12, the length of which corresponds exactly to the length of a codeword of the coding 12. In other words, each code segment 22-4 to 22-6, which lies completely within the detection range of the sensors 20-1 to 20-8, is located within the detection range of at least two, in particular three, adjacent sensors 20-1 to 20-8. This enables a particularly fine scanning of the coding 12 and thus a particularly precise resolution of the relative position between the first part 14 and the second part 18.

[0055] A measurement signal generation unit 28 is connected downstream of the sensors 20-1 to 20-8 or the readout unit to generate corresponding measurement signals from the output signals. In the simplest case, the measurement signal generation unit 28 can be one or more analog-to-digital converters. These then generate digital measurement signals as digital representations of the output signals by regularly comparing the values ​​of the output signals with one or more threshold values.

[0056] These measurement signals can then be used in a known manner by a downstream position determination unit 24 to determine the current relative position between the two parts 14 and 18.

[0057] Typically, the rough determination of the relative position is based on measurement signals for an absolute encoding, while measurement signals for an incremental encoding are used for fine-tuning the determined relative position. In principle, however, alternative configurations are conceivable.

[0058] For example, it is possible that the sensors 20-1 to 20-8 are not evenly spaced apart due to manufacturing tolerances, or that the sensors have different sensitivities due to manufacturing.

[0059] The aim of the modifications to this basic structure described below is to provide a relatively simple and reliable way to enable the correction of such systematic errors without having to identify them in detail and / or having to adjust the evaluation of the measurement signals within the positioning unit.

[0060] As in the Figuren 2A bis 4 As shown, according to the invention a correction unit 30 is provided in addition to the known components and is integrated into the device in a suitable manner.

[0061] Referring to Fig. 2 The correction unit 30 can be connected on the input side to the measurement signal generation unit 28 and the position determination unit 24, and on the output side to the light source 40, or an associated control unit.

[0062] During the calibration process, under a specified, and in particular constant, illumination by the light source 40, output signals generated by the sensors 20-1 to 20-8 of the readout unit 16 are transmitted to the correction unit 30. The correction unit 30 generates a calibration signal from the received output signals and compares its shape with a specified reference shape.

[0063] The reference curve shows a curve that would be expected under ideal conditions (i.e., without systematic errors). In a configuration where the output signals of two specific sensors should exhibit a sinusoidal curve with a phase shift of 90° to each other (see the dashed and dash-dotted lines in [reference]). Fig. 2B The square root of the sum of the squares of the output signals could, for example, serve as a calibration signal. Such a calibration signal is particularly suitable as a control signal for the intended light source 40.

[0064] The reference curve would then be a constant (see the solid line in Fig. 2B If the two associated sensors are not positioned at the correct distance from each other, this leads to a phase shift between the two output signals that is different from 90° (see Fig. 2C The calibration signal derived from these output signals then exhibits a shape that deviates from the reference shape (compare the solid wavy line in [reference]). Fig. 2C with the solid line in Fig 2B From the determined deviation of the calibration signal from the reference curve, the correction unit then generates a correction table and / or function. This correction table and / or function is generated in such a way that it serves as the basis for influencing measurement signals, which are ultimately to be taken into account for determining the relative position.

[0065] Returning to Fig. 2A In the present example, it is planned to couple the correction unit 30 to the light source 40 on the output side and to achieve an impairment or modification of the measurement signals by means of a specific control of this.

[0066] For this purpose, the correction unit generates a correction table and / or function from the result of the calibration, which enables the light source 40 to be controlled in such a way that, under otherwise identical conditions, the calibration signal exhibits the desired constant curve during a subsequent measurement run. In particular, the correction unit 30 varies the brightness of the light source 40 based on the generated correction table and / or function such that the square root of the sum of the squares of the output signal of the associated sensors remains constant.

[0067] With appropriate manipulation, the described error in the relative positioning of the corresponding sensors is no longer reflected by the output signals and thus by the measurement signals, and therefore does not lead to an error in determining the relative position based on the new measurement signals.

[0068] To influence the measurement signals depending on the position, the correction unit 30 is coupled to the position determination unit 24 on the input side and determines the correction to be made from the correction table and / or function based on an input signal from the latter.

[0069] Instead of the indirect manipulation of the measurement signals described above by means of an indirect manipulation of the output signals via a variation of the illumination intensity, an indirect manipulation of the measurement signals by means of a direct manipulation of the output signals is also possible, as will be shown below with reference to Fig. 3 is outlined.

[0070] Instead of varying the illumination intensity by specifically controlling the light source 40, the output signals of the sensors 20-1 to 20-8 of the evaluation unit 16 can be superimposed with a specially adapted correction signal. This results in the sensors essentially "seeing" the same thing as before the correction, but generating output signals that correspond to those obtained when the illumination intensity is varied according to the embodiment described above. This variant is particularly relevant for embodiments in which the properties of the coding 12 monitored by the sensors 20-1 to 20-8 cannot be readily varied, as is possible with the optical variant described above.If the sensors are capacitive, inductive or magnetic, it is easier to apply a corresponding correction signal to the sensors instead of varying the charge or magnetic strength of the individual code sections 20-0 to 20-9.

[0071] It should also be noted that the correction unit 30 can generate the calibration signal from measurement signals received by the measurement signal generation unit 28. This is particularly relevant if the measurement signal generation unit 28 is essentially an analog-to-digital converter.

[0072] As an example of a direct influence on the measurement signals, consider the following: Fig. 4 referred to. In the configuration there, the correction unit 30 is coupled on the output side to the measurement signal generation unit 28 in order to influence or modify its operation based on the correction table and / or function.

[0073] In the exemplary embodiment, in which the measurement signal generation unit 28 is essentially an analog-to-digital converter, the correction unit 30 can modify the threshold values ​​used for digitizing the output signals according to the generated correction table and / or function, and thus directly influence the received measurement signals.

[0074] In each of the three variants, outside of the calibration run, i.e., during "normal operation" of the device, only the influenced or modified measurement signals are considered when determining the relative position. According to the invention, this determination is not influenced or modified by the correction unit and / or based on the described correction table and / or function. However, specifying this determination based on a recalibration table and / or function generated elsewhere is not categorically excluded.

[0075] The correction unit 30 can be combined with the measurement signal generation unit 28 and / or the position determination unit 24 to form a common computing unit, or can be provided separately from these.

[0076] The higher the frequency at which the correction unit 30 makes suitable adjustments to the measurement signals, the more accurate the relative position that can be determined from the adjusted measurement signals.

[0077] A correction table can comprise a collection of "support points" which, when interpolated, generate a corresponding correction function. A corresponding correction function can also include parameters to specify it. For example, a correction table could include one or more values ​​for the amplitude and one or more values ​​for the wavelength (and optionally a value for a phase shift) to specify a sine wave signal.

[0078] The correction unit can also be configured to identify different error sources and / or patterns from one or more different calibration signals and to generate separate correction values ​​and / or functions for these, which are then combined to achieve the desired effect on the measurement signals. A predefined model or one determined through machine learning can be used as the basis for this. Bezugszeichenliste

[0079] 10 Device for determining position, length, or angle 12 Coding 14 First part 16 Readout device 18 Second part 20-1 to 20-8 Sensors 22-0 to 22-9 Code sections 24 Position determination unit 28 Measurement signal generation unit 30 Correction unit 40 Light source

Claims

1. An apparatus (10), in particular for position, length or angle determination, comprising: - a first and a second part (14, 18) which are movable relative to one another; - a coding (12) which is applied to the first part (14) and which has a plurality of code sections (22-0 to 22-9) of a first kind and a second kind; - a readout apparatus (16), which is attached to the second part (18), for detecting at least one part of the coding (12), wherein the readout apparatus (16) comprises a plurality of sensors (20-1 to 20-8) which are each configured to detect the individual code sections (22-0 to 22-9) and to output a corresponding output signal; - a measurement signal generating unit (28) which is configured to generate at least one corresponding measurement signal from the respective output signals; and - at position determining unit (24) which is configured to determine a relative position between the first and the second part (14, 18), taking into account, in particular on the basis of, the at least one measurement signal, characterized in that the apparatus (10) further comprises a correction unit (30) which is configured to effect an influencing, in particular a direct or indirect influencing, of the course of the at least one measurement signal before the latter is used by the position determining unit (24) to determine the relative position; wherein the correction unit (30) is configured to effect a known predefined influencing, in particular a constant and / or unchanging influencing, of the measurement signals in a calibration run-through and to generate a calibration signal from the output signals of the sensors (20-1 to 20-8) during the calibration run-through, to compare the course of the generated calibration signal with a predefined reference course and to generate a correction table and / or a correction function from the result of the comparison; wherein the correction unit (30) is further configured to effect an influencing of the course of the at least one measurement signal beyond the calibration run-through in accordance with the generated correction table and / or correction function.

2. An apparatus (10) according to claim 1, wherein the calibration signal is the sum of two squared output signals or the square root of such a sum and the reference course is a constant.

3. An apparatus (10) according to claim 1 or 2, wherein the calibration signal corresponds to a control signal which is provided to serve as the basis for effecting the intended influencing of the at least one measurement signal beyond the calibration run-through.

4. An apparatus (10) according to any one of the preceding claims, wherein the correction unit (30) is configured to generate the correction table and / or correction function such that the hereby displayed influencing results in a corresponding calibration signal having the predefined reference course.

5. An apparatus (10) according to any one of the preceding claims, wherein the correction unit (30) is configured to generate the correction table and / or correction function such that the latter, when it is applied to the output signals, in particular directly by a superposition of said output signals or indirectly in the course of a digitization of said output signals, results in least one adapted measurement signal; wherein the position determining unit (24) is configured to use the at least one adapted measurement signal to determine the relative position beyond the calibration run-through.

6. An apparatus (10) according to any one of the preceding claims, wherein the position determining unit (24) is configured, when determining the relative position, to use a readjustment table and / or readjustment function which is different from the correction table and / or correction function generated by the correction unit.

7. An apparatus (10) according to any one of the preceding claims, wherein the two parts (14, 18) are only movable in a translatory or rotary manner relative to one another.

8. An apparatus (10) according to any one of the preceding claims, wherein the sensors (20-1 to 20-8) are optical, capacitive, inductive or magnetic sensors and the coding (12) is designed accordingly, in particular at least optically, capacitively or magnetically.

9. An apparatus (10) according to any one of the preceding claims, wherein the coding (12) is a periodic coding, in particular an incremental coding.

10. An apparatus (10) according to any one of the preceding claims, wherein the correction unit (30) is configured to generate the correction table and / or correction function such that the latter maps and / or compensates for variations in a relative positioning of the different sensors (20-1 to 20-8) and / or in the sensitivity of the different sensors (20-1 to 20-8).

11. A method for determining the relative position between two parts (14, 18) movable relative to one another by means of an apparatus (10) according to one of the preceding claims, wherein the method comprises the following steps: performing a calibration run-through in the form of a relative movement of the two parts (14, 18) relative to one another with a simultaneous output of corresponding output signals by the sensors (20-1 to 20-8); generating the calibration signal from the output signals of the sensors (20-1 to 20-8); comparing the course of the generated calibration signal with the predefined reference course; generating a correction table and / or a correction function from the result of the comparison; generating at least one measurement signal, which is influenced in accordance with the correction table and / or correction function, from the output signals beyond the calibration run-through; and determining the relative position between the two parts (14, 18) taking into account, in particular on the basis of, the at least one influenced measurement signal.

12. A method according to claim 11, wherein the calibration run-through comprises a plurality of changes in direction of a relative movement of the two parts (14, 18) relative to one another.

13. A method according to claim 11 or 12, wherein the calibration run-through takes place under constant boundary conditions, in particular under a constant temperature and / or lighting.

14. A method according to any one of the preceding claims 11 to 13, wherein both a correction function and a correction table are generated, wherein the correction table comprises a plurality of values which, when they are inserted into the correction function, specify the correction function locally.

15. A method according to any one of the preceding claims 11 to 14, wherein a correction table is generated that contains at least one correction value, in particular exactly one correction value, for each determinable relative position between the two parts (14, 18).