Method for operating an absolute measuring position detection system having a single-track magnetic code object

Abstract

In a method for operating an absolute measuring position detection system having a sensor arrangement (100) and a single-track magnetic code object (105) with non-repeating code regions, wherein the sensor arrangement (100) is formed by a substantially linear arrangement of a plurality of magnetic field sensors (110), it is provided in particular that the relative position of the sensor arrangement (100) with respect to the respective code object (105) is determined by searching for a partial pattern which is most similar to a currently sensor-detected partial pattern on the basis of available reference data containing magnetic curve progressions or magnetic patterns of magnetic field vector components detected by sensors for the entire code object (105) depending on the position on the code object (105).

Claims

1. Method for operating an absolute measuring position detection system having a sensor arrangement (100) and a single-track magnetic code object (105) with non-repeating code areas, wherein the sensor arrangement (100) is formed by a substantially linear arrangement of a plurality of magnetic field sensors (110), the method comprising: determining a relative position of the sensor arrangement (100) with respect to the respective code object (105) by searching for a partial pattern (710, 720) of a detected magnetic field signal of a present complete measurement curve which is most similar to a predetermined pattern previously stored in a reference map based on magnetic curve progressions or magnetic patterns of magnetic field vector components (500, 505) detected by sensors for the entire code object (105) depending on the position on the code object (105).

2. Method according to claim 1, searching for a position of the sensor arrangement (100) on the code object (105) at which a deviation between the previously stored partial pattern and the currently sensor-detected partial pattern forms a minimum and comparing the respective partial patterns.

3. Method according to claim 2, characterised in that the deviation is formed as a square deviation between the two partial patterns (705-720).

4. Method according to claim 2, characterised in that the deviation is formed as a sum of absolute values of a difference between corresponding values of two partial patterns.

5. Method according to claim 2, characterised in that the deviation is determined by means of an artificial neural network as a square or absolute value of a distance between the detection positions of two partial patterns estimated by the artificial neural network.

6. Method according to claim 2, characterised in that finding a minimum of the deviations between the partial patterns (705-720) to be compared for the entire code object (105) occurs on a basis of a (dis)similarity curve (725).

7. Method according to claim 1, characterised in that, during operation of the position detection system, partial patterns serving as reference data for the code object (105) are automatically learned, which, taken together, describe a course of a magnetic field along the single-track code object.

8. Method according to claim 1, characterised in that a self-diagnosis of the position detection system is carried out on a basis of determined quality values of a similarity between two partial patterns to be compared.

9. Method according to claim 1, characterised in that learned reference data are relearned in an event of any changes to the code object (105) during operation of the position detection system.

10. Method according to claim 1, characterised in that learned reference data comprises values of at least two sensor-detected magnetic field vector components (500, 505) and respective phase information (600).

11. Method according to claim 1, characterised in that the reference data are stored in a reference table or on a reference map.

12. Method according to claim 1, characterised in that a corresponding bit pattern is generated from at least two field vector components (500, 505) detected by sensors and phase-shifted with respect to each other, taking into account a phase progression (600).

13. Method according to claim 12, characterised in that the at least two phase-shifted field vector components (605, 610) detected by sensors are brought into as good a phase-related match as possible by means of a linear integral transformation in order to generate a corresponding bit pattern therefrom.

14. Absolute measuring position detection system having a sensor arrangement (100) and a single-track, magnetic code object (105) with non-repeating code regions, wherein the sensor arrangement (100) is formed by a substantially linear arrangement of a plurality of magnetic field sensors (110), characterised in that the position detection system is operable according to the method of claim 1.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1 shows a position detection system concerned here in a schematic, isometric depiction;

(2) FIGS. 2a, b schematically show two possible designs of a magnetic code tape or strip concerned here;

(3) FIG. 3 shows typical changes in the magnetic field vector when moving along a code tape shown in FIGS. 2a and 2b by means of a sensor arrangement of a position detection system concerned here, on the basis of which a phase evaluation is performed;

(4) FIGS. 4a, b show typical measurement curves obtained with a sensor arrangement of a position detection system concerned here with phase transitions shown in FIG. 3 as well as corresponding position data (4a) and bit patterns (4b) detected on these measurement curves;

(5) FIGS. 5a-c show an exemplary embodiment of the learning method according to the invention on the basis of typical magnetic sensor data measured on a code object concerned here, in this case in the form of a code tape, depending on the position of the sensor arrangement along the code tape;

(6) FIGS. 6a-e show the creation of a first map from magnetic field components detected by sensors and a second map with phase angles obtained from the detected sensor data;

(7) FIGS. 7a-d show an exemplary embodiment of the procedure according to the invention for finding a partial pattern stored on a mentioned map based on a similarity comparison;

(8) FIGS. 8a-g show exemplary measurement curves illustrating the immunity or robustness of the method according to the invention to an external magnetic field present near the code tape.

DETAILED DESCRIPTION OF THE EXEMPLARY EMBODIMENTS

(9) The sensor arrangement (or sensor head) 100 shown in FIG. 1 and the magnetic target object shown there, in the present case a stretched magnetic code tape 105, together form a linear, absolute measuring position detection system.

(10) The magnetic code tape 105 has a plurality of magnetic poles, with either the pole direction upwards 107 or the pole direction downwards 108. The linear arrangement of these different poles in the x-direction represents the encoding of the magnetic code tape 105.

(11) The sensor arrangement or the sensor head 100 has a plurality of, in the present exemplary embodiment, eighteen (18) magnetic field sensor elements 110, which are irregularly spaced apart in the x-direction as indicated by the arrows 125. The sensor head 100 also includes a measurement unit and digital signal processing unit (DSP unit) 115, and a digital communication interface 120.

(12) Additionally, the typical spatial arrangement of the axes of a coordinate system 130 of the sensor arrangement 100 with respect to the magnetic code tape 105 provided in the present exemplary embodiment is marked.

(13) The measurement/DSP unit 115 arranged on the sensor arrangement detects and processes the raw signals from the magnetic field sensor elements 110 and communicates with external devices (not shown here) via a digital communication interface 120, namely for transmitting sensor data, parameter data and diagnostic data. The magnetic field sensor elements 110 are designed to be magnetically sensitive, in particular in two axes, in order to be able to perform a phase evaluation of the measurement signals as mentioned and described in more detail below.

(14) The magnetic field sensor elements 110 have in particular the following technical properties or features: they are designed to be substantially equal; they are arranged in the direction of movement of the sensor elements along the mimetic code object; depending on the spatial configuration of the code object or the movement trajectory of the respective target object to be detected, they are arranged either along a straight line or along a curved trajectory; they are arranged either with a substantially constant distance between the individual sensor elements thereof or with different or varying distances between the individual sensor elements; they each have at least two sensitive axes for detecting the magnetic field generated by the magnetic target object. The sensitive axes thereby span a plane substantially coincident with both the arrangement of the magnetic field sensor elements and a line connecting the arrangement of the magnetic field sensor elements and the centre of the respective magnetic: target object. Said centre is either the centreline of a magnetic code tape or the centre of a discrete magnet, depending on the type of target object.

(15) However, the sensor arrangement proposed herein can also be applied to sensor elements which detect magnetically only on a single axis. Also, the sensor arrangement can (optionally) still have a third, sensitive axis oriented substantially perpendicularly to the first two axes.

(16) In particular, the signal processing unit 115 has the following technical properties or features: it has a programmable component, e.g. a microcontroller, an FPGA or similar, or a combination of such components, as well as an operational memory, e.g. a RAM, which is as fast as possible, and a rewritable, non-volatile memory, e.g. a FLASH, FRAM or similar; it cyclically reads out the signal from the magnetic field sensor elements; it converts the signals detected by sensors into a series of regular sensor signals in a self-regulating manner, so to speak, by eliminating, small sensory differences between the magnetic field sensor elements of the sensor arrangement by means of a background correction and by means of a gain compensation, e.g. on the basis of a spatial rotation of the rectified signals with respect: to the coordinate system of the sensor arrangement; it determines, based on the detected sensor signals, the relative position of the sensor arrangement or sensor head with respect to the magnetic target object; it provides diagnostic information and tools for installation, maintenance and normal operation of the position detection system; it is capable of communicating bidirectionally with external devices via a digital interface.

(17) FIGS. 2a and 2b schematically illustrate two exemplary magnetic code tapes (or magnetic code strips) of a position detection system concerned here.

(18) The absolute code tape shown in FIG. 2a has only a single track 200 having an absolute code having a uniform bit length. The absolute code is composed of a linear arrangement of poles with pole direction (see FIG. 1) upwards 205 as well as with pole direction downwards 210. Thus, the coding shown comprises both single poles 207 surrounded by poles of different polarity and multiple poles, i.e. multiple connected poles of the same polarity 205, 210.

(19) In contrast, the code tape shown in FIG. 2b, which is also suitable for the sensor arrangement according to the invention or a corresponding position detection system, has both a uniform, incremental bit length code 215 and a relatively short code segment 220 or a corresponding code target object having a bit code also of uniform bit length in each case.

(20) According to FIGS. 2a and 2b, only a single absolute code track 200, 220 is required in each case according to the invention. This has various advantages with respect to the magnetic code strip. First, it results in lower manufacturing costs and also total operating costs compared to the prior art. In addition, it results in simpler installation as well as greater tolerances to assembly inaccuracies.

(21) FIG. 3 illustrates the magnetic field vector resulting when moving along a code tape shown in FIGS. 2a and 2b in the shown x-direction 300 by a sensor arrangement according to the invention. The points 305 included in the diagram indicate the respective position of the scanning. The lines 310 also shown there are in each case aligned in the direction of the field vectors occurring, during magnetic induction, corresponding to the possible phase values 320 indicated below the diagram. The length of the lines has been normalised to the maximum of the absolute field values present.

(22) The step-shaped line 315 drawn in the lower part of the diagram corresponds to the magnetic code resulting from the scanning. The angle of the magnetic induction vectors is measured with respect to the x-axis. In the following, this angle is referred to as the phase angle or phase of the magnetic induction vectors.

(23) At any distance of the sensor elements from the respective code tape (namely in the vertical z-direction shown in FIG. 1), wherein this distance should not be greater than the longitudinal dimension of a magnetic code bit, the magnetic induction vector rotates in a negative direction when the sensor elements move from left to right, i.e. clockwise in the present representation. Above code bit boundaries with alternating (bit) polarity, the field is substantially horizontal. At greater distances, however, certain bits are difficult or impossible to detect or measure due to the broadening of the magnetic field distribution of the individual magnetic code bits (see the bits contained in the top line).

(24) FIG. 4a illustrates typical measurement curves for the phase transitions at the code bits detected in each case. Here, the curved line 400 corresponds to the phase progression of the magnetic induction vector along the (horizontal) position on the code tape and the step-shaped line 405 corresponds to the corresponding value of detected magnetic code bits.

(25) If the sensor arrangement or the sensor head moves from left to right over the respective target object, e.g. a magnetic code tape, a dipole magnet, or the like, then the magnetic induction vector rotates in a negative direction, i.e. clockwise in the present case. The phase progression 400 now has characteristic features which correspond to the structure of the magnetic code tape. Plateau-shaped phase progression areas 410 correspond to longer, magnetically homogeneous sections. At magnetic transitions 415 between poles or pole areas of different polarity, however, the respective code bits are inverted. Therefore, it is possible to determine the respective magnetic bit sequence in the present case by means of a “reverse analysis”.

(26) In FIG. 4b, a code tape bit pattern 420 is depicted in the upper part of the diagram shown, wherein 425 and 430 represent the respective x and z components of a corresponding regular (and already equalised) signal detected by the sensors. A detected binary pattern 435 is depicted in the middle of the diagram. Here, the points 440 correspond to detected bits and the point 445 to the actual reference position of the sensor. In the lower part of the diagram, a detected bit pattern 450 and a monotonic phase 455 generated from said regular signals detected by sensors are depicted.

(27) FIGS. 5a-5c show an exemplary embodiment of the learning method according to the invention using typical magnetic sensor data measured on a code tape and depending on the position of the sensor arrangement along the code tape.

(28) According to the method, a reference table (herein referred to as a “reference map”) with magnetic curve progressions or magnetic patterns is generated on the basis of sensor-detected, magnetic field vector components, in the exemplary embodiments described below the field components B.sub.x and B.sub.z, as well as on the basis of phase angles of these vector components, during operation of the position detection system, i.e. during the movement of the sensor arrangement along the code object. Thus, new (local) magnetic partial patterns are learned by finding, on an existing map, already existing partial patterns corresponding or correlating to these partial patterns by means of a similarity check and by replacing or improving or correcting these corresponding partial patterns by the corresponding currently detected partial patterns.

(29) It should be noted here that the identification of the actual position based on dissimilarity does not depend on the number of sensitive axes of the sensor elements. This applies to both single-axis and triple-axis sensitive sensor elements.

(30) In the present exemplary embodiment, the mentioned learning process comprises the following four process steps based on a said already existing reference map:

(31) 1. Reading out the existing reference map for the code object concerned in each case (note: an individual reference map is created for each code object).

(32) 2. Finding a partial pattern in the reference map by means of a correlation calculation, wherein this partial pattern is as similar as possible to a currently measured partial pattern or a corresponding magnetic curve progression,

(33) 3. Learning new or correspondingly changed patterns or corresponding complete curve progressions by mathematical interpolation, and

(34) 4. Idealisation of the thus changed patterns or curve progressions on the basis of given, physical models fora magnetic system concerned here, and corresponding renewal of the already existing reference map.

(35) This procedure has the advantage that the position detection system can already be operated in normal mode even during the learning process, and that the system can preserve or even improve its measurement properties, even in the case of strongly changing external conditions (e.g. due to a parasitic magnetic field described below), as a result of the learning process.

(36) In FIGS. 5a-5c, field vector components B.sub.x (black dots) 500 and B.sub.z (white dots) 505 detected by sensors on a code tape are depicted in arbitrary units above the actual position in each case in the unit [mm] of the sensor arrangement along the code tape. The lines 507, 509, also drawn in, correspond to simple connecting lines between the measuring points, depicted differently for each of the two field components (B.sub.x: bold, B.sub.z: dashed).

(37) It should be noted that the number of points 500, 505 shown in FIGS. 5a-5c in the present example is 11 (eleven), corresponding to a sensor arrangement 100 shown in FIG. 1 with only eleven (11) sensor elements 110, wherein the distance between the individual sensor elements in the present exemplary embodiment is around 1.38 times greater than the bit length.

(38) Thus, FIG. 5a shows currently detected spatially limited (local) partial patterns (point distributions) or corresponding curve progressions for the two field vector components. These patterns or corresponding curve progressions are stored or saved on the reference map.

(39) In the measurement situation shown in FIG. 5b, the sensor arrangement has moved around 40 mm to the right above the code tape. This results in the shown further partial patterns or corresponding curve progressions for the two field vector components. This further reference data is also stored on the reference map in the present exemplary embodiment.

(40) In the measurement situation shown in FIG. 5c, the sensor arrangement above the code tape has moved around 130 mm to the left, starting from the last position according to FIG. 5b. This results in the shown still further partial patterns or corresponding curve progressions 510 for the two field vector components. This further data is also stored on the reference map, in addition to the data 515 already stored.

(41) This results in the progressions of the two field vector components B.sub.x and B.sub.z depicted in FIG. 5c by the solid line 520 and the dashed line 525 for the present code tape as a whole.

(42) FIGS. 6a-6e show the creation of a first map with complete progressions of the two field vector components B.sub.x and B.sub.z along the whole code tape, starting from the (local) partial pattern of the two magnetic field components 500, 505 shown in FIG. 5a. In addition, the creation of a second map with corresponding phase angles is shown from this data.

(43) From the point values 500, 505 (FIG. 6a), the phase shifts or corresponding phase angles of the measuring points in the unit ‘rad’ (FIG. 6b) between the measured two field vector components are calculated depending on the position of the sensor arrangement by respective difference formation.

(44) By moving or traversing the sensor arrangement along the entire code tape in the manner shown in FIGS. 5a-5c, the complete, first reference map (FIG. 6c) of the two magnetic field vector components is generated. Correspondingly, the second reference map for the phase course or phase shift 615 between the two field vector components along the entire code band is generated (FIG. 6d). In this depiction, the local phase course 600 according to FIG. 6b is delineated or highlighted by a window 620.

(45) From the two phase-shifted curve progressions 605, 610 for the two field vector components B.sub.x and B.sub.z shown in FIG. 6c, a corresponding bit pattern 625 is calculated or formed, taking into account the phase progression 615. The (binary) bit sequence 630 emerging in the present example is also listed above the bit pattern 625. This is because these two field components represent almost harmonic conjugates and can therefore preferably be brought into as good a phase-related match or overlap as possible by means of a linear integral transformation, e.g. the integrally known “Hilbert transformation”, so that they can thereby be evaluated together with regard to the determination of a corresponding bit pattern.

(46) It should be noted here that the joint evaluation is also possible if the sensor elements each have only one sensitive direction. Thus, in the case of sensor elements sensitive to only one field vector component, e.g. Bx, the course of the respective other field component, in this case Bz, can be at least approximately calculated or simulated from the course of the field vector component Bx by a Hilbert transformation.

(47) FIGS. 7a-7d show a method for finding a partial pattern stored on a map by means of a similarity check. Here, the position detection system determines the position of the sensor arrangement relative to the code tape by finding the part or partial pattern stored on the reference map that has the greatest similarity to a currently sensor-detected partial pattern. In the present exemplary embodiment, the similarity is calculated using the known method of the least deviation or error squares. However, other known correlation methods, such as the methods mentioned at the beginning, can also be considered.

(48) FIG. 7a shows a code tape 700, assumed here and having only one track, which has an irregular sequence of magnetic poles 108, 108 shown in FIG. 1.

(49) FIG. 7b again shows only locally detected sensor signals for currently detected sensor data 705 (black dots) and for sensor data 710 (white dots) stored on the reference map and very similar to the current sensor data 705, both values in arbitrary units, and in fact depending on the relative position in [mm] between the sensor arrangement and the code tape (see FIG. 1). The two measurement curves, again generated by simple connecting lines between the measuring points, show a clearly visible high match.

(50) In contrast, FIG. 7c shows sensor data 715 (black dots) currently detected locally at a different position on the reference map and corresponding data 720 (white dots) stored at this position on the reference map, both values again in arbitrary units. However, compared to FIG. 7b, considerable deviations between the two curves can be seen.

(51) FIG. 7b shows a dissimilarity curve 725 calculated mm the matching results according to FIGS. 7b and 7c, which can be used to determine the position on the reference map with the highest degree of matching. Thus, local values according to FIG. 7b result in relatively low values of dissimilarity, and local values according to FIG. 7c result in relatively high values of dissimilarity. In the present example, the two partial patterns shown in FIG. 7b result in an extremely low value 730 of dissimilarity due to the relatively high degree of matching.

(52) As a result, the current code tape position is assumed to be the position concerned on the reference map with the partial pattern shown in FIG. 7b as the most probable current position of the sensor arrangement with respect to the code tape.

(53) Finally, FIGS. 8a-8g show typical measurement curves to illustrate the robustness of the described method with respect to an external magnetic dipole field present only locally in the vicinity of the code tape in the tape direction of the code tape or the corresponding parasitic interference field.

(54) FIGS. 8a and 8b show characteristic curves 800, 805 of the sensor arrangement concerned here, in which positions determined using the method described (y-axis) are plotted against actual positions (x-axis) in respective units [mm]. The two diagrams also show curve progressions 810, 815 (each in arbitrary units) of the relative position deviations which occurred between the determined and the actual position data.

(55) FIG. 8a shows the situation without the aforementioned interference field and FIG. 8b shows the situation with the presence of the interference field. As can be seen by comparing the two characteristic curves 800, 805, the influence of the interference field on the characteristic curve 800 of the sensor arrangement is relatively small or negligible.

(56) FIGS. 8c-8e show phase-shifted curves 820, 825 for the two field vector components B.sub.x and B.sub.z already depicted in FIG. 6c, and indeed in FIG. 8c without a so-called parasitic interference field and in FIG. 8e in the presence of such an interference field. In this case, the curve progression 825 was relearned during the contact of the interference field in the manner described. It should be noted that the difference 830 between the two curves 820, 825 corresponds substantially to the field course 835 of the interference field shown in FIG. 8d, which was also detected by means of the sensor arrangement.

(57) In FIGS. 8f and 8g, non-similarity curves 840, 845 already shown in FIG. 7d are depicted, and indeed in FIG. 8f without the presence of the interference field and in FIG. 8g with the presence of the interference field, wherein a reference map (with the corresponding, curve progressions) created in a situation not disturbed by the parasitic interference field is used as a basis for creating the non-similarity curve 845. In the first non-similarity curve 840, a minimum 850 with the value zero of non-similarity (and thus a maximum of similarity) occurs at the position −25 mm of the code tape, whereas in the second non-similarity curve, minima 855, 860 occur at the two positions −53 mm and +55 mm of the code tape, but with the non-similarity value 0.12. Accordingly, even in the presence of the interference field, a curve progression of the non-similarity curve 845 emerges, which is thoroughly evaluable and yields an absolute position value. In the case of a closer look, the position 855 would even be selected, which is relatively close to the position 850, since the minimum value at the position 855 is slightly higher than the minimum value at the position 860.