DE10247319B3 - Determining movement or angle of rotation from indicator tracks on cylinder, takes into account angular error in phase sensor mounting and corrects it - Google Patents

Abstract

Tilt angle (phi) is determined, being the angular position of the sensors at right angles to the direction of motion of the component. Measurement error (DELTAalphai) caused in this way is determined using a factor (q), which is evaluated as follows. Departures (dTi) of the transformed, non-integer measured values of phase are calculated, relative to adjacent measured integer values of phase. From N-1 comparisons with the departures (dTi) the method determines the minimum error squared, for the factor (q). From this value and from the measured phase measurement values (ai), the measurement error (DELTAalphai) for each phase indicator track is determined.

Description

State of the art

The invention relates to a method for detecting a movement or an angle of rotation, in particular an angle of rotation on axes or shafts, according to the generic term of Main claim.

For example, to record the a steering wheel axis of a motor vehicle acting torque during the Rotation of the steering wheel very small changes in angle in both directions of rotation Steering wheel can be measured. Incremental angle encoders can be used here, an angular position based on the evaluation of optical, magnetic or otherwise as generated by the rotation and by suitable means evaluate detected impulses. To increase the measuring accuracy and especially for measuring the torque due to torsion a plurality of such incremental, measured values that occur periodically are evaluated so that Here several phase measured values occur, from which the one to be measured Size how e.g. the angle of rotation, an angle difference or the distance to one The goal is to determine.

In the case of more than two phase signals, for example, one is used to evaluate such phase measurement values DE 101 42 449 A1 described method proposed. Here the measured phase values are mathematically transformed using a linear transformation method and evaluated with a predetermined weighting. A method is thus described here which generates a highly precise, robust and unambiguous phase measured value from N ambiguous, possibly also disturbed phase measured values. The method is used, for example, in an optical angle sensor in which N parallel tracks are applied to a cylinder. Each of the N tracks (i = 1 ... N) contains n _i periods of phase information, which is represented, for example, in the optical case by n _i periods of light-dark transitions. Other sensor principles, for example magnetic or capacitive, are also possible here. The traces of the sensor can also be applied on one level instead of on a cylinder, for example in the case of a displacement sensor.

It is also known from the DE 195 06 938 A1 that the phase signals can be evaluated by the simple or multiple use of a classic or modified vernier method. Furthermore, to determine an angle difference from the DE 101 42 448 A1 It is also known that the phase measurement values are weighted and the integer and non-integer part is then determined.

Furthermore, it is also from the DE 43 01 971 A1 and the DE 100 18 298 A1 It is known that in the case of path detection with sensors, the angular position of the sensors perpendicular to the direction of movement of the component to be detected and any measurement error resulting therefrom are taken into account in the evaluation.

The generic method can here, for example with a corresponding sensor arrangement, as mentioned on the Steering shaft of a vehicle can be used as a so-called torque angle sensor (TAS), this phase information is recorded optically or magnetically and converted into corresponding, here electrical phase signals become. In these and other comparable applications with a corresponding one Sensor arrangement for measurement errors can be detected, for example, by the acquisition of the phase signals the sensor head is skewed relative to the sensor tracks, where the angle of the skew as the so-called tilt angle is the measurement result distort can.

With the generic method mentioned at the beginning for detecting the movement or the angle of rotation on moving mechanical Components can Phase measured values are evaluated by scanning several one above the other arranged phase encoder tracks also perpendicular to the direction of movement of the component one above the other built-in sensors assigned to the tracks become. The phase measurements can by means of a linear transformation of a vector from N phase measurement values can be mathematically transformed into an N-1 dimensional space.

With the invention can be advantageous Way the tilt angle is determined by that by means of a factor a calculation of the tilt angle caused phase errors and a subsequent correction becomes. The following procedural steps are proposed for this purpose, the based on the embodiment even closer explained become.

Deviations of the transformed non-integer measured values from the neighboring integer measured values are identified with one from the one mentioned above DE 101 42 449 A1 Known methods quantized and from N-1 equations with the deviations a value for the factor is determined using the method of least squares. The value of the factor becomes the measurement error for each phase encoder track is determined and subtracted from the measured phase measured values.

It is particularly advantageous here if the determination of the tilt angle by measurements at different Angular positions of the sensor is carried out on the component to be measured.

The invention is particularly related to this suitable the phase signals of an optical torque angle sensor (TAS) in an optimal way and the adjustment of the tilt angle in sensors of the previously described Way to perform. The adjustment can be made once during production or during Installation as well as online in the operation of the sensor. In the Online compensation can e.g. also compensate tilt angle, caused by temperature or age-dependent deformation of the sensor geometries be effected.

The adjustment is carried out according to the invention with a separate computer arrangement, for example in a control unit, so that no mechanical intervention in the sensor is necessary and for the tilt adjustment also no otherwise required reference angle encoder is required. The adjustment parameters can e.g. can also be stored in a flash memory of the control unit.

An embodiment of a sensor arrangement to carry out of the method according to the invention is explained using the drawing. Show it:

1 a basic arrangement of phase encoder tracks on a shaft, for example in the case of an optical torque angle sensor (TAS), and

2 the arrangement of three superimposed phase encoder tracks in detail with the installation position of a sensor or the angular position of a cylinder, for example.

description of the embodiment

In 1 is an axis in a schematic view 1 Shown as a rotating component, the angle of rotation of which can be determined from phase sensor tracks with a sensor arrangement which in principle derives from the state of the art mentioned in the introduction to the description DE 101 42 448 A1 is known.

The method according to the invention is described on the basis of a multidimensional phase evaluation of the signals from in 2 optical signaling devices shown 2 to 4 as phase encoder tracks. In the section of a development of 3 tracks of a cylinder shown here, n ₁ , n ₂ and n ₃ periods per revolution are run through in the individual tracks. The target orientation of a read head (not shown here) for the tracks 2 to 4 is with the line 5 drawn in dashed lines; the actual installation position (actual orientation) is with the solid line 6 shown. The tilt angle Φ denotes the angle between the lines 5 and 6 ,

The read head measures in every track 2 to 4 the phase angles α _{i of} the light / dark transitions. If the cylinder in a track rotates by one period, ie a black and white pair, the corresponding phase angle α _i runs through a range of 2 ?. From the periodic and thus ambiguous phase angles of the N phase sensor tracks (here tracks 2 to 4 ) with the help of the prior art mentioned in the introduction DE 101 42 449 A1 described algorithm, the unique angular position of a cylindrical phase encoder or sensor or the respective position in the case of a linear position encoder can be determined.

For example, if the reading head has a tilt angle Φ, with point D as the known pivot point, then the phase angle α _i applies to the measurement errors Δα _i caused thereby: Δα i = Q * d i * n i (1)

Here, d _i denote the respective distance of the i-th track from the pivot point D, ie track 2 with the distance d _1. etc. The factor q is now to be determined from measured values; it depends on the tilt angle Φ and the sensor geometry. For example, in the case of a cylinder with radius R as the carrier of the phase encoder tracks:

The method according to the invention now aims to determine the factor q from measured values (Δα _i + α _i ). The systematic measurement errors Δα _i can then be determined from equation (1) and calculated from the measured values. After the aforementioned DE 101 42 449 A1 The first step in calculating the unique angle of the cylinder is to multiply the vector of the ambiguous angle values α _i by a matrix A and subsequently by a matrix B. These transformations transform a vector from N measured values into an (N-1) dimensional space T mapped according to the following equation.

Abbreviated M ₁ = B * A (4) M ₁ is a specially chosen (N-1) times N matrix, with M ₁ in particular:

If the measured values are assumed to be ideal, the coordinates of the transformed measured values are always an integer because of equation (5). However, the measurement errors Δα _i mean that the transformed measured values are no longer an integer. The deviations dT _{i of} the values T _i with respect to integer quantized, for example with respect to the nearest integer neighboring points, are now considered. dT i = T i - (T i ) quantized (6)

The following now applies to the deviation:

Hereby there are N-1 equations for the unknown q and an optimal value for the factor q can be given according to the method of least squares. The sum of squares with respect to q must now be minimized. e 2 1 (q) + e 2 2 (q) + ... + e 2 N-1 (q) (8)

The individual errors e _{i are} calculated as follows:

The constants g _i occurring here are calculated as:

A solution for the factor q now results from the relationship:

From the factor q calculated in accordance with equation (11), the measurement errors Δα _i are then calculated _in accordance with equation (1) and subtracted from the measured phase angles. For a more precise determination of the measurement error caused by the tilt angle Φ, it is advantageous if several measurement values, for example 100, are recorded at different angular positions, for example of the cylinder with the phase encoder tracks or the position of the linear sensor, and then in each case to determine the values for dT _i . The equations described above are then to be applied to the mean values of the values dT _i .

Claims (2)

Method for detecting the movement or the angle of rotation on moving mechanical components, in which - phase measurement values are evaluated, which are generated by scanning several phase transmitter tracks using sensors which are arranged one above the other perpendicular to the direction of movement of the component, and in which - the phase measurement values by means of a linear transformation of a Vector from N phase measurement values (α _i ) are transformed into an N-1 dimensional space, characterized in that - a determination of the tilt angle (Φ) as the angular position of the sensors perpendicular to the direction of movement of the component and the measurement errors (Δα _i ) caused thereby by means of a Factor (q) is carried out, which is determined using the following method steps: - Deviations (dT _i ) of the transformed non-integer phase measured values compared to the adjacent integer phase measured values are calculated, - N-1 equations with the deviations (dT _i ) are converted into Using the method of least squares, a value for the factor (q) is determined and - from the value of the factor (q) the measurement errors (Δα _i ) for each phase encoder track are determined and subtracted from the measured phase measurement values (α _i ). A method according to claim 1, characterized in that - the determination of the tilt angle (Φ) by measurements at different angular positions of the sensors on the component to be measured performed becomes.