JPH0629723B2 - Absolute position measuring method and device - Google Patents

Description

Description: FIELD OF THE INVENTION The present invention relates to position measurement for machines with numerical control. In particular, the invention relates to measuring the absolute position of a moveable mechanical member with respect to a fixed reference position.

(Prior Art) In order to numerically control the movement of each member of a machine, it is desirable to generate a unique position measurement signal for all positions within one range of motion. This type of measurement, referred to in the text as absolute position measurement, refers to whether the position was recorded and held during periods when the movable mechanical member was not numerically controlled, such as during maintenance of the numerical control device. Regardless, it has the advantage of providing accurate information regarding the position of movable mechanical members.

(Problems to be Solved by the Invention) The absolute position measurement with respect to a numerically controlled movable mechanical member is to increase the resolution and to increase the size of the axial range in which the measurement is performed. These are constrained because they are conflicting requirements.
Absolute position measurement systems are known in all machine arts, which use a plurality of measuring devices, each operating at different resolutions and each acting to make a unique measurement over a part of the axial distance of the machine. When such a system is configured with a measuring device that incorporates a rotating element, the gear reduction ratio from the mechanical drive of the machine will be quite large relative to the measuring device used to obtain the coarsest measurement resolution. . If such a system is configured with a linear measuring device, the number of graduations required to produce a unique measuring signal over the entire machine axial distance is enormous.

Another disadvantage of the known absolute position measuring system is that it is susceptible to errors due to the measuring device itself or its associated mechanical drive mechanism. Errors resulting from such sources adversely affect the overall machine positioning accuracy.

Accordingly, one object of the present invention is to require only two measuring instruments which are operable over the entire range of motion of a movable mechanical member and which each produce a unique measuring signal which is repeated over the range of motion of the mechanical member. Not providing an absolute position measuring device and method.

Another object of the invention is to use two circular measuring instruments,
Another object of the present invention is to provide an absolute position measuring device that has a large tolerance for an error in relative position measurement performed by two measuring instruments.

(Means for Solving the Problem) In the present invention, two measuring instruments, a main measuring instrument and a sub-measuring instrument, are used. Each of these measures repeatedly in the range of motion of the movable mechanical member. The number of cycles of the signal generated when the sub-measuring instrument moves in the same range is larger than the number of cycles that periodically generate the measurement signal when the main measuring instrument moves in the entire movement range.
Then, the difference between the values of the signals from these two measuring instruments is calculated, and this difference is used to calculate how many cycles have already elapsed in the main measuring instrument before the time of measurement.

(Function) Since the maximum difference between the number of cycles of the main measuring instrument and the number of cycles of the sub-measuring instrument is 1 over the entire range of measurement, the value of the signal obtained from these measuring instruments (the value taken within one cycle) Since the difference of the signal changes depending on how many cycles the signal is from, it is possible to judge how many cycles have already passed before the signal of the main measuring instrument is actually obtained based on this difference, The absolute position can be known from the number of cycles and the signal from the main measuring instrument.

(Embodiment) In the block diagram of FIG. 1, the main measuring instrument 12 and the sub-measuring instrument 14 are respectively the interface circuits 1 thereof.
6 and 18 are shown connected. The measuring device selected by the applicant for use in industrial manipulators is of the known type, such as a resolver with electromagnetically coupled rotor and stator elements.
As shown, the moveable element or rotor of the resolver has an AC signal applied to its winding which induces an output signal in the stator winding. The rotor and stator elements are such that the periodic amplitude variation of the output based on the rotor displacement angle is unique or unique everywhere within one revolution of the rotor, or for every revolution of the rotor. The windings can be provided so that they can be repeated more than once. Each rotor of the resolver has one AC signal applied to it and produces two AC signals in the quadrature of the stator as outputs. That is, the input signal Esi to the rotor
n (wt) is the output signal Esin (wt) sin (θ) of the stator
And Esin (wt) cos (θ). However, E = strength of signal wt = instantaneous angle of input signal θ = displacement angle of rotor displacement angle θ is calculated from an inverse trigonometric function of the ratio of output signals. That is, θ is Esin (wt) sin (θ) / Esin (wt) cos
It is equal to the arctangent of (θ).

It will be appreciated that another connection of the resolver (not shown) could be substituted if the input signal is applied to the stator winding and the output signal is caused by the rotor winding. In this configuration, the stator quadrature windings have 90 ° phase difference input signals, ie, Esin (wt) and Esin (wt + 90 °).
Driven by. The displacement angle of the rotor is determined by the output signal sin (w
A corresponding phase shift in (t + θ) results, which produces a measurement signal which is detected by phase discrimination and which represents a part θ of the full cycle of the displacement angle of the rotor.

It will be appreciated that an output signal of the same shape can be obtained using an inductively coupled linear measuring device such as the Inductosyn ™ device available from Farrand.

Interface circuits 16 and 18 provide input signals to the resolver and receive output signals that are sampled and converted to appropriate levels for subsequent analog to digital converters 20 and 22. The analog-to-digital converters 20, 22 each produce an output signal representative of the instantaneous value of the output signal of the moving element of the measuring instrument. analog/
The output of the digital converter is input to the angle calculation circuits 30, 32 for calculating the displacement angle from the resolver output and for calibrating the displacement according to the selected resolution for the measured position. . It will be appreciated that the output signals of the angle calculation circuits 30, 32 represent the angle of displacement of the rotor with respect to the stator and are unique only to the range of rotation which is defined as one cycle. That is, if one revolution of the resolver rotor produces more than one repetitive cycle of displacement angle in the output signal, the output signal of the angle calculation circuits 30, 32 corresponds to one revolution cycle of the resolver rotor. A unique signal will be generated only for the parts. In the preferred variant, one A / D converter and an angle calculation circuit are used for both resolver outputs, where the conversion and calculation functions depend on the time multiplexing operation.

Any device capable of producing a unique measurement signal which is periodically repeated over the range of motion of a movable mechanical member, a conventional measuring instrument 12, 14, an interface circuit 1
6,18 Analog / digital converter 20,22
And a suitable alternative to the combination of the angle calculation circuits 30, 32 will be apparent to those skilled in the art. Thus, the measuring instrument can be a potentiometer, a rheostat, a variable transformer, or any measuring instrument that produces a unique measuring signal that is periodically repeated over the range of motion of a movable mechanical member.

As for the measuring instruments 12 and 14, the sub-measuring instrument 14 is the main measuring instrument 1.
It is arranged that its measured signal over a range of motion of the movable mechanical member than according to 2 produces a cycle number greater than one cycle. As the movable mechanical member travels from its reference position to the extreme opposite its range of motion, the difference between the two measuring instruments of the signal changing in one cycle of the measuring signal changes. For this reason, the difference in the measurement signals always gives a unique indication of how many cycles the signal of the main measuring instrument has already passed when the mechanical member moves with respect to a fixed reference value. In order to be able to identify one cycle so that the difference between the measurement signals indicates the cycle number of the measurement, the difference between the measurement signals exceeds one complete cycle over the entire range of motion. Request not to. However, while the difference in the measured angles is proportional to the number of cycles associated with the rotation given to the rotor of the measuring instrument, only subtracting the main measurement signal from the secondary measurement signal guarantees that the required difference will result. It will not be done. That is, the main measuring instrument produces a measurement signal in the A-th cycle, while the sub-measuring instrument has already generated A.
When entering the + 1st cycle, the difference of the sub-measurement signal minus the main measurement signal is negative, and it is therefore necessary to be able to handle this negative difference.

Continuing with the description with respect to FIG. 1, the difference calculation is carried out by the difference calculation circuit 24, which compensates for the values as required and produces a positive output in all cases. If the result of just the difference calculation is negative,
A value equal to one complete cycle of the main measurement signal is added to this difference. The difference signal represents a portion of the complete cycle and is directly proportional to the number of cycles of the main measurement signal corresponding to the displacement of the movable mechanical member from the fixed reference position to the current position.

In this way, the angular position (difference between two measuring signals) is proportional to the number of cycles of the main measuring signal, so that the absolute position is easily calculated by knowing the difference in the measuring signals. That is, the absolute position can be calculated by a combination of the calculated number of elapsed cycles of the main measurement signal and the value of the current measurement signal, which is calculated by the absolute position calculation circuit 26 of FIG.
The absolute position is represented with the same resolution as the main measuring device.

The absolute position value can be calculated based on the determined cycle number only by judging the cycle number of the main measurement signal against a certain standard. A slightly more complex algorithm is used in the preferred embodiment to remove the error in the measurement signal. The flow chart of FIG. 2 illustrates the procedure used in the preferred embodiment for absolute position measurement. This procedure calculates the number of complete cycles of the main measurement signal up to the nearest cycle corresponding to the distance from the reference position to the current position to synthesize the absolute position value.

In the flowchart of FIG. 2, the procedure is step 4
Starting with 0, the values of the main measurement signal X and the sub-measurement signal Y are read, each representing a part of a complete cycle. The measurement signals X and Y are to represent a value indicating the required resolution, for example, one thousandth of a length of about 24.5 mm (1 inch), or one thousandth of a rotation angle of one degree. Scaled to fit. In step 42, the difference value is calculated by subtracting the value of the primary measurement signal from the value of the secondary measurement signal. At decision step 44, the result of processing step 42 is examined to determine if a negative value has occurred. If not, execution of the procedure will continue in process step 48. However, if a negative value is detected, execution of the procedure will proceed to process step 46.
, Where a value equal to one complete cycle of the main measurement signal is added. Either the value obtained in step 46 or the value obtained in step 42 for positive cases is then used by step 48 to calculate the integer number of complete cycles of the measured signal of the main resolver.

The general number of cycles of the main measurement signal can be expressed as the sum of full cycle I and partial cycles. Only partial cycles are represented by the main measurement signal X. The secondary measuring instrument rotor shall be associated with the primary measuring instrument rotor by a relative ratio (M + 1) / M, where M is equal to the number of cycles of the primary measuring signal which is completely divisible within the range of motion of the movable mechanical member. , And assuming that each revolution of the rotors of both instruments results in one cycle of each measuring signal, the cycle of the sub-measuring instrument is equal to (I + X + I / M + X / M). Here again, it is the same that only part of the complete cycle is represented by the sub-measurement signal Y. When the measurement signal Y of the sub-measuring device is greater than or equal to the measurement signal X of the main measuring device, the last three terms (X + I / M + X /
The sum of M) is equal to the magnitude of the sub-measurement signal. Sub measurement signal Y
Is smaller than the main measurement signal X, the sum of the last three terms is one (cycle) larger than the magnitude of the sub measurement signal Y. Therefore, an integral number of cycles I of the main measurement signal can be added using known quantities from the equation below. I = M (DIFF) -X where DIFF is the difference value (Y-X) adjusted to produce a positive result as previously described.

In processing step 48, the difference signal is used to represent an integral number of cycle signals I representing an integral number of cycles of the main resolver.
Cause The difference value DIFF is multiplied by the number M of cycles of the main resolver over the measuring range and a part of one cycle represented by the measured value of the main measurement signal X is subtracted from the product result. The resulting difference is divided by the scale factor S, which corresponds to the number of units that represent the resolution within one cycle of the main resolver measurement signal. The relative error in the measured signal appears as the remainder R, which is the remainder of the integer obtained as the quotient.
In process step 49, the magnitude of the residual R is measured. If the residual R is greater than or equal to half the cycle of the main measurement signal, the integer value I is incremented by 1 in process step 50. If the residual R is less than half of one cycle in the main measurement signal, there is no change to the integer value I. In processing step 52, the product of the integer value I resulting from processing steps 48 to 50 and the scale factor S corresponding to the number of units of resolution described above contained in one cycle of the main measurement signal is: In process step 40 it is added to the value of the main measurement signal X read.

The action of the processing steps 48, 50, 52 is to eliminate the relative error of the values read in the action step 40. The numerical rounding action of steps 49, 50 yields an integer by effectively adding or subtracting a portion of the cycle of the main measurement signal that is less than or equal to half a cycle. This integer value I is corrected if necessary,
The partial cycles thus removed are replaced by the actual measured values of the master measuring device. Thus, this procedure is valid for differential errors in the measurement of the primary and secondary measuring instruments that are no greater than one cycle of the primary resolver measurement signal and the corresponding part of the rotation.

When used in a control unit for an industrial manipulator, the procedure of the flowchart of FIG. 2 is converted into a program for a microprocessor capable of executing integer arithmetic. In this embodiment, the difference calculation circuit 24 and the absolute position calculation circuit 26 in the block diagram of FIG.
It is configured by using a microprocessor and a program therefor in order to achieve the procedure of the flowchart in the figure. Furthermore, the temporary storage of the values used in this procedure is done by a memory location in random access memory.

In FIG. 3, the drive mechanism is shown along with the mechanical construction of the primary and secondary measuring instruments. The main resolver 60 and the sub-resolver 62 are geared to a common drive shaft via their respective reduction gears 64, 66. The drive gear 68 can be driven by a shaft that is driven directly by a motor that drives the movable mechanical member, or by the movement of the movable mechanical member by a clutch and pinion or other suitable device.
In either case, if the primary resolver 60 and the secondary resolver 62 are the same, then a comparison of the gear ratio of the primary resolver gear 66 to the drive gear 68 and the secondary resolver gear 64 to the drive gear 68 is made. The small difference causes the required difference in the number of cycles of the secondary measurement signal compared to the primary measurement signal over the range of motion of the movable mechanical member. Specifically, if the number of cycles of the main measurement signal that is completely divisible over the range of motion of the movable mechanical member is M, the required relative gear reduction ratio of the rotor of the sub-resolver to the rotor of the main resolver is (M + 1) / M. Becomes
As mentioned above, this simplified gear train is required in known systems to obtain the required resolution via one measuring instrument and to cover the entire machine axis by the second or other measuring instrument. It offers significant advantages over large gear reductions. From the above description, a unique measurement signal is derived from the amount of difference measured by the primary and secondary measuring instruments, rather than by dedicating a single measuring instrument to cover the selected resolution and axis range. It will be easy to understand.

When used in the manipulator described in this text, an absolute position measuring device is used to measure the position of the rotational axis of the manipulator. The measuring device is driven by gear reduction directly from a motor that produces movement of a moveable mechanical member. Nevertheless, by driving a measuring device with rotating elements such as the resolvers described herein, or by using a linear measuring device with two scales and two read heads, the same device and method described above is linear. It is also suitable for application to mechanical members having a general axis of motion.

(Effect of the Invention) According to the present invention, it is not necessary to set a large reduction ratio between the main and auxiliary measuring instruments, and therefore, the mechanism for providing the reduction ratio between them can be downsized, which reduces the inertial force. It is possible, and because there is no need to significantly change the resolution between them, the error in the measuring instrument which has a much coarser resolution than the other as in the conventional one leads to an overall significant error. there is no problem.

[Brief description of drawings]

FIG. 1 is a block diagram showing an apparatus for absolute position measurement according to the present invention, FIG. 2 is a flow chart showing the method used to determine the absolute position from two periodic measurement signals, and FIG. FIG. 3 is a schematic diagram showing two measuring instruments with rotating elements and associated drive mechanisms. 12 ... Main measuring instrument, 14 ... Sub measuring instrument, 16 ... Interface circuit, 18 ... Interface circuit,
20, 22 ... Analog / digital converter, 2
4 ... Difference calculation circuit, 26 ... Absolute position calculation circuit, 3
0, 32 ... Angle calculation circuit, 60 ... Main resolver, 6
2 ... Secondary resolver, 64, 66 ... Reduction gear, 68 ...
Drive gear.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (56) Reference JP-A-59-88612 (JP, A) JP-A-57-171207 (JP, A) JP-A-54-99528 (JP, A)

Claims (6)

[Claims] 1. For all positions within the range of motion of said mechanical member, which are movable by means of measurements of two measuring instruments each comprising a stator and a rotor which move relative to each other in response to the movement of the mechanical member. A method for producing an absolute position signal having a unique value from a reference position, comprising: (a) producing a main measurement signal representative of the relative position of the stator and rotor of the first measuring instrument, the main measurement The signal is a step which is cyclically repeated over the range of motion of the mechanical member and is unique within one cycle, (b) a sub-measurement signal representative of the relative position of the stator and rotor of the second measuring instrument. The sub-measurement signal is cyclically repeated over the range of motion of the mechanical member and is unique within one cycle, and the number of cycles of the sub-measurement signal is One cycle or less greater than the number of cycles of the main measurement signal over the range of motion of the mechanical member, (c) producing a difference signal representing the difference between the value of the main measurement signal and the value of the sub-measurement signal, (d) Calculating, from the difference signal, an integer representing the number of complete cycles in which the main measurement signal cyclically changes during movement of the mechanical member from the reference position; and (e) the integer value and the main measurement. And a signal to generate an absolute position signal. 2. The method according to claim 1, wherein the step of producing the difference signal comprises the steps of (a) subtracting the main measurement signal from the sub-measurement signal, and (b) examining the result of the subtraction. Detecting whether the value is negative,
And (c) in response to detecting a negative value, adding a value equal to one complete cycle of the main measurement signal to the result of said subtraction. 3. A device for measuring an absolute position of a movable mechanical member with respect to a reference position, comprising: (a) a stator and a rotor which relatively move in accordance with the motion of the mechanical member, and a relative position of the stator and the rotor. A main measuring instrument for producing a representative main measuring signal, said main measuring signal repeating periodically over the range of motion of the mechanical member and being unique within one cycle; (b) ) A sub-measuring instrument comprising a stator and a rotor which move relative to each other in response to the movement of the mechanical member, and which produces a sub-measurement signal representative of the relative position of the stator and the rotor, the sub-measurement signal being cyclic over the range of motion of the mechanical member Repeatable and unique within one cycle, and the number of cycles of the secondary measurement signal is greater than the number of cycles of the primary measurement signal over the range of motion of the mechanical member. (C) means for producing a difference signal in response to the main measurement signal and the sub measurement signal, the difference signal representing the difference between the value of the main measurement signal and the value of the sub measurement signal; ) Means for calculating from the difference signal an integer representing the number of complete cycles in which the main measurement signal cyclically changes while the mechanical member moves from the reference position, and (e) the integer value and the main Means for producing an absolute position signal representing a unique value indicative of the absolute position of the mechanical member by means of the measurement signal. 4. The apparatus according to claim 3, wherein the main measuring instrument is (a) an angle measuring means provided so as to rotate the stator and the rotor relatively, and (b). Apparatus comprising drive means for providing said relative rotation when movement is imparted to said movable mechanical member. 5. The apparatus according to claim 4, wherein the sub-measuring instrument is (a) an angle measuring unit provided so as to relatively rotate the stator and the rotor, and (b). Apparatus comprising drive means for providing said relative rotation when movement is imparted to said movable mechanical member. 6. The apparatus according to claim 3, wherein the means for producing the difference signal is (a) means for subtracting the main measurement signal from the sub-measurement signal, and (b) checking the result of the subtraction. An apparatus comprising means for detecting whether a value is negative, and (c) means for responding to the detection of a negative value and adding to that value a value equal to one cycle of the main measurement signal.