JPS6110715A - Absolute-position detecting system - Google Patents

Abstract

PURPOSE:To correct the interpolation error of a resolver, by obtaining the correcting value at an arbitrary point from a value read out of the resolver and the interpolation error in the resolver, and obtaining the corrected absolute value at the arbitrary point based on said value and a specified reference value. CONSTITUTION:Paper tape 101, in which NC command data is stored, is connected to an NC device 102. A pulse distributing device 103 is operated by the command from the NC device 102. The signal from the device 103 is supplied to a servomotor 105 through an acceleration and deceleration circuit 104, an error computing and memory part 107, a D/A converter 108 and a speed control circuit 109. Thus a resolver 106 is driven. The resolver 106 is rotated and its interplation error is measured. The characteristic curve is obtained and stored in the device 102. At first, the reference value for detecting the absolute-position, which is corrected based on the value read out of the resolver 106 and the interpolation error of the resolver 106, is obtained. The corrected value at an arbitrary point is obtained based on the value, which is read out of the resolver 106 at the arbitrary point, and the interpolation error. Based on said value and the reference value, the absolute-position of the arbitrary point is obtained.

Description

Detailed Description of the Invention (Industrial Application Field) The present invention relates to an absolute position detection method, and in particular, in order to minimize the influence of interpolation errors of the resolver, a characteristic curve to the resolver is measured, The present invention relates to an absolute position detection method that can obtain accurate absolute positions based on the measurement results.

(Prior art and its problems) As shown in Fig. 6, the resolver has a pair of stator windings SA.
, SR and rotor winding RA.
is connected to the rotating shaft of the servo motor.

The rotor winding RA is configured to be rotated by the rotation of the servo motor, and the resolver excitation circuit 1 applies a sine wave (sinωot) to one stator winding SA and a cosine wave to the other stator winding SB. When (cosωot) is applied, a sine wave of 5 inches (ωot+θ) delayed by the mechanical angle θ of the rotor RA is output from the rotor winding RA. This sine wave output is sent to the stator winding SA by the position detection circuit 2.
The phase difference θ between the sine wave sinωOt and the applied sine wave sinωOt is detected, and the pulses corresponding to the phase difference are counted and output as a position signal θ. However, such a multi-pole resolver requires a pair of stator windings SA and SB shown in FIG. 6 for the number of poles, so it is necessary to arrange a large number of stator windings on the rotor winding RA of the resolver. , high positional accuracy is required. Resolvers themselves are generally miniaturized, and it is difficult to arrange the stator windings with high precision, and placement errors, that is, interpolation errors, are unavoidable, which causes the phase of the output wave to shift by the error amount. It turns out. This situation is shown, for example, in FIG. In other words, the resolver error changes depending on the phase difference. Since such interpolation errors of the resolver exist, there is a drawback that accurate position detection is difficult.

(Objective of the Invention) An object of the present invention is to provide an absolute position detection method capable of correcting the interpolation error of the resolver and performing accurate position detection.

(Summary of the Invention) The present invention comprises a measuring means for rotating a resolver to obtain characteristic curve data of interpolation error of the resolver, a data storage means for storing data regarding the characteristic curve in a bubble memory,
Value r read from the resolver when setting the reference point for absolute position detection. ' and the interpolation error of the resolver obtained from the measuring means (0)
. If you want to find the absolute position at any point,
A value r' read by the resolver at that time, a corrected value r at an arbitrary point obtained based on the interpolation error ε' of the resolver obtained from the measuring means, and the reference value r. A means for determining a corrected absolute position at an arbitrary point from the above is provided to obtain an accurate absolute position of the servo control system.

(Example) An example of the present invention will be described in detail below with reference to the drawings.

FIG. 1 is an overall block diagram of a servo control system to which the absolute position detection method according to the present invention is applied.
A paper tape with C command data perforated, positioning information for processing, and M.

Accumulated NC command data such as S and T function information, 1
Reference numeral 02 denotes an NC device, which performs tape league (described later) from a paper tape 101 to read NC data, and deciphers the read NC data. For example, if it is an M, S, or T function command, etc. If it is a movement command Zc, it is output to the subsequent pulse distributor.

The NC device 102 includes a processor 102a that executes arithmetic processing according to a control program, a program memory 102b that stores a predetermined control program, a data memory 102c that stores data, and an operation panel 102 for operation.
d, tape reader/puncher 102e, teaching operation panel 102f, input/output port 102g, bubble memory 102
h, a current position counter 102i, a position detection signal input port 102j, a display device 102k, an address, and a data bus 102. Reference numeral 1034 denotes a pulse distributor, which executes a known pulse distribution calculation based on the movement command Zc to generate a distribution pulse Ps of a frequency corresponding to the command speed. 104 is a known acceleration/deceleration circuit that linearly accelerates the pulse speed of the distributed pulse train Ps when the pulse train is generated, and decelerates the pulse speed linearly at the end of the pulse train to generate the pulse train Pi; 105 drives the operating axis; It is a servo motor, and the movable part of an industrial robot is provided as a load, for example. Reference numeral 106 indicates a resolver as a position detection means, and reference numeral 107 indicates an error calculation storage unit, which is composed of, for example, a reversible counter, and stores the number of input pulses Pi generated from the acceleration/deceleration circuit 104 and the error Er of the pulses obtained by feedback. do. Incidentally, this error calculation storage section may be composed of the illustrated calculation circuit 107a and an error register 107b that stores the error Er, and 108 is the error register 1.
A digital-to-analog (DA) converter 109 generates an analog voltage proportional to the contents of 07b, and 109 is a speed control circuit. Note that the error calculation storage unit 107 and the DA converter 108
constitute the motor position control circuit.

In particular, in the present invention, a characteristic curve of interpolation error of the resolver is obtained by rotating the resolver lO6, data regarding the characteristic curve is stored in the bubble memory 102h, and a reference point for absolute position detection is set, that is, an origin value. When setting , the value ro ' (third
A reference value ro for absolute position detection is obtained by correcting the interpolation error of the resolver from the data obtained by the measurement (see figure) and the data obtained by the measurement.

Furthermore, when determining the absolute position at an arbitrary point, the arbitrary point obtained using the value r' read by the resolver at the arbitrary point and the interpolation error ε' of the resolver obtained from the measurement data. The corrected value r and the reference value r
. Find the corrected absolute position at any point from .

This point will be explained in more detail based on the drawings.

First, the resolver is set to interpolation error measurement mode by operating the operation panel, and the interpolation error is measured by rotating the resolver 106. That is, by rotating the resolver 106, the rotor winding RA shown in FIG. An output wave of 5 inches (ωOt 10+θ') (where θ' is an interpolation error) and a sine wave sinωot applied to the stator winding SA are input to the position detection signal input capo 102j shown in FIG.

Figure 2 is. FIG. 2 is a circuit configuration diagram of an embodiment of such a position detection signal input board.

In the figure, 20 and 21 are zero slice circuits each composed of a comparator, and each output wave 5i of the rotor winding RA.
n(ωot+θ+θ'), the sine wave sinωOt applied to the stator winding SA is sliced to zero volts and converted into a rectangular wave T 1 + T 2 (see FIG. 4). 2
2.23 is a rising edge detection circuit, which detects each rectangular wave T, ,T
2 and outputs rising pulses T, , T4, and is composed of a NOT gate 22a, a pair of shift registers 22b, 22c, and an AND gate 22d. 24 is a flip-flop, rising pulse T3
It is set by the rising pulse T4 and outputs the gate signal T5 corresponding to the phase difference. 25 is a gate circuit, which outputs a clock pulse C during the output period of the gate signal T5.
Output LK. 26 is a counter, and gate circuit 2
5 and is reset by the rising pulse T4. The output from counter 26 is output to address e-data bus 102 (see FIG. 1). Furthermore, the rising pulse T5 is similarly output to the address/data bus 1021.

Therefore, the output from the counter 26 is stored in the bubble memory 102h via the address/data bus 102.

Next, correction of the interpolation error θ' output from the resolver 106 will be explained.

The relationship between the phase difference of the resolver 106 and the interpolation error ε, that is, the correction value θ' is as shown in FIG. 3, and the error of the sampled phase difference of the resolver is measured and stored in the bubble memory 102h. has been done.

Next, connected to the servo motor 105, for example,
The reference point for absolute position detection of an articulated robot, that is, the origin setting, is performed from the operation panel.

FIG. 5(a) is a diagram showing the origin value setting state of the articulated robot. In FIG. 6, 1 is the W-axis unit, 2 is the W-axis arm, 2a is the bearing, 3a is the bearing, and 4 is the wrist. It is a mechanism.

The origin setting of this articulated robot is performed by driving the servo motor 105 so that the W-axis arm 2 is vertical.

Therefore, the read value at this time when resol/< l O6 is read at this origin setting is r. ′ (see FIG. 3), and on the other hand, the interpolation error ε of the resolver corresponding to the origin setting, which is stored in the bubble memory mentioned above. is output from the bubble memory 102h to the address/data bus 10211, and the read value r is output. ′ to interpolation error ε. is subtracted by the CPU 102a to obtain the corrected reference value r for absolute position detection of the origin setting position. seek.

Next, as shown in FIG. 5(b), in order to find the absolute position of the arm of the articulated robot at an arbitrary point, a bubble is generated from the read value r' from the resolver 106 at the arbitrary point. A corrected value r is obtained by subtracting the interpolation error ε' of the resolver 106 corresponding to the arbitrary point stored in the memory 102h. Therefore, the obtained correction value r and the reference value r described above. Find the exact absolute position at any point from .

In addition, although servo control of only the l-axis is explained in FIG. 1, it goes without saying that it can also be applied to control of multiple axes. The movable parts of various machines equipped with NC devices, such as machines, can be connected.

(Effects of the Invention) According to the present invention, even though there is an interpolation error inherent in the resolver, it is possible to reliably correct the interpolation error, so the position of the robot, machine tool, etc. Detection can be performed accurately, and position control can therefore be performed accurately. Therefore, reliability in controlling robots, machine tools, etc. can be improved. Further, the circuit configuration for this purpose requires only a simple circuit to be installed in a part of the conventional NC device, and the increase in cost for expanding the functions is small.

[Brief explanation of the drawing]

FIG. 1 is an overall block diagram of a servo control system to which the absolute position detection method according to the present invention is applied, FIG. 2 is a circuit configuration diagram of an embodiment of the absolute position detection signal input port in FIG. 1, and FIG. Characteristic curve diagram explaining interpolation error of resolver, 4th
The figure is a waveform diagram of each part of the embodiment configuration in Figure 2, and Figure 5 (a) is,
A side view showing the reference point position setting state of the articulated robot,
FIG. 5(b) is a side view showing the operating state of a similar multi-jointed robot at an arbitrary point, and FIG. 6 is a diagram illustrating the principle of a conventional multipolar resolver. 1.02 h... Bubble memory, 1023... Position detection signal input port, 105... Servo motor, 10
6... Resolver. Patent applicant Representative of FANUC Co., Ltd. Patent attorney Minoru Tsuji (1 other person) Figure 3

Claims (2)

[Claims] (1) An interpolation error measuring means that rotates the resolver to measure the interpolation error of the resolver, a data storage means that stores the data obtained by the measuring means in an auxiliary storage device, and a reference point setting for absolute position detection. means for obtaining a reference value r_0 for absolute position detection corrected based on the value r_0' read from the resolver at the time and the interpolation error ε_0 of the resolver obtained from the measuring means; and obtaining the absolute position at an arbitrary point. In doing so, an arbitrary value is calculated from the value r' read by the resolver at that time, the corrected value r at any point obtained based on the interpolation error ε' of the resolver obtained from the measuring means, and the reference value r_0. An absolute position detection method comprising: means for determining a corrected absolute position at a point. (2) The absolute position detection method according to claim (1), wherein the auxiliary storage device is a bubble memory.