JP4290405B2 - Anomaly detection device for location information - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an abnormality detection device for position information output from an absolute encoder, and for example, relates to an abnormality detection device that improves the reliability of a linear absolute encoder that detects the position of a feed axis of a machine tool in an absolute manner.
[0002]
[Prior art]
In recent years, drive systems using linear motors have been actively used for machine tools and the like for high speed and high accuracy. In a drive system using a linear motor, a linear absolute encoder that detects the position of the movable part in an absolute manner is used for detecting the position and speed of the movable part and the motor magnetic pole position (see Patent Document 1).
[0003]
As a conventional linear absolute encoder, as shown in the block diagram of FIG. 6, an incremental position sensor that outputs a signal having a phase difference of 90 degrees according to a change in position and an irregularity such as an M series according to a change in position. There is an absolute encoder equipped with an absolute position sensor that outputs a cyclic random number code signal. Although not shown in the figure, the LED 3, the lens 2, the light receiving element 4, and the linear image sensor 5 are fixed to the same member and can move relative to the scale 1. The light emitted from the LED 3 is converted into parallel light by the lens 2 and irradiated to the scale 1. On the scale 1, there are formed an incremental pattern 11 in which light and dark repeats at intervals of 20 μm and an M-sequence pattern 12 having a circulation period 2 ¹⁵ -1 indicating a 1-bit code at 20 μm.
[0004]
The light that has passed through the incremental pattern 11 is converted into an electrical signal by the light receiving element 4, and is output as a signal SA proportional to the sine value of the displacement and a signal CA proportional to the cosine value in accordance with a change in the relative position with respect to the scale. Is done. The signals SA and CA are converted into digital values SD and CD by AD converters 6 and 7, respectively. The digital values SD and CD are subjected to arc tangent calculation by the interpolation processing means 10 to convert the relative position within 20 μm into a numerical value PAL indicating absolute. The signals SA and CA are converted by the comparators 8 and 9 into a pulse signal SP and a pulse signal CP, respectively. The pulse signal SP and the pulse signal CP are counted by the up / down counter 13 as a numerical value PCH indicating the change of the relative position in increments of 5 μm in accordance with the phase change of the pulse signal SP and the pulse signal CP. The digit aligning means 14 corrects the numerical value PCH to a numerical value in units of 20 μm that changes in synchronization with the change of the numerical value PAL, and outputs it as incremental position information PI with the corrected numerical value PIH as the upper digit and the numerical value PAL as the lower digit. . Here, the entire components from the incremental pattern 11, the light receiving element 4, the signal SA and the signal CA to the output of the incremental position information PI will be referred to as an incremental position sensor.
[0005]
On the other hand, the circulation period 2 ¹⁵ light passing through the M-sequence pattern 12 -1 by the linear image sensor 5, the light amount change range exceeding the 15-bit M-sequence pattern of the circulating period 2 ¹⁵ -1 circulation period 2 ¹⁵ -1 Are detected with a resolution equal to or less than half the pitch corresponding to one bit of the M series and output as light amount data VO. In the linear image sensor signal processing means 15, among the light quantity data obtained from the M-sequence pattern of the circulation cycle 2 ¹⁵ -1 detected by the linear image sensor 5, the location most suitable for encoding is determined according to the value of the numerical value PAL. Data is selected every 20 μm and binarized to convert it into an M-sequence code MH having a circulation period of 2 ¹⁵ -1 of 15 bits or more. Here, the entire component until the M-sequence code MH is output is referred to as an absolute position sensor.
[0006]
The decoder 16 uses the code read out at the origin position of scale 1 as an initial value, and generates an M sequence until the M sequence code obtained by repeating the M sequence code generation operation of the circulation period 2 ¹⁵ -1 matches the code MH. The calculation is repeated, and a numerical value PH indicating the number of repetitions of the calculation when they coincide is output. At this time, the numerical value PH is a numerical value indicating the relative movement amount from the origin in an absolute manner. The subtracter 17 subtracts the upper digit PIH of the incremental position information PI from the numerical value PH indicating the absolute position, thereby outputting a numerical value POF indicating the offset position amount of both. The numerical value POF is set in the storage device 18 immediately after the linear absolute encoder is turned on or reset. The adder 19 adds the number MOF indicating the offset position amount stored in the storage unit 18 to the incremental position information PI, thereby correcting the relative movement amount with respect to the scale 1 and outputting the absolute position information PAO.
[0007]
Further, the unit conversion means 20 converts the position information PAO into a bit unit value of the absolute position sensor and outputs it as a numerical value PAM. The M-sequence conversion means 21 has a look-up table that stores in advance the M-sequence of the circulation cycle 2 ¹⁵ -1 corresponding to the position information, and is a code that is an M-sequence code of the circulation cycle 2 ¹⁵ -1 corresponding to the numerical value PAM. MP is output. The abnormality determiner 22 compares the coincidence state between the code MP and the code MH output from the linear image sensor signal processing means 15, and if there is a large difference between the two codes, outputs an abnormality signal ER to the position information PAO. Inform the outside that there is an abnormality.
[0008]
The linear absolute encoder shown in FIG. 6 detects absolute position information based on a signal from an absolute position sensor only immediately after power-on or reset, and thereafter performs position processing by incremental processing based on an output signal from the incremental position sensor. Information is being updated. As described above, the reason for using the incremental position sensor signal without using the signal from the absolute position sensor to update the position information is that it takes a lot of time to output the signal from the linear image sensor and to decode the M-sequence code. This is because the position information update cycle is inappropriate for controlling the linear motor or the like. However, the position information based on the output signal of the incremental position sensor is not reliable as absolute position information. For this reason, by comparing the difference between the position information and the data based on the absolute position sensor signal in a cycle longer than the position information update cycle, it is possible to detect an abnormality in the position information and ensure the reliability of the position information. ing.
[0009]
[Patent Document 1]
Japanese Patent No. 311546 [0010]
[Problems to be solved by the invention]
In the position information abnormality detection device of the conventional example, when converting the position information into the M-sequence code, a memory having a large storage capacity for realizing the lookup table (for example, 2 ¹⁵ -1 M sequence: 32,767 bytes) ), The abnormality detection circuit becomes larger and the cost is increased. Further, as a method of converting position information into an M-sequence code, there is a method of obtaining an M-sequence code corresponding to position information by repeating the operation of the M-sequence code calculation circuit a number of times corresponding to the position information. However, this method does not require a look-up table, but the processing time for obtaining the M-sequence code is long and the period for detecting an abnormality is long. Therefore, the reliability of position information necessary for controlling a linear motor or the like is increased. There was a problem that it could not be secured.
[0011]
The present invention has been made under the circumstances as described above, and an object of the present invention is to detect position information detected by an absolute encoder having an absolute position sensor that outputs an irregular cyclic random number code signal and an incremental position sensor. The present invention relates to an abnormality detection apparatus for ensuring reliability, and in particular, to provide an abnormality detection apparatus for position information that can perform abnormality detection at low cost and at high speed.
[0012]
[Means for Solving the Problems]
The position information abnormality detection device according to the present invention is based on the exclusive OR of the M sequence of the circulation cycle 2 ⁱ −1 and the M sequence of the circulation cycle 2 ^j −1 having a relatively simple circulation cycle according to the change in position. Do Ri, a product of the circulation cycle of both the absolute position sensor than any of circulation cycles of the M series code and outputs the irregular circulation random code signals that have a large circulation period, depending on a change in the position An incremental position sensor that outputs a periodic signal, a unit converter that converts the position information obtained based on the output signal of the incremental position sensor into a value in bit units of the absolute position sensor, and a unit converter after the unit conversion a remainder K obtained by dividing the positional information in a circulating cycle 2 ⁱ -1 and calculating means for calculating a remainder L divided by the circulation cycle 2 ^j -1, to correspond to the K-th M sequence code of the circulating cycle 2 ⁱ -1 N bits (N ≧ i + j) one of an encoder for outputting a code M _i of the, other encoder for outputting a code M _j of N bits corresponding to the L-th M sequence code of the circulating cycle 2 ^j -1 The N-bit code M _i output from one encoder, the N-bit code M _j output from the other encoder, and the N-bit irregular cyclic random code signal output from the absolute position sensor. An abnormality detection apparatus for position information including an abnormality determination device for detecting an abnormality in position information.
[0013]
[Action]
As described above, according to the present invention, the irregular circulation output from the absolute position sensor can be obtained simply by obtaining two M-sequence codes having a circulation period smaller than the circulation period of the irregular cyclic random number code output from the absolute position sensor from the position information. Abnormality of the random number code can be detected. For this reason, it is possible to use a lookup table having a smaller capacity than the conventional apparatus. Even when two M-sequence codes are obtained using an M-sequence code calculation circuit, it is possible to obtain an M-sequence code at a relatively high speed because of a small circulation cycle. As described above, according to the present apparatus, position information abnormality detection can be performed at low cost and at high speed.
[0014]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0015]
FIG. 1 is a block diagram of a linear type absolute encoder to which the position information abnormality detection device of the present invention is applied. In the figure, components having the same functions as those in FIG. FIG. 2 is a table showing an irregular cyclic random number code formed on the scale 101 and two M sequences that are the basis thereof. FIG. 3 is a table showing a lookup table of the encoder 103 in FIG. FIG. 4 is a table showing a lookup table of the encoder 105 of FIG. FIG. 5 is a block diagram illustrating the operation of the decoder 116.
[0016]
In FIG. 1, the scale 101 is formed with an incremental pattern 11 that repeats light and dark at intervals of 20 μm and an irregular cyclic random code pattern 112 that indicates a 1-bit code at 20 μm. In the table of FIG. 2, the light transmissive portions at intervals of 20 μm of the irregular cyclic random number code pattern 112 are shown as 1 and the light shielding portion 0. As can be understood from the table, the irregular cyclic random number code is a code that is an exclusive OR of the M sequence having 2 ⁷ -1 cycles and the M sequence having 2 ⁸ -1 cycles. In this way, a code based on an exclusive OR of M sequences having a relatively simple circulation cycle is called a Gold sequence, and it is known that the circulation cycle is the product of both M sequence circulation cycles. Therefore, the cyclic cycle of the irregular cyclic random number code pattern 112 is 32,385, which is almost the same as the M series of 2 ¹⁵ -1 cyclic cycle (32,768) in FIG. is there.
[0017]
The linear image sensor 5 outputs a light amount change in a range of 16 bits or more in 1 bit conversion of the pattern 112 out of the light passing through the irregular circulation random number code pattern 112 as the light amount data VGO. In the linear image sensor signal processing means 15, among the light quantity data VGO obtained from the irregular circulation random number code pattern detected by the linear image sensor 5, the data of the most suitable place for encoding according to the value of the numerical value PAL is 20 μm. By selecting every other and binarizing it, it is converted into a 16-bit irregular cyclic random number code GH. The code GH is input to the decoder 116.
[0018]
In the decoder 116, as shown in the block diagram of FIG. 5, the M-sequence generator 24 of the circulation cycle 2 ⁷ -1, the M-sequence generator 25 of the circulating cycle 2 ⁸ -1, these M-sequence generator From an exclusive logic unit 26 that performs an exclusive OR operation on the signals M7 and M8 output from the signal 16 and a 16-bit shift register 23 that converts the serially output signal X15 after the exclusive OR operation into a 16-bit parallel code GD. It has a gold series generator 30 configured. When the code GH is input, the gold series generation unit 30 once initializes the code GD to a value corresponding to the code read at the origin position of the scale 101. The comparator 27 compares the 16-bit codes GH and GD, and outputs a clock generation command ST to the clock signal generator 28 until both values match. Clock signal generator 28, while the clock generation command ST is input, the M-sequence generator 25 and 16-bit shift register 23 with the M sequence generator 24 of the circulation cycle 2 ⁷ -1 circulation cycle 2 ⁸ -1 A clock signal CK for shifting and counting is output to the counter 29 at regular intervals. From the above, the decoder 116 repeats the gold sequence generation calculation until the irregular cyclic random number code GH and the gold sequence code GD match, and the counter 29 outputs a numerical value PH indicating the number of repetitions of the calculation when the match occurs. . At this time, the numerical value PH is a numerical value indicating the relative movement amount from the origin in an absolute manner.
[0019]
The unit converter 20 converts the position information PAO obtained based on the output signal of the incremental position sensor into a value in bit units of the absolute position sensor and outputs it as a numerical value PAM. When the numerical value PAL obtained by the incremental position sensor as shown in FIG. 1 or FIG. 6 is equal to the bit unit of the absolute position sensor, the unit conversion means 20 discards the lower data of the position information PAO and only the upper data of 20 μm or more. The unit conversion means 20 can also be expressed by deleting it from the block diagram.
[0020]
The dividing means 102 outputs a remainder K obtained by dividing the numerical value PAM by 127 (2 ⁷ −1). The encoder 103 has a look-up table indicating the first 8 bits of the M-sequence of the K-th cycle 2 ⁷ −1 shown in FIG. 3, and from the input numerical value K, the K-th is an upper code and the K + 8-th A 16-bit code Mi having a lower code of (K + 8> 126 only, K + 8-127th) is output. As a result, the 16-bit code Mi becomes an M-sequence code of the Kth circulation cycle 2 ⁷ -1. The exclusive OR calculator 106 calculates the exclusive OR of the 16-bit irregular cyclic random number code GH and the 16-bit 2 ⁷ −1 M-sequence code Mi, and outputs a 16-bit code GX.
[0021]
On the other hand, the dividing means 104 outputs a remainder L obtained by dividing the numerical value PAM by 255 (2 ⁸ −1). The encoder 105 has a lookup table indicating the first 8 bits of the M-sequence of the L-th circulation period 2 ⁸ −1 shown in FIG. 4, and from the input numerical value L, the L-th is an upper code, and the L + 8-th A 16-bit code Mj having a lower code of (L + 8> 255 only, L + 8-255th) is output. As a result, the 16-bit code Mj becomes the M-sequence code of the Lth circulation cycle 2 ⁸ -1. The abnormality determiner 122 compares the coincidence state of the 16-bit code GX and the 16-bit code Mj, and if there is a large difference between the two codes, outputs an abnormality signal ER to indicate that the position information PAO is abnormal. To inform. From the above, the memory capacity required for the lookup table of the encoder 103 and the encoder 105 is 382 bytes in total of 127 bytes and 255 bytes, respectively, which is much smaller than the conventional 32,767 bytes. Anomaly detection is possible with memory.
[0022]
In the embodiment of FIG. 1, the result GX after the exclusive logical operation of the 16-bit code Mi and the 16-bit irregular cyclic random number code GH is compared with the 16-bit code Mj. Anomaly detection is possible even if the result after the exclusive logical operation of the 16-bit code Mj is compared with the 16-bit irregular cyclic random number code GH.
[0023]
Further, the encoder 103 and the encoder 105 can be implemented using a circulating cycle 2 ⁷ -1 M-sequence generator 24 and the circulation period of 2 ⁸ -1 M sequence generator 25 of FIG. 5, respectively It is.
[0024]
Further, in this embodiment, the irregular cyclic random number code GH from the absolute position sensor is processed with a size of 16 bits. However, the reliability of abnormality detection can be improved by increasing the bit size as in the present embodiment. Furthermore, by increasing the bit size, it is possible to increase redundancy in the case where the scale is partially dirty.
[0025]
【The invention's effect】
As described above, according to the absolute position abnormality detection device of the present invention, it is possible to achieve high-speed abnormality detection with a small and low-cost realization means.
[Brief description of the drawings]
FIG. 1 is a block diagram of a linear absolute encoder to which an abnormality detection method of the present invention is applied.
FIG. 2 is a table showing irregular cyclic random number codes formed on a scale 101;
FIG. 3 is a table showing a lookup table of the encoder 103 in FIG. 1;
FIG. 4 is a table showing a look-up table of the encoder 105 of FIG.
5 is an internal block diagram of the decoder 116 of FIG. 1. FIG. 6 is a block diagram of a conventional linear absolute encoder.
[Explanation of symbols]
1,101 scale, 2 lens, 3 LED, 4 light receiving element, 5 linear image sensor, 6,7 AD converter, 8,9 comparator, 10 interpolation means, 11 incremental pattern, 12,112 irregular cyclic random code pattern , 13 Up / down counter, 14 digit alignment means, 15 linear image sensor signal processing means, 16,116 decoder, 17 subtractor, 18 memory, 19 adder, 20 unit converter, 21, 103, 105 encoding , 22, 122 abnormality determination unit, 23 shift register, 24, 25 M-sequence generator, 26, 106 exclusive OR calculator, 27 comparator, 28 clock signal generator, 29 counter.

Claims (4)

It consists of the exclusive OR of the M sequence of the circulation cycle 2 ⁱ −1 and the M sequence of the circulation cycle 2 ^j −1 whose circulation cycles are relatively disjoint according to the change in position , and absolute position sensor than any of circulation cycles of the M series code and outputs the irregular circulation random code signals that have a large circulation period,
An incremental position sensor that outputs a periodic signal in response to a change in position;
A unit converter for converting the position information obtained based on the output signal of the incremental position sensor into a value in bit units of the absolute position sensor;
Calculating means for obtaining a remainder K obtained by dividing the position information after the unit conversion by a circulation cycle 2 ⁱ −1 and a remainder L obtained by dividing the position information by a circulation cycle 2 ^j −1;
One encoder that outputs a code M _i of N bits (N ≧ i + j) corresponding to the Kth of the M-sequence code of the cycle period 2 ⁱ −1 ;
The other encoder that outputs an N-bit code M _j corresponding to the L-th of the M-sequence code of the circulation period 2 ^j −1 ;
The position is determined by an N-bit code M _i output from one encoder, an N-bit code M _j output from the other encoder, and an N-bit irregular cyclic random code signal output from the absolute position sensor. An apparatus for detecting an abnormality in position information, including an abnormality determiner for detecting an abnormality in information. An arithmetic unit that performs an exclusive OR operation on an N-bit irregular cyclic random number code signal output from the absolute position sensor and an N-bit code M _i output from one encoder, and a code after the exclusive OR operation The position information abnormality detection device according to claim 1, further comprising: an abnormality determination unit that detects abnormality by comparing an N-bit code M _j output from the other encoder . An arithmetic unit that performs an exclusive OR operation on an N-bit code M _i output from one encoder and an N-bit code M _j output from the other encoder, and an N-bit code after the exclusive OR operation. The position information abnormality detection device according to claim 1, further comprising: an abnormality determination unit that detects an abnormality by comparing a code and an N-bit irregular cyclic random number code output from the absolute position sensor. Other for outputting the M-sequence code of N bits of the circulating cycle 2 ^j -1 corresponding to one encoder or L-th outputs the M-sequence code of N bits of the circulating cycle 2 ⁱ -1 corresponding to K th The position information abnormality detection apparatus according to claim 1, wherein a lookup table is used as the encoder .

Images