WO2020026832A1

WO2020026832A1 - Photoelectric sensor

Info

Publication number: WO2020026832A1
Application number: PCT/JP2019/028266
Authority: WO
Inventors: 火炎木焦
Original assignee: オムロン株式会社
Priority date: 2018-08-01
Filing date: 2019-07-18
Publication date: 2020-02-06
Also published as: JP2020022091A; TW202008724A; JP7035890B2; TWI691166B

Abstract

Provided is a photoelectric sensor that determines the state of a target by means of a simple configuration and without a time lag. A photoelectric sensor that comprises a light projection unit, a light reception unit, FIFO memory, a model storage unit, and a determination unit. The light projection unit emits light toward a detection area to which a target will arrive. The light reception unit acquires chronological signal values that are based on received light. The FIFO memory stores a prescribed number of signal values in the order of the acquisition thereof and periodically updates the prescribed number of signal values using newly acquired signal values. The model storage unit stores a determination model that determines a rank for the agreement between: a waveform that is formed from the prescribed number of signal values stored in the FIFO memory; and a reference waveform that corresponds to a specific state for the target. The determination unit uses the determination model to make a determination each time the FIFO memory is updated or once every time the FIFO memory is updated a plurality of times and determines the state of the target on the basis of the rank for the agreement.

Description

Photoelectric sensor

The present invention relates to a photoelectric sensor having a function of determining a state of an object.

Conventionally, as a sensor that detects the presence or absence of an object, it irradiates the object with light, detects light passing through the object, detects blocking of light by the object, and detects light reflected by the object Photoelectric sensors are used. In addition, when detecting the state of the target object instead of the presence or absence of the target object, a visual sensor that images the target object with a camera and performs image analysis may be used.

Regarding the photoelectric sensor, for example, Patent Document 1 below stores a detection value corresponding to a background level as a zero reset reference value so that an arbitrary detection value can be displayed as a relative value based on the background level. A structured photoelectric sensor is described.

Further, in Patent Document 2 below, a steel sheet surface is scanned with a laser beam, a plurality of characteristic amounts representing a reflected light waveform are calculated, and the characteristic amount is added to a neural network trained in advance, and a flaw / An inspection method for performing no output is described.

JP 2001-124594 A JP-A-2-298840

It is more difficult to detect the presence or absence of an object with a single popular photoelectric sensor, but the state of the object is not so large as to require various capabilities of a visual sensor that is large and expensive compared to the photoelectric sensor. There is a demand for detection. For example, when discriminating an object having a significantly different shape or pattern, simply detecting the presence or absence of the object is not sufficient, but the various capabilities of the visual sensor are not required.

Here, by combining the conventional photoelectric sensor with a method of analyzing the waveform of the amount of received light as disclosed in Patent Document 1, it can be considered that the state of the target object can be determined in principle. However, since an appropriate trigger for acquiring only a waveform of a portion necessary for determination (a portion corresponding to the size of a flaw in Patent Document 2) cannot be obtained, a waveform in a time range longer than that required for determination is not obtained. After completion of the acquisition, waveform analysis is performed. For example, real-time performance is lacking when applied to an object that is successively carried on a transport line.

Therefore, the present invention provides a photoelectric sensor that determines the state of an object with a simple configuration and with a small time delay.

A photoelectric sensor according to an aspect of the present disclosure is a light emitting unit that emits light toward a detection range where an object arrives, a light receiving unit that acquires a time-series signal value based on light reception, A FIFO memory that stores a predetermined number of signal values in order and periodically updates the predetermined number of signal values with newly acquired signal values, and a predetermined number of signal values stored in the FIFO memory. Storage unit for storing a determination model for determining the rank of the degree of coincidence between the waveform and the reference waveform corresponding to the specific state of the object, and a frequency once every time the FIFO memory is updated one or more times A determination unit that performs a determination by the determination model and determines the state of the target object based on the rank of the degree of coincidence.

According to this aspect, the waveform composed of the signal values stored in the FIFO memory and the reference waveform corresponding to the specific state of the target object are updated once every time the FIFO memory is updated one or more times. By determining the rank of the degree of coincidence with, it is possible to determine the state of the objects that are successively transported on the transport line with a simple configuration and with little time delay.

{In the above aspect, the judgment model} may be a learned model generated by machine learning.

According to this aspect, the learned model determines the rank of the degree of coincidence between the waveform constituted by the signal values stored in the FIFO memory and the reference waveform corresponding to the specific state of the target object, thereby enabling the conveyance. It is possible to more flexibly determine the state of an object that is successively carried on a line.

In the above aspect, the determination model includes calculating a degree of coincidence from a difference between a predetermined number of signal values stored in the FIFO memory and a reference value corresponding to the predetermined number of signal values and representing a reference waveform. It may be.

According to this aspect, the rank of the degree of coincidence between the waveform constituted by the signal values stored in the FIFO memory and the reference waveform corresponding to the specific state of the object is determined by a relatively simple model. It is possible to determine the state of an object that is successively transported on a transport line at a higher speed.

In the above aspect, the determination model is a model that determines that the degree of coincidence is high when the degree of coincidence is higher than a predetermined value. When the result of the determination of the high rank is obtained, the determination unit The state of an object may be determined to be a specific state.

In the above aspect, the determination model is a model that further determines that the matching degree is the middle rank when the matching degree is not higher than the predetermined value but is higher than a second predetermined value that is smaller than the predetermined value. In the range of time required for the object to pass through the detection range, when there is no case of the high rank and there is a case of the middle rank, even if it is determined that an object that is not in a specific state has arrived Good.

According to this aspect, even when the state of the target is not the specific state, it is possible to determine that the target has arrived and that the state of the target is not the specific state.

In the above aspect, an operation control unit that generates a determination model based on the signal value and stores the generated determination model in the model storage unit may be further provided.

According to this aspect, since the photoelectric sensor can generate the determination model by itself, the determination model generated according to the actual target object can be used without acquiring the determination model from outside.

In the above aspect, the operation control unit is configured to output a signal belonging to the fluctuation period when a fluctuation period in which the fluctuation of the time-series signal value is relatively large follows the stable period in which the fluctuation of the time-series signal value is relatively small. The determination model may be generated based on the value.

According to this aspect, the determination model can be generated by selectively using the signal value generated by the object from the signal values.

In the above aspect, the operation control unit may be capable of outputting the determination model to the outside.

According to this aspect, since the generated determination model can be used by another photoelectric sensor, it is not necessary to repeat generation of the determination model for each of a plurality of photoelectric sensors used in the same target object and installation situation.

In the above aspect, the operation control unit may be capable of outputting a time-series signal value to the outside.

According to this aspect, it is possible to output the signal value to the outside and generate the judgment model by the external device. This eliminates the need for the photoelectric sensor itself to have computational resources for the process of generating the determination model.

In the above aspect, the operation control unit may acquire the determination model from outside and store the determination model in the model storage unit.

According to this aspect, the generation of the determination model can be omitted by diverting the determination model generated by another device, for example, another photoelectric sensor.

According to the present invention, there is provided a photoelectric sensor that determines the state of an object with a simple configuration and with a small time delay.

It is a figure showing the outline of the detection system containing the photoelectric sensor concerning the embodiment of the present invention. It is a figure showing composition of a photoelectric sensor concerning this embodiment. It is a figure showing an example of composition of a processing part of a photoelectric sensor concerning this embodiment. 5 is a flowchart of processing in a learning mode and a determination mode of the photoelectric sensor according to the embodiment. 5 is a flowchart of a first example of a process of determining a state of an object by the photoelectric sensor according to the embodiment. It is a flowchart of the 2nd example of the process which determines the state of a target object by the photoelectric sensor which concerns on this embodiment. FIG. 9 is a diagram illustrating an example of a signal value measured in an n-th cycle of the photoelectric sensor according to the embodiment. FIG. 7 is a diagram illustrating an example of a signal value measured in an (n + 1) th cycle of the photoelectric sensor according to the embodiment. FIG. 9 is a diagram illustrating another example of a signal value measured in an n-th cycle of the photoelectric sensor according to the embodiment. It is a figure showing other examples of the composition of the processing part of the photoelectric sensor concerning this embodiment. It is a figure showing the example which installs a judgment model from the outside in the photoelectric sensor concerning this embodiment.

Hereinafter, an embodiment according to one aspect of the present invention (hereinafter, referred to as “the present embodiment”) will be described with reference to the drawings. In each of the drawings, the components denoted by the same reference numerals have the same or similar configurations.

[Configuration example]
An example of the configuration of the photoelectric sensor 10 according to the present embodiment will be described with reference to FIGS. FIG. 1 is a diagram illustrating an outline of a detection system 1 including a photoelectric sensor 10 according to the present embodiment. The detection system 1 includes a photoelectric sensor 10, a controller 20, a computer 30, a robot 40, and a transfer device 50.

The photoelectric sensor 10 is a device that detects that the object 100 has arrived in the detection range 10a of the photoelectric sensor 10 based on the acquired signal value, and determines the state of the object 100. The photoelectric sensor 10 may be a reflection type photoelectric sensor, a transmission type photoelectric sensor, or a regression reflection type photoelectric sensor. The photoelectric sensor 10 may be a displacement sensor that emits a laser beam to the object 100 and obtains a signal value corresponding to the distance to the object 100 based on the principle of triangulation. Further, the photoelectric sensor 10 may be a distance measuring sensor that obtains a signal value corresponding to the distance to the object 100 based on the round trip time of light reflected by the object 100. In this specification, the “signal value” includes a signal value corresponding to the distance to the object 100 in addition to the value of the amount of received light.

The target object 100 is an object to be detected by the photoelectric sensor 10, and may be, for example, a completed product to be produced or an unfinished product such as a part. The object 100 illustrated in FIG. 1 is an object having a shape with a projection on a base. In addition, it is assumed that an object having the same base but having no protrusion is mixed and transported as a heterogeneous object. When the photoelectric sensor 10 is, for example, a reflective photoelectric sensor, when the target object 100 reaches the detection range 10a of the photoelectric sensor 10, the amount of reflected light detected increases. When the object 100 has a shape with a projection on the base, the amount of reflected light further increases if the object 100 has a projection in the detection range 10a.

(4) The controller 20 controls the robot 40 and the transfer device 50. The controller 20 may be composed of, for example, a PLC (Programmable Logic Controller). The controller 20 detects that the object 100 has arrived based on the output from the photoelectric sensor 10, and further controls the robot 40 according to the determined state of the object 100.

(4) The computer 30 sets the photoelectric sensor 10, the controller 20, and the robot 40. Further, the computer 30 acquires, from the controller 20, an execution result of the control by the controller 20. Furthermore, the computer 30 may include a learning device that generates a determination model for determining the state of the target object 100 by the photoelectric sensor 10 by machine learning. Here, the judgment model may be constituted by, for example, a neural network or a decision tree.

The robot 40 operates and processes the target object 100 under the control of the controller 20. The robot 40 may, for example, pick up the object 100 and move it to another location, or cut or assemble the object 100. The robot 40 may change the processing content or the destination depending on whether or not the target object 100 has a protrusion.

The transport device 50 is a device that transports the target object 100 under the control of the controller 20. The transfer device 50 may be, for example, a belt conveyor, and may transfer the target object 100 at a speed set by the controller 20.

FIG. 2 is a diagram illustrating a configuration of the photoelectric sensor 10 according to the present embodiment. The photoelectric sensor 10 includes a light emitting unit 11, a light receiving unit 12, a processing unit 13, an operation unit 14, and an output unit 15.

<Light emitting part>
The light projecting unit 11 emits light toward the detection range 10a where the object 100 arrives. The light projecting unit 11 may include a light projecting element 11a and a drive circuit 11b. The light emitting element 11a may be configured by an LED (Light Emitting Diode) or a laser diode, and the drive circuit 11b controls a current for causing the light emitting element 11a to emit light. The drive circuit 11b may cause the light emitting element 11a to emit pulses intermittently, for example, at a period of 0.1 ms. The light emitted from the light projecting element 11a may be applied to the detection range 10a via a lens or an optical fiber (not shown).

<Light receiving section>
The light receiving unit 12 acquires a time-series signal value based on light reception. The light receiving section 12 may include a light receiving element 12a, an amplifier 12b, a sample / hold circuit 12c, and an A / D converter 12d. The light receiving element 12a may be constituted by a photodiode, and converts the amount of received light into an electric output signal. The light receiving unit 12 may make the light reflected or transmitted in the detection range 10a incident on the light receiving element 12a via a lens or an optical fiber (not shown). The amplifier 12b amplifies the output signal of the light receiving element 12a. The sample / hold circuit 12c holds the output signal of the light receiving element 12a amplified by the amplifier 12b in synchronization with the timing of pulse emission by the light projecting unit 11. Thereby, the influence of disturbance light is reduced. The A / D converter 12d converts the analog signal value held by the sample / hold circuit 12c into a digital value of the amount of received light.

<Processing unit>
The processing unit 13 includes an operation control unit 13a, a FIFO (First In First Out) memory 13b, a model storage unit 13c, and a determination unit 13d. The processing unit 13 may be configured as, for example, a computer including a microprocessor, a memory, and a program stored in the memory.

The operation control unit 13a may integrally control the operation of the photoelectric sensor 10 as a whole, in addition to processing for handling a determination model described later.

The メモリ FIFO memory 13b stores a predetermined number of signal values in the order of acquisition, and periodically updates the predetermined number of signal values with newly acquired signal values. Here, the number of signal values stored in the FIFO memory 13b, that is, the predetermined number is arbitrary, but may be, for example, about 100. The FIFO memory 13b may be realized by dedicated hardware, or may be realized on a memory of the processing unit 13 according to a program of the processing unit 13. In this case, the shift of the signal value to the subsequent stage of the FIFO memory 13b can be performed not by physically shifting the stored data but by updating the access location on the memory.

The model storage unit 13c stores a determination model for determining a rank of the degree of coincidence between a waveform composed of a predetermined number of signal values stored in the FIFO memory 13b and a reference waveform corresponding to a specific state of the object 100. I do. Here, the reference waveform may be a waveform of a typical signal value corresponding to a specific state of the object 100, for example, an average of waveforms obtained for a plurality of objects 100 in a specific state. Good.

The model storage unit 13c may store a learned model generated by machine learning as a determination model. The learned model is learned so as to determine the rank of the degree of coincidence between a waveform composed of a predetermined number of signal values stored in the FIFO memory 13b and a reference waveform corresponding to a specific state of the object 100. Good. Here, the learned model may be generated by the computer 30 and stored in the model storage unit 13c.

The determination unit 13d performs the determination based on the determination model once every time the update of the FIFO memory 13b is performed once or a plurality of times, and generates a waveform composed of a predetermined number of signal values stored in the FIFO memory 13b. The state of the object 100 is determined based on the rank of the degree of coincidence with the reference waveform corresponding to the specific state of the object 100. For example, when the reference waveform is a waveform of an object having a protrusion on the base, the determination unit 13d performs a determination using a determination model, and when the degree of coincidence is sufficiently high, the object 100 is It may be determined that the state is attached. In this manner, the waveform constituted by the signal values stored in the FIFO memory 13b and the reference corresponding to the specific state of the object 100 are updated once every time the FIFO memory 13b is updated one or more times. By determining the rank of the degree of coincidence with the waveform, it is possible to determine the state of the object 100 that is successively transported on the transport line with a simple configuration and with little time delay. Accordingly, the state of the target object 100 can be determined with a simple configuration close to a widely used photoelectric sensor, that is, without the need for image processing or a separate trigger unit and with a small time delay.

The judgment model is a model including calculating the degree of coincidence from a difference between a predetermined number of signal values stored in the FIFO memory 13b and a reference value corresponding to the predetermined number of signal values and representing a reference waveform. May be.

The determination model may be a model that determines that the degree of coincidence is high when the degree of coincidence is higher than a predetermined value. When the determination result of the high rank is obtained, the determination unit 13 d May be determined to be a specific state. When a high rank determination result is obtained, the determination unit 13d may maintain the output indicating that the state of the target object 100 is in a specific state for a predetermined period even after the high rank determination result disappears. Here, the predetermined period may be approximately the time required for the object 100 to pass through the detection range 10a. The determination unit 13d may determine the state of the target object 100 a plurality of times within a time range required for the target object 100 to pass through the detection range 10a. Here, "the range of time required for the object 100 to pass through the detection range 10a" may be incorporated in the determination model as the duration of the signal value fluctuation actually measured when the determination model is generated. Further, at the time of the determination operation, without detecting the start of the signal value fluctuation, the determination result obtained from the present to the past by the number of shifts corresponding to the duration is referred to as “the time required to pass through the detection range 10a”. Range ". Note that the period from when the start of the signal value fluctuation is detected to when the number of shifts corresponding to the duration is shifted may be defined as "the range of time required to pass through the detection range 10a". As described above, within the time required for the object 100 to pass through the detection range 10a, the waveform constituted by the signal values stored in the FIFO memory 13b and the reference corresponding to the specific state of the object 100 It can be determined whether or not the rank of the degree of coincidence with the waveform is higher than a predetermined value.

The determination model may be a model that determines that the degree of coincidence is the middle rank when the degree of coincidence is not higher than the predetermined value but is higher than a second predetermined value that is smaller than the predetermined value. When there is no high rank and there is a middle rank within the time required for the object 100 to pass through the detection range 10a, it may be determined that an object that is not in a specific state has arrived. The determination unit 13d determines the state of the target object 100 a plurality of times within the time required for the target object 100 to pass through the detection range 10a, and the determination result is not high rank in any of the plurality of determinations. Alternatively, when the result of the determination is at least one of the middle ranks among the plurality of determinations, it may be determined that an object that is not in a specific state has arrived. The second predetermined value is a value greater than zero. The second predetermined value may be a threshold value that is determined based on a signal value when the target object 100 that is not in a specific state reaches the detection range 10a. Accordingly, even when the state of the target object 100 is not the specific state, it is possible to determine that the target object 100 has arrived and that the state of the target object 100 is not the specific state. For example, when the specific state is a state in which the base has a protrusion, or when an object with only the base without the protrusion arrives, the determination unit 13d may determine that the matching degree is the middle rank.

The determination unit 13d may adjust the signal value magnification in consideration of the light amount deterioration due to window contamination or the like, and determine whether or not the coincidence is sufficiently high. That is, the determination unit 13d determines that the degree of coincidence between the waveform obtained by changing the magnification of the waveform constituted by the signal values stored in the FIFO memory 13b and the reference waveform corresponding to the specific state of the object 100 is higher than a predetermined value. The state of the target object 100 may be determined depending on whether or not. By doing so, the state of the target object 100 can be stably determined even when the amount of projected light or the amount of received light changes due to window contamination or the like.

When a plurality of types of objects are mixed and conveyed, the determination model includes a waveform composed of signal values stored in the FIFO memory 13b and a reference waveform corresponding to a specific state of any type of the object 100. If the degree of coincidence with is higher than a predetermined value, the model may be determined to have a high degree of coincidence. Further, when the degree of coincidence is a high rank or a middle rank, the determination model itself indicates which type of object is the determination result, so that the determination unit 13d is determined to be a high rank or a middle rank. The type of the target object 100 may be specified. Alternatively, the judgment model is prepared for each type of the object 100, and the judgment unit 13d specifies the type of the object 100 judged as the high rank or the middle rank depending on which judgment model judges the high rank or the middle rank. May be.

For example, when the first type of object and the second type of object are mixed and conveyed, the determination unit 13d determines the waveform constituted by the acquired signal values and the reference corresponding to the first type of object. When the degree of matching with the waveform is high, the state of the transported object may be determined to be a specific state of the first type of object. In addition, when the degree of coincidence between the waveform constituted by the acquired signal values and the reference waveform corresponding to the second type of object is high, the determination unit 13d determines the state of the object being conveyed. May be determined to be a specific state of the second type of object. Further, the determination unit 13d does not determine the high rank of any of the objects, and determines that the degree of coincidence between the waveform formed by the acquired signal values and the reference waveform corresponding to the first type of object is medium. In the case of the rank, the transported target object may be determined to be a first type target object that is not in a specific state. In addition, the determination unit 13d does not determine a high rank for any of the objects, and determines that the degree of coincidence between the waveform formed by the acquired signal values and the reference waveform corresponding to the second type of object is medium. In the case of the rank, the state of the transported object may be determined to be a second type of object that is not a specific state. When the first type of object and the second type of object are similar or the second predetermined value for determining the middle rank is a low value, it is configured by the acquired signal value. The coincidence between the generated waveform, the reference waveform corresponding to the first type of object, and the coincidence between the reference waveform corresponding to the second type of object may both result in a middle rank determination result. sell. In such a case, the determination unit 13d determines that the transported object is a first type of object that is not in a specific state or a second type of object that is not in a specific state. May not be specified.

<Operation section>
The operation unit 14 operates the photoelectric sensor 10 and may include an operation switch, a display, and the like. The operator of the photoelectric sensor 10 can use the operation unit 14 to input an instruction such as setting an operation mode of the photoelectric sensor 10 and to confirm an operation state. The photoelectric sensor 10 according to the present embodiment includes, as operation modes, a learning mode for generating a determination model and a determination mode for determining the state of the target object 100 using the generated determination model. Good.

<Output section>
The output unit 15 outputs various data including the determination result by the determination unit 13d. The output unit 15 may simply output the determination result by the determination unit 13d in binary. The photoelectric sensor 10 may include a communication unit instead of the output unit 15 so that a large amount of data can be input and output.

FIG. 3 is a diagram illustrating an example of a configuration of the processing unit 13 of the photoelectric sensor 10 according to the present embodiment. In the first cycle, the processing unit 13 shifts the signal value stored in each stage of the FIFO memory 13b to the next stage by one, and converts the digital value of the received light amount output from the A / D converter 12d. It is stored in the first stage q0. In the figure, the number of stages of the FIFO memory 13b is set to ten from q0 to q9 to explain the principle, but the number of stages of the FIFO memory 13b may be larger, for example, about 100. .

The first cycle for updating the FIFO memory 13b may be the same as or different from the cycle of pulse emission by the light projecting unit 11. The first cycle for updating the FIFO memory 13b may be the same as or different from the cycle of pulse emission of the light projecting unit 11 and conversion by the A / D converter 12d (referred to as a second cycle). You may. For example, the second cycle may be fixed to a value unique to the photoelectric sensor 10 (for example, 0.1 ms). The first cycle may be settable from the computer 30 shown in FIG. The first cycle needs to be determined so that the range of signal value waveforms to be processed simultaneously falls within the FIFO memory 13b. The first cycle is often longer than the second cycle, and may be, for example, 1 ms.

The determination unit 13d determines the rank of the degree of coincidence between the waveform formed by the signal values stored in the plurality of stages of the FIFO memory 13b and the reference waveform corresponding to the specific state of the object using a determination model. The state of the object is determined based on the rank of the coincidence, and the determination result is output to the operation control unit 13a in the first cycle.

The model storage unit 13c may store, as a determination model, a learned model that is generated by machine learning and determines a rank of a matching degree. Here, the determination model may be configured by, for example, a neural network or a decision tree, and may include a learned model generated by another known machine learning method. The determination unit 13d may perform the determination using the learned model once every time the FIFO memory 13b is updated one or more times, and determine the state of the target object based on the rank of the matching degree. .

The operation control unit 13a generates a judgment model based on the signal value, and stores the generated judgment model in the model storage unit 13c. For example, the operation control unit 13a may execute machine learning of a learning model based on the acquired signal value, generate a learned model, and store the generated learned model in the model storage unit 13c. As described above, the determination model can be generated by the operation control unit 13a. That is, since the photoelectric sensor can generate the determination model by itself, the determination model generated according to the actual target object can be used without acquiring the determination model from outside.

The operation control unit 13a may be capable of outputting the determination model to the outside. Thus, the generated determination model can be used by another photoelectric sensor, so that it is not necessary to repeat generation of the determination model for each of a plurality of photoelectric sensors used in the same target object and installation situation. Therefore, it is possible to efficiently prepare the photoelectric sensor that determines the state of the object.

The operation control unit 13a, based on a signal value belonging to the fluctuation period, when a fluctuation period in which the fluctuation of the time-series signal value is relatively large follows a stable period in which the fluctuation of the time-series signal value is relatively small. May be used to generate a judgment model. Here, the stable period in which the change in the time-series signal value is relatively small may be a period in which there is substantially no change in the time-series signal value when the influence of noise is subtracted. Further, the fluctuation period in which the fluctuation of the time-series signal value is relatively large may be a period in which the fluctuation of the time-series signal value is substantially present when the influence of noise is subtracted. As described above, by generating the determination model based on the signal values belonging to the fluctuation period, it is possible to generate the determination model by selectively using the signal value generated by the object from the signal values. Note that specific examples of a stable period in which the fluctuation of the time-series signal value is relatively small and a fluctuation period in which the fluctuation of the time-series signal value is relatively large will be described with reference to FIGS. 7A and 7B.

FIG. 4 is a flowchart of processing in the learning mode and the determination mode of the photoelectric sensor 10 according to the present embodiment. First, the photoelectric sensor 10 determines whether or not it is in a learning mode for generating a determination model (S10). The switching between the learning mode and the determination mode may be performed by the operation unit 14.

When the photoelectric sensor 10 is in the learning mode (S10: YES), the photoelectric sensor 10 acquires a time-series signal value, and generates a determination model by the operation control unit 13a (S11).

On the other hand, when the photoelectric sensor 10 is not in the learning mode (S10: NO), that is, when the photoelectric sensor 10 is in the determination mode, the photoelectric sensor 10 updates the FIFO memory 13b with a new signal value (S12), and stores the data in the FIFO memory 13b. The state of the object is determined by applying a determination model to the stored signal values (S13). Here, the determination process (S13) will be described in more detail with reference to FIGS.

(4) Thereafter, the photoelectric sensor 10 determines whether to end the determination mode (S14). The termination of the determination mode may occur when the operation of the photoelectric sensor 10 is terminated or when the mode is switched from the determination mode to the learning mode. If the determination mode is not ended (S14: NO), the photoelectric sensor 10 acquires the time-series signal values again (S12), and applies the determination model to the acquired signal values to change the state of the target object. A determination is made (S13). On the other hand, when ending the determination mode (S14: YES), the processes in the learning mode and the determination mode are ended.

FIG. 5 is a flowchart of a first example of the process (S13) of determining the state of the target by the photoelectric sensor 10 according to the present embodiment. First, the photoelectric sensor 10 determines the rank of the degree of coincidence between the waveform constituted by the acquired signal values and the reference waveform using the determination model (S131). The rank of the degree of coincidence may be represented by discrete values such as “high rank”, “medium rank”, and “low rank”.

If the matching degree is high (S132: YES), the photoelectric sensor 10 determines that the state of the target object is a specific state corresponding to the reference waveform (S133). On the other hand, if the degree of coincidence is not high (S132: NO), the photoelectric sensor 10 ends the determination process and determines whether the degree of coincidence is high again when determining the next cycle. Thus, the first example of the determination processing ends.

FIG. 6 is a flowchart of a second example of the process (S13) of determining the state of the target by the photoelectric sensor 10 according to the present embodiment. First, the photoelectric sensor 10 determines the rank of the degree of coincidence between the waveform constituted by the acquired signal values and the reference waveform using the determination model (S134).

(4) If the determined degree of coincidence is high (S135: YES), the photoelectric sensor 10 determines that the state of the target object is a specific state corresponding to the reference waveform (S136). On the other hand, if the degree of coincidence is not high (S135: NO) and the degree of coincidence is determined to be medium (S137: YES), the photoelectric sensor 10 determines without immediately determining the state of the target object. The process ends.

The degree of coincidence is not high (S135: NO), the degree of coincidence is not middle (S137: NO), and there is no case where the degree of coincidence is high during the time when the object passes through the detection range. When it is determined that there is a middle rank case (S138: YES), the photoelectric sensor 10 determines that an object that is not in a specific state has arrived (S139). On the other hand, even if the degree of coincidence is not high (S135: NO) and the degree of coincidence is not middle (S137: YES), the degree of coincidence is high within the time when the object passes through the detection range. If there is no case and it is not determined that there is a case of the middle rank (S138: NO), the determination processing ends.

In order to determine whether or not there is a case where the degree of coincidence is high and there is a case where the degree of middle is within the range of the time when the object passes through the detection range (S138), the photoelectric sensor 10 A series of determination results obtained by determining the rank of the matching degree by the determination model within the time required for the object to pass through the detection range may be stored in a not-shown FIFO memory for storing the determination results. Also, the photoelectric sensor 10 does not store a series of determination results in the FIFO memory, and measures the time required for the object to pass through the detection range when the determination model determines that the matching degree is the middle rank. If the high-rank judgment result does not appear before the timing ends, it is determined that an object that is not in a specific state has arrived, and if the high-rank judgment result appears before the timing ends. Alternatively, it may be determined that an object in a specific state has arrived. Thus, the second example of the determination processing ends.

FIG. 7A is a diagram illustrating an example of a signal value measured in the n-th cycle of the photoelectric sensor 10 according to the present embodiment. FIG. 7B is a diagram illustrating an example of a signal value measured in the (n + 1) th cycle of the photoelectric sensor 10 according to the present embodiment. 7A and 7B, the ordinate represents the value of the amount of received light, and the abscissa represents the time and the corresponding stage of the FIFO memory 13b. As shown in both figures, the latest value of the received light amount (value at time t9) is stored in the first stage q0 of the FIFO memory 13b, and the oldest received light amount value (time t0 value) is stored in the FIFO memory 13b. 13b is stored in the last stage q9. In this example, the FIFO memory 13b stores ten signal values in the order of acquisition.

The shape S1 of the object indicated by the broken line schematically shows the shape of the object in accordance with the timing at which the value of each received light amount is obtained. According to the shape S1 of the object, it can be read that the object has a shape with a projection on the base.

波形 A waveform W1 indicated by a solid line in FIG. 7A is a waveform obtained in the n-th cycle and configured by the signal values stored in the FIFO memory 13b. Further, a waveform W2 indicated by a solid line in FIG. 7B is a waveform obtained in the (n + 1) th cycle and configured by the signal values stored in the FIFO memory 13b. As shown in both figures, the signal value stored in the FIFO memory 13b in the n-th cycle is shifted to the next stage by one in the (n + 1) -th cycle and stored in the FIFO memory 13b.

Since the detection range 10a has a certain extent, near the timing corresponding to the step of the object 100, reflected light from both the upper surface and the lower surface of the step is received, and the signal values forming the waveforms W1 and W2 are It is an intermediate value. The value of the intermediate amount of received light is likely to fluctuate greatly due to a slight difference in the acquisition timing. Therefore, the value of the amount of received light may vary each time even for the target 100 having the same shape. In generating the judgment model, it is preferable that the object 100 be transported a certain number of times and the acquisition of the value of the amount of received light be repeated so that an averaging effect is obtained.

The operation control unit 13a, based on a signal value belonging to the fluctuation period, when a fluctuation period in which the fluctuation of the time-series signal value is relatively large follows a stable period in which the fluctuation of the time-series signal value is relatively small. May be used to generate a judgment model. In the case of the example of FIG. 7A, the stable period in which the fluctuation of the time-series signal value is relatively small is from time t0 to t1, and the fluctuation period in which the fluctuation of the time-series signal value is relatively large is from time t2 to t8. It is. In the case of the example of FIG. 7B, the stable period in which the fluctuation of the time-series signal value is relatively small is after time t8, and the fluctuation period in which the fluctuation of the time-series signal value is relatively large is from time t2 to t7. It is. The operation control unit 13a compares values stored in adjacent stages from the last stage to the first stage of the FIFO memory 13b, and when there is an adjacent stage whose difference is greater than or equal to the threshold, the operation control unit 13a It is determined that the signal value belonging to the fluctuation period from the stage closer to the first stage toward the first stage is stored, and the signal belonging to the stable period from the stage closer to the last stage of the adjacent stages toward the last stage. It may be determined that a value is stored. Specifically, in the case of the example of FIG. 7A, the operation control unit 13a compares the values stored in the final stage q9 and the eighth stage q8 and finds that the difference is 0, which is equal to or smaller than the threshold. The value stored in the stage q8 and the value stored in the seventh stage q7 may be compared, and the difference may be determined to be 2 and equal to or greater than the threshold. Here, the threshold value may be 1, for example. Then, the operation control unit 13a belongs to the fluctuation period from the seventh stage q7, which is closer to the first stage q0, to the first stage q0 of the eighth stage q8 and the seventh stage q7 in which the difference between the stored values is equal to or larger than the threshold. It is determined that the signal value is stored, and the signal value belonging to the stable period from the eighth stage q8, which is closer to the final stage, of the eighth stage q8 and the seventh stage q7 toward the final stage q9 is stored. May be determined.

The determination unit 13d is configured to perform the update of the FIFO memory 13b once or a plurality of times, once each time, the waveform W1 composed of a predetermined number of signal values stored in the FIFO memory 13b, and the specific The degree of coincidence with the reference waveform corresponding to the state may be determined by a determination model, and the state of the target object may be determined based on the rank of the degree of coincidence. For example, when the reference waveform is a waveform substantially equal to the shape S1 of the object shown by the broken line in FIG. 7A, the determination unit 13d compares the signal value stored in each stage of the FIFO memory 13b with the signal value of the reference waveform. The sum of the absolute values of the differences may be determined, and the smaller the value is, the higher the matching degree may be, and it may be determined whether the state of the target object is a specific state. As described above, when the conveyed object 100 is an object having a protrusion when the photoelectric sensor 10 is operating in the determination mode, the amount of received light in the specific shift cycle of the FIFO memory 13b is determined. The degree of coincidence between the waveform constituted by the value and the reference waveform constituted by the value of the amount of received light at the time of model generation becomes high.

FIG. 8 is a diagram illustrating another example of the signal value measured in the nth cycle of the photoelectric sensor 10 according to the present embodiment. In the figure, the vertical axis indicates the value of the amount of received light, and the horizontal axis indicates the time and the corresponding stage of the FIFO memory 13b.

波形 The waveform W3 indicated by the solid line in FIG. 8 is a waveform obtained in the nth cycle and configured by the signal values stored in the FIFO memory 13b. The shape S2 of the object indicated by the broken line schematically shows the shape of the object in accordance with the timing at which the value of each received light amount is obtained. According to the shape S2 of the object, it can be read that the object is a shape of only the base having no projection.

Comparing the waveform W3 with the waveform W1 shown in FIG. 7A, there is a difference in the value of the amount of received light between the times t4 and t6. If the object to be conveyed while operating in the determination mode is an object without a protrusion, the light reception acquired in several shift cycles of the FIFO memory 13b within the time required for the object to pass through. The degree of coincidence between the waveform composed of the value of the amount and the reference waveform composed of the value of the amount of received light at the time of generation of the judgment model is larger than the second predetermined value and is medium, but if it is higher than the predetermined value, Not high. Accordingly, the determining unit 13d may determine that an object having some commonality with the object having the protrusion has arrived, although the object is not the object having the protrusion, but the degree of coincidence may be medium. . In this example, the degree of coincidence was medium because the base portions of the objects are common.

FIG. 9 is a diagram illustrating another example of the configuration of the processing unit 13 of the photoelectric sensor 10 according to the present embodiment. The example of the configuration of the processing unit 13 shown in FIG. 3 is different from the example of the configuration of the processing unit 13 shown in FIG. 3 in that the reference value R is stored in the model storage unit 13c. The configuration is common.

The model storage unit 13c calculates a degree of coincidence from a difference between a predetermined number of signal values stored in the FIFO memory 13b and a reference value R corresponding to the predetermined number of signal values and representing a reference waveform, as a determination model. The model may be stored. In this example, the model for calculating the degree of coincidence calculates the sum of the absolute values of the differences between the five reference values R and the signal values stored in the corresponding stages q0 to q4 of the FIFO memory. Alternatively, a model that calculates the degree of coincidence such that the smaller the value is, the higher the degree of coincidence may be. Reference values r4, r3, r2, r1, and r0 are stored in the model storage unit 13c, and may correspond to values stored in q0, q2, q4, q6, and q8 of the FIFO memory 13b, respectively.

The determination model provided in the determination unit 13d determines the absolute value of the difference between the values in the corresponding relationship, determines that the sum of the absolute values of the differences is smaller than a first threshold as a first criterion, and matches when the first criterion is satisfied. A model that determines that the degree is a high rank may be used. The judgment model provided in the judgment unit 13d further uses the fact that the sum of the absolute values of the differences is between the first threshold and a second threshold larger than the first threshold as a second criterion. If the second criterion is satisfied but the first criterion is not satisfied within the required time range, the model may determine that the matching degree is the middle rank.

The judgment unit 13d may execute the judgment by the judgment model once every time the FIFO memory 13b is updated once or plural times, and may judge the state of the target object based on the rank of the matching degree. As described above, by using a relatively simple model to determine the degree of coincidence between the waveform composed of the signal values stored in the FIFO memory 13b and the reference waveform corresponding to the specific state of the target object 100, the transport is performed. It is possible to determine the state of the object 100 that is successively carried on the line at a higher speed.

The determination unit 13d determines the state of the target object 100 a plurality of times within a time period required for the target object 100 to pass through the detection range 10a, and determines a predetermined number of signal values stored in the FIFO memory 13b, When the first criterion for determining that the difference from the predetermined number of reference values R representing the reference waveform respectively corresponding to the signal values of the numbers is small is at least once, the state of the target object 100 is a specific state. It may be determined that there is. For example, the determination unit 13d corresponds to a predetermined number of signal values stored in the FIFO memory 13b and a predetermined number of signal values, respectively, within a time period required for the target object 100 to pass through the detection range 10a. When the first criterion for determining that the difference from the predetermined number of reference values R representing the reference waveform is small is at least once, it is determined that the target object 100 is a target object having a shape with a protrusion on the base. Good. As described above, within the time required for the object 100 to pass through the detection range 10a, the waveform constituted by the signal values stored in the FIFO memory 13b and the reference corresponding to the specific state of the object 100 It can be determined whether or not the degree of coincidence with the waveform is higher than a predetermined value.

Also, the determination unit 13d determines the state of the target object 100 a plurality of times within the time required for the target object 100 to pass through the detection range 10a, and the first criterion is not satisfied in all of the plurality of determinations. , For determining that the difference between the predetermined number of signal values stored in the FIFO memory 13b and the predetermined number of reference values R corresponding to the predetermined number of signal values and representing the reference waveform is moderate. When the two criteria are satisfied at least once, it may be determined that the target object 100 that is not in the specific state has arrived. For example, the determination unit 13d determines that a predetermined number of signal values stored in the FIFO memory 13b and a predetermined number of reference values R representing the reference waveform are within a range required for the object 100 to pass through the detection range 10a. Does not meet the first criterion for determining that the difference is small, but satisfies at least once the second criterion for determining that the difference is moderate, the object 100 has no protrusion and only the base May be determined. As described above, even when the state of the target object 100 is not the specific state, it is possible to determine that the target object 100 has arrived and that the state of the target object 100 is not the specific state.

FIG. 10 is a diagram illustrating an example in which a determination model is externally installed in the photoelectric sensor 10 according to the present embodiment. The example of the configuration of the processing unit 13 of the photoelectric sensor 10 illustrated in FIG. 3 is different from the example of the configuration of the processing unit 13 illustrated in FIG. 3 in that the operation control unit 13a outputs a time-series value of the received light amount to the outside. However, the difference is that a judgment model generated by an external computer is input based on that, and the input judgment model is stored in the model storage unit 13c, and the other configurations are common.

The operation control unit 13a may be capable of outputting a time-series signal value to the outside. The signal value output to the outside may be a signal value stored in the FIFO memory 13b. The signal value can be output to the outside, and the judgment model can be generated by the external device. This eliminates the need for the photoelectric sensor itself to have computational resources for the process of generating the determination model.

The operation control unit 13a may acquire the determination model from the outside and store it in the model storage unit 13c. The operation control unit 13a may obtain a determination model generated by an external computer, or may obtain a determination model generated by another photoelectric sensor. By diverting a determination model generated by another device, for example, another photoelectric sensor, generation of a determination model can be omitted.

The operation control unit 13a does not output a time-series signal value to the outside, and outputs an external model based on a model generated by another photoelectric sensor or a time-series signal value acquired by another photoelectric sensor. A computer-generated model may be input and used.

The embodiments described above are intended to facilitate understanding of the present invention, and are not intended to limit and interpret the present invention. The components included in the embodiment and their arrangement, material, condition, shape, size, and the like are not limited to those illustrated, but can be appropriately changed. It is also possible to partially replace or combine the configurations shown in the different embodiments.

[Appendix]
A light projecting unit (11) for emitting light toward a detection range (10a) where an object (100) arrives;
A light receiving unit (12) for acquiring a time-series signal value based on the light reception,
A FIFO memory (13b) that stores a predetermined number of the signal values in an order of acquisition and periodically updates the predetermined number of the signal values with the newly acquired signal values;
A determination model for determining a rank of the degree of coincidence between a waveform constituted by a predetermined number of the signal values stored in the FIFO memory (13b) and a reference waveform corresponding to a specific state of the object (100) is provided. A model storage unit (13c) for storing;
The determination by the determination model is performed once every time the FIFO memory (13b) is updated one or more times, and the state of the object (100) is determined based on the rank of the matching degree. A determining unit (13d) for performing
A photoelectric sensor (10) comprising:

Claims

A light-emitting unit that emits light toward a detection range where an object arrives,
A light receiving unit that acquires a time-series signal value based on the light reception of the light,
A FIFO memory that stores a predetermined number of the signal values in the order of acquisition and periodically updates the predetermined number of the signal values with the newly obtained signal values;
A model storage unit for storing a determination model for determining a rank of a degree of coincidence between a waveform composed of a predetermined number of the signal values stored in the FIFO memory and a reference waveform corresponding to a specific state of the object; ,
A determination unit that performs determination by the determination model once every time the FIFO memory is updated one or more times, and determines a state of the target object based on the rank of the matching degree;
A photoelectric sensor comprising:
The determination model is a learned model generated by machine learning,
The photoelectric sensor according to claim 1.
The determination model calculates the degree of coincidence from a difference between the predetermined number of signal values stored in the FIFO memory and a reference value representing the reference waveform corresponding to each of the predetermined number of signal values. Including the model,
The photoelectric sensor according to claim 1.
The determination model is a model for determining that the degree of coincidence is high when the degree of coincidence is higher than a predetermined value,
The determination unit, when the determination result of the high rank is obtained, determines that the state of the object is the specific state,
The photoelectric sensor according to claim 1.
The determination model is a model that further determines that the matching degree is a middle rank when the matching degree is not higher than the predetermined value but is higher than a second predetermined value smaller than the predetermined value,
The determination unit is configured such that, within the time required for the object to pass through the detection range, when there is no case of the high rank and there is a case of the middle rank, the object not in the specific state Is determined to have arrived,
The photoelectric sensor according to claim 4.
An operation control unit that generates the determination model based on the signal value, and stores the generated determination model in the model storage unit,
The photoelectric sensor according to claim 1.
The operation control unit, when following a stable period in which the fluctuation of the signal value in time series is relatively small, when a fluctuation period in which the fluctuation of the signal value in time series is relatively large appears, the operation belonging to the fluctuation period. Generating the determination model based on the signal value,
The photoelectric sensor according to claim 6.
The operation control unit can output the determination model to the outside,
The photoelectric sensor according to claim 6.
The operation control unit can output the signal value in time series to the outside,
The photoelectric sensor according to claim 6.
The apparatus further includes an operation control unit that acquires the determination model from the outside and stores the model in the model storage unit.
The photoelectric sensor according to claim 1.