Detection system

By designing a detection system in a robot system, using mobile units to perform multiple inspections and selecting the latest or most accurate data elements, the problem of invalidation when shooting data is used multiple times is solved, the detection accuracy and work accuracy are improved, and suitable for different Job purpose and object characteristics.

CN110871442BActive Publication Date: 2024-12-03FANUC LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN201910793986.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2018-08-31
Filing Date
2019-08-27
Publication Date
2024-12-03
Estimated Expiration
2039-08-27

AI Technical Summary

Technical Problem

In the robot system, when shooting data of the same object is used multiple times, the detection results are invalid, and due to distortion of the shooting device, lighting angle and handling device incomplete, the initial detection results are not accurate enough.

Method used

A detection system is designed, using a mobile unit to perform multiple detections, and the working data containing locations and indicators is generated through the working data production unit, and the working data storage unit selects the latest or most accurate data elements as the data to be saved, with priority Consider indicators such as the shape, quality and type of the object to improve detection accuracy and accuracy.

Benefits of technology

It realizes efficient and effective use of detection data, improves the accuracy and practicality of robot operations, and is suitable for different job purposes and object characteristics.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN110871442B_ABST
    Figure CN110871442B_ABST
Patent Text Reader

Abstract

The present invention provides a detection system, which comprises: a first detection device (10), which performs multiple detections on a moving object (O) within a predetermined detection range; a working data generating unit, which generates working data having a first data element and a second data element each time the object (O) is detected using the first detection device (10), wherein the first data element represents at least the position of the object (O) obtained using the first detection device (10), and the second data element at least includes an indicator about the object (O) obtained when the detection is performed; and a working data storing unit, which stores the working data generated by the working data generating unit, wherein the working data storing unit selects, based on the indicator, either the working data newly generated for the object (O) or the working data stored by the working data storing unit as the working data to be stored.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a detection system. Background Art

[0002] At present, there is known a robot system which includes a detection unit for detecting an object carried by a carrying device, and a robot, and the robot uses a tool provided at its front end to take out the object on the carrying device (for example, see Patent Document 1). In this robot system, if the same object is photographed twice or more by a photographing device, the detection results after the second time are invalidated.

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent Document 1: Japanese Patent Application Publication No. 2016-209995 Summary of the invention

[0006] Problem that the invention aims to solve

[0007] In the robot system, when the same object is imaged twice or more by the imaging device of the detection unit, the detection result obtained from the imaging data after the second time is not used.

[0008] However, due to the influence of the distortion of the lens of the imaging device, the angle of illumination of the object, and the imperfection of the transport device, the detection results initially obtained based on the imaging data may not be the most useful.

[0009] The present invention has been made in view of the above-mentioned circumstances. An object of the present invention is to provide a detection system that can efficiently and effectively use detection data of an object detected by a detection section, the object being moved by a moving unit.

[0010] Solutions for solving problems

[0011] In order to solve the above problems, the present invention adopts the following solutions.

[0012] The detection system of the first scheme of the present invention comprises: a detection unit, which performs multiple detections of an object moved by a moving unit within a detection range; a working data creation unit, which creates working data having a first data element and a second data element each time the detection of the object is performed by the detection unit, the first data element indicating at least the position of the object obtained by the detection unit, and the second data element at least including an indicator about the object obtained when performing the detection; and a working data storage unit, which stores the working data created by the working data creation unit, the working data storage unit selecting, based on the indicator, either the working data newly created for the object or the working data stored by the working data storage unit as the working data to be stored.

[0013] The detection system of the second scheme of the present invention comprises: a detection unit, which detects an object moved by a moving unit multiple times within a detection range; a working data creation unit, which creates working data having a first data element and a second data element each time the detection of the object is performed by the detection unit, the first data element indicating at least the position of the object obtained by the detection unit, the second data element at least including an indicator about the object obtained when performing the detection; and a working data storage unit, which stores the working data created by the working data creation unit, the working data storage unit selecting, based on the indicator, any one of the data elements of the working data newly created for the object and the data elements of the working data stored by the working data storage unit as the data element to be stored, the data element being any one of the first data element and the second data element.

[0014] For example, when there is a possibility that the object moves on the mobile unit, it is preferred to use the latest position of the object and perform the operation using a robot or the like. In addition, when it is desired to identify the shape of the object as accurately as possible in order to enable the robot or the like to perform the operation appropriately, it is preferred to reflect that the detection of the shape of the object is the most accurate detection among multiple detections of the object. In addition, when the quality, type, etc. of the object are important, it is preferred to reflect that the quality, type, etc. of the object is the most accurate detection among multiple detections of the object. Here, the detection accuracy, quality, and type of the shape of the object are examples of indicators about the object included in the second data element.

[0015] In the first embodiment, based on the index of the object included in the second data element, either the newly created work data for the object or the work data stored by the work data storage unit is selected as the work data to be stored. Since the work data is converted using the index of importance among the detection accuracy, quality, and type of the shape of the object, the detection data of the detection unit can be used efficiently and effectively.

[0016] In the second scheme, based on the index of the object included in the second data element, any one of the data element of the working data newly created for the object and the data element of the working data stored by the working data storage unit is selected as the data element to be stored. In addition, the data element is any one of the first data element and the second data element. Since the data element of the working data is converted using the index of importance among the detection accuracy, quality, and type of the shape of the object, the detection data of the detection unit can be used efficiently and effectively.

[0017] In the above aspect, preferably, the work data creating unit creates the work data having a work-related score as the indicator of the second data element, the work-related score being associated with the difficulty of the work on the object.

[0018] In order to properly perform work using a working machine such as a robot, generally speaking, it is preferred to perform work on an object with a low degree of difficulty. Using a work relevance score associated with the degree of difficulty of work on an object as an index is conducive to performing work on an object more properly.

[0019] In the above scheme, preferably, the work data producing unit produces the work data further having another score as the indicator of the second data element, and the other score is different from the job-related score.

[0020] According to the purpose of the work performed by the work machine such as a robot, it is preferred to flexibly convert the work data or its data elements. In this scheme, the work data has a score different from the work-related score as an indicator. Therefore, the work data or its data elements can be converted by preferentially using the score suitable for the purpose of the work.

[0021] In the above aspect, preferably, the work data creating unit creates the work data having a quality relevance score as the indicator of the second data element, the quality relevance score being associated with at least one of the quality and the type of the object.

[0022] It is preferred to flexibly convert work data or its data elements according to the purpose of the work performed by a work machine such as a robot. In this scheme, the work data has a quality-related score associated with at least one of the quality and type of the object, and the quality-related score is a score different from the work-related score. Therefore, the work data or its data elements can be converted preferentially using a score suitable for the purpose of the work.

[0023] In the above solution, preferably, the work data saving unit selects the work data to be saved based on the job-related score and any one of the other scores or the quality-related score with a higher priority.

[0024] By changing the priority score, the operation of a working machine such as a robot can be made simple and efficient to suit the purpose of the work.

[0025] In the above scheme, preferably, when the index of the newly produced working data is the same as the index of the working data saved by the working data holding unit, the working data holding unit selects the newly produced working data as the working data to be saved.

[0026] In this case, when the newly created work data is saved, even if the index of the object does not change, the position data of the first data element in the work data becomes the latest. For example, in the case where there is a possibility that the object moves on the mobile unit, this structure can use the latest position data, which is conducive to improving the accuracy and reliability of the work.

[0027] In the above scheme, preferably, the job-related score is associated with at least one of the following: the contrast of the object in the image obtained by the shooting device of the detection unit, the distortion of the object in the image, and the position of the object in a direction orthogonal to the moving direction of the moving unit.

[0028] When a robot or other working machine is used to work on an object, if the contrast of the object in the image obtained by the camera is low or the distortion of the object is large, it may have an adverse effect on the work performed by the working machine. In addition, when the distance from the working machine to the object in the orthogonal direction is appropriate, it is easy to work with the working machine. Therefore, this structure is conducive to improving the accuracy and reliability of the work.

[0029] Effects of the Invention

[0030] According to the present invention, detection data of an object detected by a detection unit, the object being moved by a moving unit, can be used efficiently and effectively. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a diagram showing the configuration of a robot system using the detection system according to the first embodiment of the present invention.

[0032] Figure 2 It is a block diagram of a control unit of the detection system according to the first embodiment.

[0033] Figure 3 The following is a diagram for explaining the behavior of an object detected by the detection system according to the first embodiment.

[0034] Description of Reference Numerals

[0035] 2: Transport device

[0036] 2a: Encoder

[0037] 2b: Motor

[0038] 10: First detection device (detection unit)

[0039] 10a: Shooting device

[0040] 10b: Lighting

[0041] 11: Processor

[0042] 12: RAM

[0043] 13: Memory

[0044] 13a: Image Processing Program

[0045] 20: Second detection device (detection unit)

[0046] 20a: Shooting device

[0047] 30b: Ultraviolet lighting device

[0048] 21: Processor

[0049] 22: RAM

[0050] 23: Memory

[0051] 23a: Image Processing Programs

[0052] 30: Robot

[0053] 30a: Arm

[0054] 31: Servo motor

[0055] 40: Control unit

[0056] 41: Processor

[0057] 42: Display device

[0058] 43: Storage

[0059] 43a: System Programs

[0060] 43b: Work data creation program (work data creation unit)

[0061] 43c: Work data storage program (work data storage unit)

[0062] 44: Servo controller

[0063] 45: Input

[0064] 201: Reference coordinate system

[0065] O: Object DETAILED DESCRIPTION

[0066] A detection system according to a first embodiment of the present invention will be described below with reference to the drawings.

[0067] In one example, the detection system of the first embodiment is used to Figure 1 The robot system shown in FIG. 1 is provided with a first detection device (detection unit) 10, a second detection device (detection unit) 20, and a control unit 40 for controlling a robot 30. The robot (operating machine) 30 performs a predetermined operation on an object that is an article moved by a transport device (moving unit) 2. A tool T is installed at the front end of the robot 30. As an example, the robot 30 uses the tool T to remove the object O on the transport device 2. In the present embodiment, the transport direction of the transport device 2 is the same as that of the conveying device 2. Figure 1 The X-axis direction of the reference coordinate system 201 is consistent with that of Figure 1 The Z-axis direction of the reference coordinate system 201 is consistent, and Figure 1 The Y-axis direction of the reference coordinate system 201 coincides with the width direction of the conveying device 2. The control unit 40 may be composed of one control device or a plurality of control devices.

[0068] The robot 30 is not limited to a specific type of robot. The robot 30 of the present embodiment is provided with a plurality of servo motors 31 (see Figure 2 ) is a vertical multi-joint robot. In addition, the arm 30a of the robot 30 is composed of a plurality of movable parts. Each servo motor 31 has a working position detection device for detecting its working position, and as an example, the working position detection device is an encoder. The detection value of the working position detection device is sent to the control unit 40. The robot 30 can also be a horizontal end joint robot, a multi-link robot, etc.

[0069] In an example, Figure 2As shown, the control unit 40 includes: a processor 41 such as a CPU; a display device 42; a storage unit 43 having a non-volatile memory, a ROM, and a RAM; a plurality of servo controllers 44 corresponding to the servo motors 31 of the robot 30; and an input unit 45 such as an operating panel.

[0070] The storage unit 43 stores a system program 43a, which assumes the basic functions of the control unit 40. In addition, a work data creation program (work data creation unit) 43b and a work data storage program (work data storage unit) 43c are stored in the storage unit 43. In addition, an action program and a tracking control program are also stored in the storage unit 43. The processor 41 controls the servomotors 31 and the tool T of the robot 30 based on the programs, thereby performing a removal operation for the object O transported by the transport device 2.

[0071] Furthermore, the first detection device 10 and the second detection device 20 perform the following detection at the working position of the robot 30 or at a position upstream of the working device of the robot 30 in the conveying direction.

[0072] The first detection device 10 has a camera 10a, which is a two-dimensional camera, a three-dimensional camera, or the like. Instead of the camera 10a, a three-dimensional distance sensor or the like may be used, or other sensors that can obtain data for detecting the position and posture of the object O may be used. The camera 10a of this embodiment is a two-dimensional camera, and the camera 10a is supported above the conveying device 2 by a frame not shown in the figure. In order to illuminate the shooting position, for example, the first detection device 10 has a lighting device 10b that irradiates visible light, and the lighting device 10b irradiates light only when shooting is performed by the camera 10a.

[0073] The first detection device 10 includes a processor 11, a RAM 12, and a memory 13 storing an image processing program 13a. The processor 11 performs, for example, spot processing, pattern matching processing, and posture detection processing based on the position of feature points based on the image processing program 13a. The detection data (first data element) obtained by the first detection device 10 is sent to the control unit 40. The detection data includes at least the position data of each object O, and may also include the posture data of each object O.

[0074] The first detection device 10 performs a work-related data acquisition process based on the image processing program 13a, for example, to obtain the degree of fit (work-related score) between the shape of the predetermined part measured by the pattern matching process and the predetermined model shape. If the predetermined part is a part to be grasped by the tool T, the difficulty of grasping by the tool T varies depending on its shape. Therefore, the score for each object O obtained by the work-related data acquisition process represents the difficulty of the work on each object. The work-related data (second data element) including the work-related score (index) obtained by the first detection device 10 is sent to the control unit 40.

[0075] The second detection device 20 has a photographing device 20a, which is used to detect at least one of the quality (quality), type, and inspection result (quality) of the object O. In the present embodiment, the photographing device 20a is a two-dimensional camera, and has an illumination device 20b that irradiates ultraviolet light to the photographing position. The illumination device 20b irradiates ultraviolet light only when photographing is performed using the photographing device 20a. The photographing device 20a can photograph fluorescence of a predetermined wavelength generated from the object O irradiated with ultraviolet light using the illumination device 20b. In addition, the illumination device 20b can also irradiate infrared light or visible light.

[0076] The second detection device 20 includes a processor 21, a RAM 22, and a memory 23 storing an image processing program 23a. The processor 21 performs a quality-related data acquisition process to obtain the area (quality-related score) of the portion where fluorescence of a predetermined wavelength is generated in the object O when the object O is detected, using image processing such as pattern matching processing and binarization processing based on the image processing program 23a. In the case where the object O is fresh food, the wavelength and intensity of the fluorescence vary depending on its freshness, the presence or absence of corrosion, etc. Therefore, the area obtained for each object O by the quality-related data acquisition process represents the quality of each object O. The quality-related data (second data element) including the quality-related score (index) obtained by the second detection device 20 is sent to the control unit 40. The wavelength and intensity of the fluorescence vary depending on the material, so the processor 21 can also obtain the type of the object O based on the wavelength of the fluorescence from the object O by the quality-related data acquisition process.

[0077] The conveying device 2 includes an encoder 2 a that can detect the amount of movement of the object O. The encoder 2 a is provided in a motor 2 b that drives the conveying device 2 , for example.

[0078] In one example, whenever the encoder 2a detects a predetermined amount of movement, the first detection device 10 performs position detection processing, posture detection processing, and work-related data acquisition processing of the object O, and the second detection device 20 performs quality-related data acquisition processing just before, just after, or simultaneously with the position detection processing and posture detection processing performed by the first detection device 10. Alternatively, the result of the first detection device 10 can be associated with the result of the second detection device 20, and the second detection device 20 can perform detection at any time.

[0079] Furthermore, the amount of movement may be detected by the first detection device instead of using an encoder.

[0080] The control unit 40 receives the detection data and operation-related data of each object O from the first detection device 10, and receives the quality-related data of each object O from the second detection device 20 in a manner corresponding to the detection data and operation-related data of each object O. For example, each quality-related data includes approximate position data of the corresponding object O. Therefore, the control unit 40 can associate the detection data and operation-related data with the quality-related data.

[0081] The processor 41 of the control unit 40 generates work data having each detection data (first data element), corresponding operation-related data (second data element) and quality-related data (second data element) for each object O based on the work data generation program 43b. In addition, the processor 41 saves the generated work data in the storage unit 43 based on the work data storage program 43c. In one example, the work data is stored in a work queue, and a plurality of work data are arranged in the work queue based on their position data. More specifically, in the work queue, a plurality of work data are arranged in sequence from the downstream side of the conveying direction of the conveying device 2.

[0082] The control unit 40 receives detection data and operation-related data from the first detection device 10 , and receives operation-related data and quality-related data from the second detection device 20 , and generates working data having the received detection data, corresponding operation-related data, and quality-related data using the processor 41 .

[0083] The processor 41 determines whether the work data for the object O identical to the newly created work data already exists in the work queue. Whether the object O is the same is determined by a known method using the position data of each work data, the movement amount of the object O obtained from the encoder 2a, etc. For example, objects determined to be objects O whose distances are less than a certain threshold are determined to be identical. Furthermore, if the work data for the same object O already exists, the processor 41 selects either the newly created work data or the already saved work data as the work data to be saved based on the work data saving program 23c, and saves the selected work data in the work queue.

[0084] Figure 3 Indicates a situation where the same object O is detected three times within a predetermined detection range in the field of view of the first detection device 10, the first detection is at one end side of the conveying direction (X-axis direction) in the detection range, the second detection is at the center side of the conveying direction in the detection range, and the third detection is at the other end side of the conveying direction in the detection range. In this case, due to the influence of the distortion of the lens of the first detection device 10, etc., there is a tendency that the operation correlation score on the center side of the detection range is better than that on one end side and the other end side, and there is a tendency that the area of ​​the part of the fluorescence generated as the mass correlation score of the object on the center side of the detection range is larger than that on one end side and the other end side. On the other hand, the other end side of the detection range is closer to the operation area where the robot 30 is used for operation than the center side. When there is a possibility that the object O moves on the conveying device 2, it is preferred to use the detection data obtained in the third detection to improve the accuracy of the operation performed by the robot 30.

[0085] The selection criteria described above vary depending on various situations. Examples of the selection criteria are shown below.

[0086] Selection based on assignment-related scores

[0087] When work data for the object O identical to the newly created work data already exists in the work queue, the processor 41 compares the degree of fit (job association score) of both parties and keeps (saves) the party with a higher degree of fit in the work queue. A higher degree of fit means that it is easier to hold the object O using the tool T, and the control unit 40 can more accurately grasp the shape and position of the object O. In addition, the party with a lower degree of fit can be kept in the work queue according to the purpose. For example, the score of the quality data is prioritized.

[0088] On the other hand, when the matching degree of both parties is the same, the work data already saved in the work queue is left (saved) in the work queue. At this time, the newly created work data can also be left (saved) in the work queue. There is a case where it is advantageous to leave the newly created work data in the work queue when the position of the item (object O) may change.

[0089] Selection based on quality correlation score

[0090] When the work data for the object O identical to the newly created work data already exists in the work queue, the processor 41 compares the areas (quality correlation scores) of the parts where fluorescence occurs on both sides, and retains (saves) the one with the larger area in the work queue. The larger area means: the area of ​​the part where the quality has decreased in the object O, for example. In this case, the object O with lower quality is taken out by the robot 30 first. When the same object (object O) is adjusted and the lighting is changed and several inspections are performed, as long as the quality decreases in one time, the part with its quality data is left in the work queue, thereby improving the quality of the inspected items. In addition, it is also possible to leave the one with a smaller area (the one with better quality) in the work queue according to the purpose. In this case, the object O with the best quality is given priority.

[0091] On the other hand, when the areas of both parties are the same, the work data already stored in the work queue is left (stored) in the work queue. At this time, the work data newly made can also be left (stored) in the work queue.

[0092] Selection based on job-related scores and quality-related scores

[0093] When the work data for the same object O as the newly created work data already exists in the work queue, the processor 41 compares the areas (mass correlation scores) of the parts where fluorescence occurs on both sides, and keeps (stores) the one with the larger area in the work queue. In addition, the one with the smaller area can also be kept in the work queue according to the purpose.

[0094] At this time, if the areas of both are the same, the degrees of agreement (task correlation scores) of both are compared, and the one with higher agreement is retained (stored) in the work queue. Alternatively, the one with lower agreement can be retained in the work queue according to the purpose.

[0095] On the other hand, when the areas and the matching degree of both parties are the same, the work data already stored in the work queue is left (stored) in the work queue. At this time, the newly created work data can also be left (stored) in the work queue.

[0096] Furthermore, it is also possible to perform the same processing as described above using a plurality of types of work-related scores and a plurality of types of quality-related scores.

[0097] A robot system according to a second embodiment of the present invention will be described below.

[0098] In the detection system of the second embodiment, it is not necessary to select either the already saved work data or the new work data as in the first embodiment, but rather to select either the data element of the already saved work data or the data element of the new work data and save the selected data element in the work queue.

[0099] For example, if there is a possibility that the object O moves on the transport device 2, Figure 3 In the above, it is preferable to use the detection data (position data) of the first data element of the object O obtained by the third detection. Furthermore, when it is desired to reliably remove the object O with reduced quality by using the robot 20, it is desired to obtain the area (quality correlation fraction) of the portion where fluorescence occurs in the object O as accurately as possible. However, due to the distortion of the lens of the second detection device 20, etc., there is a tendency that the area of ​​the portion where fluorescence occurs becomes more accurate in the second detection.

[0100] In order to further improve the operation of the robot 30 in this situation, in one example, when the work data for the object O identical to the newly created work data is already stored in the work queue, the processor 41 replaces the detection data (position data) of the first data element of the work data already in the work queue with the detection data of the newly created work data. Thus, when there is a possibility that the object O moves on the transport device 2, the robot 30 performs the removal operation of taking out the object O using the latest position data.

[0101] At the same time, when the work data for the same object O as the newly created work data already exists in the work queue, the processor 41 compares the areas (quality correlation scores) included in the quality-related data (second data element) and retains (stores) the quality-related data with a larger area in the work queue. That is, when the area of ​​the quality-related data included in the newly created work data is larger, the quality-related data of the work data already existing in the work queue is replaced with the quality-related data of the newly created work data.

[0102] In the second embodiment, when the same object O is photographed twice by the first detection device 10 or the second detection device 20, the first lighting and the second lighting can be changed. For example, the first lighting can be set to lighting suitable for position detection, and the second lighting can be set to lighting suitable for accurate acquisition of work-related scores, quality-related scores, etc.

[0103] Furthermore, even when selecting and replacing data elements in the operation data of the second embodiment, the same selection criteria as those of the first embodiment can be used.

[0104] Furthermore, in each of the above-described embodiments, the work-related score of each object O can be obtained based on the Y-axis position of each object O. For example, it is often difficult to work at a position farther from the robot 10 in the Y-axis direction, and it is often easy to work at a position closer to the robot 10. Therefore, if the work-related score is determined based on the Y-axis position data of each object O, the accuracy and reliability of the work can be improved.

[0105] In addition, in each of the above-described embodiments, it is also possible to obtain a work-related score based on the posture of each object O. Due to the shape of the object O, the shape of the tool T of the robot 30, etc., there may be a posture of the object O that is difficult to work with the robot 30. In this case, if the work data or its data elements are converted based on the work-related score according to the posture of the object O, the accuracy and reliability of the work performed with the robot 30 can be improved.

[0106] In addition, in each of the above embodiments, the first detection device 10 can also obtain the degree of fit (quality correlation score) between the shape of the detected object O and the predetermined model shape. For example, when the object O is fresh food, the shape of the object O changes depending on its freshness, the presence or absence of corrosion, etc. In addition, the first detection device 10 can also perform quality inspections for defects, cracks, etc. In this case, even if the second detection device 20 is not provided, the quality correlation score can be obtained.

[0107] In the above-described embodiments, instead of the encoder 2a, a camera, a three-dimensional measuring instrument, a three-dimensional distance sensor, etc. can be used to detect the movement amount of the object O. For example, by detecting a mark on the conveying device 10, the object O, etc., using a camera, the movement amount of the object O can be detected. The movement amount can also be detected using other means.

[0108] In addition, in each of the above embodiments, as a substitute for the transport device 2, the object O may be moved by other robots or AGV (Automated Guided Vehicle). Even in this case, the same effects as described above can be achieved. Furthermore, in the case where the object O is a car, a frame of a car, etc., the object O that performs the predetermined operation may be moved by its engine, wheels, etc. In these cases, other robots, engines, wheels, etc. function as a moving unit.

[0109] In addition, when a frame of a car or the like is moved by a moving unit, holes provided in the frame, welding points in the frame, and the like may be the object O. In this case, the position data of the object O, the degree of conformity with the model shape (work-related score), the roundness of the hole, the color of the welding point, and the quality-related score such as the presence or absence of defects around the hole are transmitted to the control unit 40 by the first detection device 10 .

[0110] Instead of the transport device 2, a chute that uses gravity to make the object O slide, roll, or fall can be used to move the object O. In this case, the chute is vibrated by a vibration device, thereby making it possible to smoothly move the object O on the chute. In these cases, the chute, the vibration device, etc. function as a moving unit, and the object O moved by the chute is taken out by the tool T mounted on the robot 30.

[0111] Furthermore, in the above-described embodiments, when another robot exists on the downstream side, the robot 30 does not need to forcibly take out all the objects O on the transport device 2. In this case, the first data element of each operation data may also be set for each robot.

[0112] In addition, in each of the above-mentioned embodiments, the control unit 40 or other computers may be used to implement the functions of the image processing program 13a based on the first detection device 10 and the functions of the image processing program 23a based on the second detection device 20.

[0113] In the first embodiment, based on the index about the object O included in the second data element, either the newly created work data for the object O or the work data saved by the work data saving unit is selected as the work data to be saved. Since the work data is converted using the index of importance among the detection accuracy, quality, type, etc. of the shape of the object O, the detection data of the first detection device 10 and the second detection device 20 can be used efficiently and effectively.

[0114] In the second embodiment, based on the index about the object O included in the second data element, one of the data element of the newly created work data for the object O and the data element of the work data stored by the work data storage unit is selected as the data element to be stored. In addition, the data element is either the first data element or the second data element. Since the data element of the work data is converted using the index of importance among the detection accuracy, quality, type, etc. of the shape of the object O, the detection data of the detection unit can be used efficiently and effectively.

[0115] Furthermore, in each of the above-described embodiments, the processor 41 of the control unit 40 creates work data having a task-related score associated with the difficulty of the task for the object O as an indicator of the second data element.

[0116] In order to properly perform work using a working machine such as the robot 30, it is generally preferred that the difficulty of the work on the object O be lower. Using the work relevance score associated with the difficulty of the work on the object O as an index is conducive to performing the work on the object O more properly.

[0117] Furthermore, in each of the above-described embodiments, the processor 41 of the control unit 40 creates work data that also has another score different from the task-related score as an index of the second data element.

[0118] It is preferred to flexibly convert the work data or its data elements according to the purpose of the work performed by the work machine such as the robot 30. In this structure, the work data has a score different from the work-related score as an indicator. Therefore, the work data or its data elements can be converted by preferentially using a score suitable for the purpose of the work.

[0119] Furthermore, in each of the above-described embodiments, the processor 41 of the control unit 40 creates work data having a quality relevance score associated with at least one of the quality and the type of the object O as an index of the second data element.

[0120] It is preferable to flexibly convert the work data or its data elements according to the purpose of the work performed by the work machine such as the robot 30. In this structure, the work data has a quality-related score associated with at least one of the quality and type of the object, and the quality-related score is a score different from the work-related score. Therefore, the work data or its data elements can be converted by preferentially using a score suitable for the purpose of the work.

[0121] In addition, in each of the above-mentioned embodiments, the processor 41 of the control unit 40 selects the work data to be saved based on either the task-related score or the other score or the quality-related score.

[0122] By changing the priority score, the operation of the working machine such as the robot 30 can be made simple and efficient to suit the purpose of the work.

[0123] In the above-described embodiments, when the index of the newly created working data is the same as the index of the working data stored in the working data storage unit, the processor 41 of the control unit 40 selects the newly created working data as the working data to be stored.

[0124] In this case, even if the newly created work data is saved, the index of the object does not change, but the position data of the first data element in the work data becomes the latest. For example, in the case where there is a possibility that the object O moves on the conveying device 2, this structure that can use the latest position data is conducive to improving the accuracy and reliability of the work.

[0125] In addition, in each of the above embodiments, the job-related score can also be associated with at least one of the following: the contrast of the object O in the image obtained by the shooting device 10a of the first detection device 10, the distortion of the object O in the image, and the position of the object O in the direction (Y-axis direction) orthogonal to the moving direction of the conveying device 2.

[0126] When a working machine such as the robot 30 performs work on the object O, there is a case where the contrast of the object O in the image obtained by the imaging device 10a is low or the distortion of the object O is large, which may have an adverse effect on the work performed by the working machine. In addition, when the distance from the working machine to the object O in the Y-axis direction is appropriate, it is easy to perform work by the working machine. Therefore, this structure is conducive to improving the accuracy and reliability of the work.

Claims

1. A detection system, characterized in that: have: a detection unit that detects an object moved by a moving unit multiple times within a detection range; a work data creating unit that creates work data having a first data element indicating at least a position of the object obtained by the detection unit and a second data element including at least an indicator about the object obtained when the detection is performed, each time the detection unit performs the detection of the object; and a working data storage unit for storing the working data produced by the working data producing unit, The working data storage unit selects, based on the indicator, either the working data newly created for the object or the working data stored by the working data storage unit as the working data to be stored, The work data creating unit creates the work data having a work relevance score as the index of the second data element, the work relevance score being associated with a degree of difficulty of the work on the object.

2. The detection system according to claim 1, characterized in that: The work data creating unit creates the work data further having another score as the indicator of the second data element, the other score being different from the task-related score.

3. The detection system according to claim 1, characterized in that: The work data creation unit creates the work data having a quality relevance score as the indicator of the second data element, the quality relevance score being associated with at least one of a quality and a type of the object.

4. The detection system according to claim 2, characterized in that: The work data creation unit creates the work data having a quality relevance score as the indicator of the second data element, the quality relevance score being associated with at least one of the quality and the type of the object, The work data saving unit selects the work data to be saved based on the task-related score and any one of the other scores or the quality-related score having a higher priority.

5. The detection system according to claim 1, characterized in that: When the index of the newly created working data is identical to the index of the working data stored in the working data storage unit, the working data storage unit selects the newly created working data as the working data to be stored.

6. The detection system according to claim 1, characterized in that: The work-related score is associated with at least one of the contrast of the object in the image obtained by the imaging device of the detection unit, the distortion of the object in the image, and the position of the object in a direction orthogonal to a moving direction of the moving unit.

7. A detection system, characterized in that: have: a detection unit that detects an object moved by a moving unit multiple times within a detection range; a work data creating unit that creates work data having a first data element indicating at least a position of the object obtained by the detection unit and a second data element including at least an indicator about the object obtained when the detection is performed, each time the detection unit performs the detection of the object; and a working data storage unit for storing the working data produced by the working data producing unit, The working data storage unit selects, based on the indicator, any one of the data element of the working data newly created for the object and the data element of the working data stored by the working data storage unit as the data element to be stored, The data element is any one of the first data element and the second data element, The work data creating unit creates the work data having a work relevance score as the index of the second data element, the work relevance score being associated with a degree of difficulty of the work on the object.

8. The detection system according to claim 7, characterized in that: The work data creating unit creates the work data further having another score as the indicator of the second data element, the other score being different from the task-related score.

9. The detection system according to claim 7, characterized in that: The work data creation unit creates the work data having a quality relevance score as the indicator of the second data element, the quality relevance score being associated with at least one of a quality and a type of the object.

10. The detection system according to claim 8, characterized in that: The work data creation unit creates the work data having a quality relevance score as the indicator of the second data element, the quality relevance score being associated with at least one of the quality and the type of the object, The work data saving unit selects the work data to be saved based on the task-related score and any one of the other scores or the quality-related score having a higher priority.

11. The detection system according to claim 7, characterized in that: The work-related score is associated with at least one of the contrast of the object in the image obtained by the imaging device of the detection unit, the distortion of the object in the image, and the position of the object in a direction orthogonal to a moving direction of the moving unit.