JP2555823B2 - High-speed picking device for piled parts - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a high-speed picking device capable of gripping a stack of parts stored in a tray by a robot one by one.

[0002]

2. Description of the Related Art Conventionally, as a means for picking one part from a plurality of parts one by one, grayscale image data is generated from a video signal captured by an image input camera, and a ridge line of differentiated edge image data is traced. Then, the contour line is extracted to obtain a line segment image. A method is known in which pattern matching is performed between the line segment image and a matching model corresponding to the shape of the gripped component, and the component located at the top is recognized and picked up.

[0003]

By the way, parts and the like in general industrial products are usually supplied in a pile in a tray or the like. Then, these parts overlap each other and exist in a state in which each part has a sharp inclination. It is extremely rare that the parts located at the top of the pile match the matching model when there is no inclination, and it is almost impossible to recognize and pick one part from the pile parts. There was a problem.

The present invention has been made to solve the above problems, and an object of the present invention is to select a specific portion consisting of a simple shape of a part that can be gripped by a robot hand from a pile of parts. Is to recognize at high speed.

[0005]

The constitution of the invention for solving the above-mentioned problems is, as the concept is shown in FIG. 9, obtaining a contour line of a piled-up component from a two-dimensional image, and calculating the contour line from the contour line. In a high-speed picking device that extracts a plurality of constituent line segments that make up the object, recognizes the component from the constituent line segments, and grips it with a robot hand, A model corresponding to the specific part for recognizing, and a matching model storing means for storing a matching model preset by data specifying a shape when the specific part assumes a reference posture; It is preset based on an allowable limit angle of inclination of the specific portion with respect to the reference posture, which represents a grippable range of the specific portion. Matching limit value storage means for storing a matching limit value indicating a range of allowable mismatch with the matching model and a range of mismatch defined by the matching limit value with respect to the matching model from the two-dimensional image. Specific part detecting means for detecting one part recognized in the body as the specific part, position determining means for determining the position of the detected specific part, and positioning the hand at the position to identify the specific part. It is characterized in that it is provided with command means for picking up.

[0006]

In the collation model storage means, a model corresponding to the specific part for recognizing a simple shape of a part that can be grasped in a two-dimensional image is provided. A collation model preset by the data for specifying the shape is stored. Further, the matching limit value storage means is preset based on the allowable limit angle of inclination of the specific part with respect to the reference posture by the hand of the robot, and shows the range of allowable mismatch between the specific part and the matching model. The matching limit value is stored. Further, the specific portion detecting means detects, as the specific portion, one portion recognized from the two-dimensional image within the non-coincidence range defined by the collation limit value with respect to the collation model. Then, the position of the specific portion detected by the position determining means is determined. Further, a command for positioning the hand at the position determined by the position determining device by the commanding device and picking up the specific portion is output to the robot side.

[0007]

EXAMPLES The present invention will be described below based on specific examples. FIG. 1 is an overall configuration diagram showing a high-speed picking device for piled parts according to the present invention, and FIG. 2 is a block diagram showing a configuration of a main part of the device of the same embodiment.
The high-speed picking device 100 mainly has an image input camera 10, an object recognizing device 20, and a hand 40 for gripping one component (hereinafter, also referred to as a work) W from among piled components with a finger arranged at the tip. A picking robot 30. Then, on the workbench, there are placed trays T that accommodate the works W in a piled state.

In FIG. 2, the trays T store the works W in a piled state, and an image input camera 10 for picking up the works W from the upper part of the tray T is provided. Further, an illuminating device L for uniformly illuminating the work W from the upper center of the tray T is provided. Object recognition device 20
Is a central processing unit 21 that performs data processing such as collation and determination.
And data for processing the video signal obtained by the image input camera 10 to detect the contour line of the detected object, extracting the constituent line segments forming the contour line, and obtaining a composite edge image. An image processing device 22 that performs processing, a storage device 23 that stores data regarding a matching model and data regarding a detected object, and an illumination control circuit 24. Further, the image processing device 22 samples the video signal output from the image input camera 10 to generate the grayscale image data in which the grayscale level is digitized to generate the grayscale image data.
1 and an edge detection device 222 that obtains edge image data representing an edge of an object image by obtaining a lightness gradient by differential calculation from the grayscale image data, and a contour line is traced from the edge image data, and the contour line is configured. And a line segment extraction device 223 that generates data regarding the position of the component line segment. Further, the storage device 23 is composed of a RAM or the like, stores a collation model preset by data that specifies the shape when the specific part of the work W takes the reference posture, and achieves the collation model storage means. A model memory area 231, a matching limit value memory area 232 for storing a matching limit value indicating a range of allowable mismatch between a specific part and a matching model, and achieving a matching limit value storage means, and a stack in the tray T. The recognition result memory area 233 stores the result of matching of the line segment images corresponding to a large number of workpieces W.

Next, after inputting video signals of a large number of workpieces W in the piled state by the image input camera 10 and extracting the constituent line segments, a circle, for example, a collation model, is selected as a specific portion from the various constituent line segment groups. The operation of the present device will be described based on the flowchart of FIG. 3 showing the processing procedure of the object recognition device 20 in the case of performing collation selection by "circle". The illumination control circuit 24 turns on the illumination device L, and the video signal obtained by the image input camera 10 is input to the image input device 221. Then, in the image input device 221, the video signal is sampled and converted into a digital signal to generate a grayscale image. The grayscale image data is input to the edge detection device 222 and differentiated to generate an edge image.
The edge image data is input to the line segment extraction device 223,
The contour line of the object is extracted by tracing the edge line. Further, the contour line is approximated by a polygonal line or a circle to obtain a line segment image. Then, in step 100, the central processing unit 1 inputs the line segment image obtained by the image processing unit 22. Next, in step 102, a series of line segment groups from the input line segment image is stored in the storage area A in the central processing unit 1.
To be stored. Next, in step 104, it is determined whether or not the storage area A is empty, that is, whether or not the series of line segments that are the object of matching remain in the storage area A stored in step 102. In the first execution cycle, as long as the series of line segment groups have been successfully extracted, the storage area A
Is not empty, the process proceeds to step 106. In step 106, n sets of two line segments in a series of line segment groups in the storage area A are selected, and an intersection G of their vertical bisectors is selected.
Calculate _i (i = 1,2, ..., n). FIG. 4 shows a method of calculating the center of a perfect circle or an ellipse. As shown in Fig. 4 (a), if a series of line segments is close to a perfect circle, n sets of two line segments (l ₁ and l _2, l ₃ and l _4, l ₅ and l _6, ...) The coordinates of the intersections of the perpendicular bisectors of are substantially coincident and become G ₀ . However, as shown in FIG. 4 (b), if the series of line segments is an ellipse, n sets of two line segments (l ₁ and l _2, l ₃ and l _4, l ₅ and l _6, ... The coordinates of the intersection of the vertical bisectors of) do not match and become G _1, G _2, G _3, ... _, G _n .
Next, in step 108, the maximum value of the distance between any two points at the intersection point G _i (i = 1,2, ..., n) obtained in step 106 is the first matching limit of the matching model. It is determined whether or not it is within a predetermined allowable range ρ ₁ which is a value. That is, it is determined whether or not the continuous line segment group is originally a circle. That is, the allowable range ρ ₁ is a portion shown in a hatched circle in FIG. 4B, and the circular hole portion which is the specific portion of the work W is inclined to be an ellipse. The collation limit value that can be collated corresponding to the collation model in each case is shown. At step 108, if all the intersections G _i (i = 1,2, ..., n) are within the predetermined allowable range ρ ₁ , the process proceeds to step 110 and the intersections G _i (i = 1,2, ..., n). Average of (G _1, G _2, … _, G _n )
/ N is calculated to be the central coordinate G _m . Next Step 11
2, the central coordinate G obtained in step 110
The average of the distances between _m and each line segment selected in step 106 is calculated to obtain the radius r _m . Next, the routine proceeds to step 114, where the deviation | r _{m with} respect to the radius r ₀ of the circle of the matching model stored in the matching model memory area 231 of the storage device 23.
It is determined whether -r ₀ | is within the predetermined allowable range ρ ₂ . Here, FIG. 5 shows a work posture that determines picking availability. That is, the work W having the work tilt angle of θ ₂ in FIG. 5B cannot be picked by the hand 40 of the picking robot 30, and the work W having the work tilt angle within θ ₁ of FIG. It indicates that picking is possible. This work inclination angle θ ₁ is an allowable limit value for a specific part of the work W. The second matching limit value of the matching model based on the allowable limit value θ ₁ of the tilt that can be gripped by the circular hole portion that is the specific portion of the work W is the allowable range ρ ₂ . The allowable limit value θ ₁ is a slightly smaller angle in consideration of the safety factor in the work inclination angle θ at which the picking success probability is 100%. Here, as shown in FIG. 6, the picking success probability of a specific portion of the work with respect to the work inclination angle θ is experimentally obtained. As can be seen from the figure, the workpiece tilt angle θ increases and the number of workpieces recognized increases up to about 50 °, but the workpiece tilt angle θ is about 35 °.
If it exceeds °, the picking success rate will gradually decrease from 100%. That is, in the past, the picking target area was a small area in which the workpiece tilt angle θ was recognized as a substantially perfect circle, but was about 10 ° or less. Will be expanded to. In step 114, if | r _m −r ₀ | <ρ ₂ ,
The process proceeds to step 116 and step 11
The central coordinate G _m obtained by 0 is transmitted to the picking robot 30 side, and this program is terminated. If the determinations at steps 108 and 114 are NO, the process proceeds to step 118, the line segment data is erased from the storage area A, the process returns to step 104, and the same processing is performed thereafter. Further, in step 104, when there is no continuous line segment group and the storage area A becomes empty, this program ends. Here, the specific portion detecting means is steps 106 and 10.
In step 8, the position determining means in steps 110 to 114,
The command means are each accomplished in step 116.

When the above program is executed, the workpiece W based on the center coordinates G _m initially recognized corresponding to the specific part of the part is picked by the picking robot 30.
Will be picked by the hand 40. As another picking method, among a plurality of specific parts of a part recognized from the entire two-dimensional image input by the image input camera 10, a specific part in a posture state with the smallest inclination suitable for picking is selected. Corresponding center coordinates G
It is also possible to find _m and send it to the robot side.
In this case, even if it takes some time to recognize the specific portion, the effect of increasing the picking success probability can be expected.

FIG. 7 shows the confirmation results of the respective cycle times in the perfect circle recognition or the ellipse recognition. still,
FIG. 7A shows the required time (seconds) for each round recognition, and FIG. 7B shows the required time (seconds) for the ellipse recognition. The high-speed picking device for piled parts of the present invention recognizes an ellipse as a “circle”, and therefore the picking repetition time is 2.7 seconds to 9.0 seconds (average 4.8 seconds) to 2.2 seconds due to conventional perfect circle recognition. It can be reduced to ~ 3.8 seconds (3.0 seconds on average). Further, as can be seen from the figure, there is an effect that variations in required time in repeated picking are extremely reduced.

FIG. 8 shows basic steps of high-speed picking from piled parts and a method of dealing with multi-shaped parts. In the above-mentioned embodiment, the picking of the part with holes whose product shape in the drawing is a “circle” group has been described. However, with respect to parts with parallel parts and parts with pins, the simple shape parts are specified as specific parts. By performing the limited recognition as, it is possible to obtain the same operation and effect. Further, although the present invention is intended for a piled-up component, only one component placed on a plane alone, a few components placed in a scattered state, and a state where they are separated in a tray with partitions. It is obvious that the same can be applied to the parts placed at.

[0013]

The present invention is configured as described above, and has a collation model preset with data for specifying the shape of a specific part of a simple shape that can be gripped by a part when it takes a reference posture, and a robot. Matching limit value that is preset based on the allowable limit angle of inclination of the specific part with respect to the reference posture that represents the range in which the specific part can be gripped by the hand, and indicates the range of allowable mismatch between the specific part and the matching model. Is stored in the two-dimensional image, and one part recognized within the range of disagreement defined by the matching limit value with respect to the matching model is detected as the specific part, and the position of the detected specific part is detected. Is determined, and a command to position the hand at that position and pick up a specific part is transmitted to the robot side. Therefore, as long as the specific part of the component exists within the tilt angle at which it can be gripped, the specific part is collated and recognized. Then, the robot hand is positioned at the position of the identified and determined specific portion, and the component is picked up. As described above, in the collation using the collation model corresponding to the specific part of the component, the time required for the collation is extremely short. Also, if the part is tilted and is within the allowable tilt angle for gripping, the specific part of the part is within the collation limit value, so it is reliably recognized and the probability of being gripped by the robot hand is greatly improved. To do.

[Brief description of drawings]

FIG. 1 is an overall configuration diagram showing a high-speed picking device for piled parts according to a specific embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a main part of the apparatus of the embodiment.

FIG. 3 is a flowchart showing a processing procedure of a central processing unit used in the apparatus of the embodiment.

FIG. 4 is an explanatory diagram showing a method of calculating the center of a perfect circle or an ellipse according to the embodiment.

FIG. 5 is an explanatory diagram showing a work posture that determines picking availability according to the embodiment.

FIG. 6 is an explanatory diagram showing a recognized number and a chuck success probability with respect to a work tilt angle according to the embodiment.

FIG. 7 is an explanatory diagram showing respective confirmation results of cycle times in perfect circle recognition or ellipse recognition.

FIG. 8 is an explanatory diagram showing a basic step of high-speed picking from a piled component and a method of dealing with a multi-shaped component.

FIG. 9 is a block diagram showing the concept of the present invention.

[Explanation of symbols]

10-camera for image input 20-object recognition device 2
1-Central Processing Unit 22-Image Processing Unit 23-Storage Device 24-Lighting Control Circuit 30-Picking Robot 40- (Robot)
Hand T-Tray W-Work (part) 100-High-speed picking device Steps 106, 108-Specific part detecting means Steps 110-114-Position determining means Step 116-Command means

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Hiroshi Harada 1-1-1, Showa-cho, Kariya city, Aichi Japan Denso Co., Ltd. (56) References JP 62-28877 (JP, A) JP 60- 116005 (JP, A) JP-A-60-194303 (JP, A)

Claims (1)

(57) [Claims] 1. A contour line of a piled-up component is obtained from a two-dimensional image, a plurality of constituent line segments constituting the contour line are extracted from the contour line, the component is recognized from the constituent line segment, and grasped by a robot hand. In the high-speed picking device, a model corresponding to the specific part for recognizing the specific part of the simple shape of the part that can be grasped in the two-dimensional image, and the shape when the specific part takes a reference posture Collation model storage means for storing a collation model preset by data for identifying, and in advance based on an allowable limit angle of inclination of the specific portion with respect to the reference posture, which represents a grippable range of the specific portion by the hand. Collation limit value storage means for storing a collation limit value that is set and indicates a range of allowable mismatch between the specific part and the collation model; Specific part detecting means for detecting, as the specific part, one part recognized within the non-coincidence range defined by the matching limit value from the two-dimensional image with respect to the matching model, and the detected part. A high-speed picking device for piled parts, comprising: position determining means for determining the position of a specific portion; and command means for positioning the hand at the position to pick up the specific portion.