JPH04148205A - Positioning device - Google Patents

Abstract

PURPOSE:To realize positioning corresponding to recognition precision by obtaining the relative positional relation of a moving body and an objective body from a shadow by the shape of the objective body, the shadow of a projection pattern obstructed by the moving body, and the shadow of reflected light from the objective body obstructed by the moving body. CONSTITUTION:An image processing means 14 obtains the relative positional relation of the moving body 3 and the objective body 8 on the basis of position information obtained by recognizing the shadow formed from the shape of the objective body 8, the shadow formed because the pattern projected toward the objective body 8 from a light source 4 is obstructed by the moving body 3, and the shadow formed because the reflected light from the objective body 8 is obstructed by the moving body 3. Accordingly, the relative positional relation of the moving body 3 and the objective body 8 can be obtained through the camera input image processing of one time, and the moving direction and the movement of a moving mechanism can be determined from the relative positional relation. Thus, the positioning not based on teached precision but corresponding to the recognition precision can be realized.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a positioning device having a visual and moving mechanism used, for example, in a visual screw tightening robot, a component inserter with a visual function, and a component mounting machine.

Conventional technology Conventionally, when aligning a moving object and a target object using a robot with vision, a positioning device is taught the reference positions of the target object, the moving object, and the movement of the robot as a moving mechanism. The position of the moving object is recognized from the image input by the camera, the position of the moving object is recognized from the image input by another camera, and the difference between the recognized position when the target object is at the reference position and the position when the moving object is at the reference position is determined. The method used was to perform coordinate transformation from each visual coordinate to the robot coordinate based on the difference from the recognized position at the time, and determine the amount of motion correction from the reference position of the moving mechanism.

Problems to be Solved by the Invention However, the conventional configuration described above has a problem in that the teaching settings of the robot motion and the reference position of the target object/moving object directly affect the accuracy of the alignment work. Therefore, when highly accurate positioning is required, teaching often requires a great deal of effort. Furthermore, the accuracy of the positioning operation cannot exceed the accuracy of the teaching settings. In addition, it is necessary to perform recognition in two steps, and in many cases, recognition must be performed in a separate step from the alignment operation, which increases the time required to perform the work operation including the recognition operation. I had an issue.

The present invention solves the above-mentioned conventional problems, and an object of the present invention is to provide a positioning device that can perform positioning with position detection accuracy and positioning accuracy of a moving mechanism, regardless of the teaching accuracy of a reference position.

Means for Solving the Problems To achieve this object, the alignment device of the present invention comprises:
A light source that projects a pattern onto the moving object and the target object at an angle to a straight line descending from the moving object to the target object, and a light source that projects the light onto the moving object and the target object at an angle that is different from the light projection direction of the light source. A camera that receives the reflected light of the pattern, an image processing means that determines the relative positional relationship between the moving object and the target object from the input image received by the camera, and a device that determines the direction and amount of movement of the moving object from the relative positional relationship and moves the object. and a control means for controlling the movement of the mechanism, and the image processing means detects the shadow generated from the shape of the target object, the shadow generated when the pattern projected from the light source onto the target object is blocked by the moving object, and the target object. The apparatus is configured to determine the relative positional relationship between the moving object and the target object based on positional information obtained by recognizing the shadow created when the reflected light from the moving object is blocked by the moving object.

Effect: With this configuration, a light source emits a pattern onto the moving object and the target object in a direction that forms an angle with a line segment connecting a predetermined position of the moving object and the target object, and a light source emits a pattern at an angle different from the light projection direction of the light source. A camera that receives reflected light, an image processing means that determines the relative positional relationship between the moving object and the target object from the input image received by the camera, and a device that determines the direction and amount of movement of the moving object from the relative positional relationship and moves the J-motion mechanism. The image processing means controls shadows generated from the shape of the target object, shadows generated when the pattern projected from the light source onto the target object is blocked by a moving object, and reflections from the target object. Since the relative positional relationship between the moving object and the target object is determined based on the position information obtained by recognizing the shadow created when light is blocked by the moving object, the movement can be performed from a single camera input image processing. The relative positional relationship between the object and the target object can be determined, and the amount of movement in the moving direction of the moving mechanism can be determined from the relative positional relationship. Therefore, positioning can be achieved in accordance with recognition accuracy, regardless of teaching accuracy.

EXAMPLE An example of the present invention will be described below with reference to the drawings.

FIG. 1 is a configuration diagram of a positioning device in an embodiment of the present invention. A gripping device 2 is attached to the tip of the robot 1, and the gripping device 2 grips a component 3. light source 4
and the camera 6 are attached to the gripping device 2 so as to face each other or perpendicularly to each other, and move together with the gripping device 2 as the robot 1 moves.

The light emitted by the light source 4 is a laser beam that travels in a straight line, and projects a rectangular pattern 6. The lens of the camera 6 is equipped with a filter that transmits the wavelength band of the laser beam, so that the projected pattern 6 is received. light source 4,
The control signals IaFi for the camera 6 and the robot 1 are each connected to the controller 7, and are controlled by the bathing power of the controller 7. Note that 8 is a workpiece, and 9 is a hole provided in the workpiece 8 into which the component 3 is inserted.

The pattern projected by the alignment device configured as above will be described below.

FIG. 2(a) is a main part configuration diagram showing the positional relationship between the light source of the alignment device and the workpiece, and FIG. 2(b) is a pattern of light projected onto the workpiece. As shown in FIG. 2(a), a part of the pattern 8 emitted from the light source 4 installed diagonally upward to the left is blocked by the component 3, creating a shadow 10. As a result, as shown in FIG. 2(g), a pattern 6 is formed which has a cutout 9' corresponding to the hole 9 of the workpiece 8 and an edge 1o blocked by the part 3.

Images of this pattern 6 captured by the camera 6 are shown in FIGS. 3 and 4.

FIG. 3(a) is a main part configuration diagram showing the positional relationship between the camera and workpiece of the positioning device in the first embodiment;
3) is a pattern in which the light enters the camera, and in this case, the light source 4 and camera 6 are in a positional relationship in which they are opposed to each other.

The pattern 6 reflected by the workpiece 8 and incident on the camera 6 is again blocked by the part 3, creating a shadow 11, as shown in FIG.
g) The shape will be as shown in K.

In this case, if the angles formed by the optical axis of the light source 4 and the optical axis of the camera 6 are made equal to the perpendicular line descending from the component 3 to the workpiece 8, the position and height of the component 3 can be determined as follows. The position of the component 3 is at the midpoint of the line segment connecting the tips of the shadows 1o and 11, and the height from the workpiece 8 is proportional to the line segment. Therefore, alignment is performed by matching the position of the component 3 determined by the above method with the missing portion 9' on the pattern 6 corresponding to the hole 9 of the workpiece 8.

That is, the relative positional deviation (dzvdp, dz) between the tip position of the component 3 and the hole 9 of the workpiece 8 is calculated as follows.

d! =”10”11)/2 “9’dy” (”1o”711)/2-79/d =
(x, .-xll)/2 Also, (!9/ l Ygl) is the coordinate of the missing part 9, (
x. e]' 1°,) is the coordinate of the shade 1o, ("11.
yll) is the coordinate of the shadow 11, and the angle between the light source 4 and the camera 5 and the perpendicular line descending from the component 3 to the workpiece 8 is 45°.

FIG. 4(a) is a main part configuration diagram showing the positional relationship between the camera of the alignment device and the workpiece in the second embodiment;
(b) is a pattern that is incident on the camera; in the case of (b) and (b), the positional relationship between the light source 4 and the camera 6 is orthogonal to each other.

Unlike the embodiment shown in FIG. 3, in FIG.
0 and Yin 11 are in a right-angled relationship.

In this case, the relative positional deviation (d-work, d-a, d2) between the tip position of the component 3 and the hole 9 of the workpiece 8 is calculated as follows.

d! = Fly. -! 9'd7""1179' d = I (!10 "11)" (710711
]/2 However, (! g / e p g / ) is the missing part 9
' coordinate, (xl.s71°) is the coordinate of shade 1o, (!1
1 e711 ) is the coordinate of the shadow 11, and the angle between the light source and the perpendicular line that the camera 6 descends from the part 3 to the workpiece 8 is 46
°, and their optical axes are orthogonal.

FIG. 6 is a configuration diagram of the controller 7. As shown in the figure, the controller 7 includes a central control section 12, a moving mechanism control section 1
3, an image processing section 14, and a system bus 16. The central control unit 12, the moving mechanism control unit 13, and the image processing unit 14 each include a microcomputer, a memory, and an input/output unit (Ilo), and communicate via a system path 16.

The robot 1 moves back and forth, left and right, and up and down by a pole screw driven by an electric motor.

``Each electric motor is controlled by the movement mechanism control section 13. The central control unit 12 sends operation commands to the moving mechanism control unit 13 and image processing commands to the image processing unit 14 according to a program stored in the memory.

Image processing means for determining the relative positional relationship between the component 3 and the workpiece 8 from the input image received by the camera 6 in this embodiment;
Control means for determining the moving direction and amount of movement of the parts 3 from the relative positional relationship and controlling the movement of the robot are realized by programs of the image processing section 14, the central control section 12, and the microcomputer of the moving mechanism control section 13, respectively. .

The image processing unit 14 determines the position of the missing portion, the position of the shadow 10 at the tip of the component 3, and the position of the shadow 11 at the tip of the component 3 from the laser beam pattern 6 received by the camera 6 according to instructions from the central control unit 12. It is recognized by template matching and the obtained recognition position coordinate data is sent to the central control unit 12.

Figure 6 (4) is prepared in advance as a template to be used for recognition processing.
), (b), (C).

The template (d) is stored in the memory of the image processing section 14.

The central control unit 12 determines the position of the missing part V from the image processing unit 14, the position of the shadow 1o at the tip of the component 3, and the shadow 1 at the tip of the component 3.
The relative positional relationship between the tip of the part 3 and the hole 9 is determined by receiving the screen coordinate values of the position 1, and based on this, the amount of movement of the robot 1 is calculated to move to the position of the part 3, and the movement mechanism is controlled. An operation command is sent to section 13.

The relative positional relationship between the tip of the part 3 as a moving object and 60 as a target object is calculated by the respective procedures when the camera 6 and the light source 4 are arranged oppositely and when the camera 5 and the light source 4 are arranged orthogonally. Then, (d!, da, d2) is obtained as the relative position in the image coordinate system, and this is further calculated from the image coordinate system to the robot 1 as a moving mechanism.
The vector k (r-work, r-a, r7) in which the robot 1 should move for positioning is determined by converting it into the coordinate system.

(r!, r,, r,) = (dx, d,, d,)T where T is a (3x3) coordinate transformation matrix.

The processing flow of the central control unit 12 is shown in FIG.

In this embodiment, if the movable object is a screw held by a screw driver and the predetermined position of the target object is a screw hole, the positioning device of the present invention can be applied to a screw tightening robot.

Furthermore, in this embodiment, if the insertion part held by the robot is a moving object and the insertion hole is set at a predetermined position on the target object,
The alignment device of the present invention can also be applied to a component insertion robot.

Further, in this embodiment, if the mounting component is a moving object and the mounting target board is a target object, the positioning device of the present invention can also be applied to a component mounting robot.

Further, in this embodiment, if the glove is used as a moving object and the terminal is placed at a predetermined position on the target object, the positioning device of the present invention can also be applied to an inspection device that positions a probe and a terminal.

Effects of the Invention As described above, in the alignment device of the present invention, the moving object and the camera/light source are moved so that the target object and the moving object are within the range where the camera captures the reflected light of the light projected from the light source. By setting the teaching in this way, the moving object and the target object can be aligned with the visual position detection accuracy and the positioning accuracy of the moving mechanism, regardless of the teaching accuracy. Therefore, there is no need to teach the operating position of the moving mechanism with positioning accuracy, and the labor for teaching is also significantly reduced.

Visually, the positions of the moving object and the target object can be determined simultaneously from the screen by one person, so visual feedback processing for the movement of the moving mechanism can be performed in a short time.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of an alignment device in an embodiment of the present invention, Fig. 2(a) is a block diagram of main parts showing the relationship between the light source of the alignment device and a workpiece, and Fig. 2(b) is a block diagram of a workpiece. Figure 3 (a) is a diagram of the pattern projected onto the top; Figure 3 (a) is a main part configuration diagram showing the positional relationship between the camera and workpiece of the positioning device in the first embodiment; Figure 3 (mountain) is the pattern incident on the camera. 4(a) is a main part configuration diagram showing the relationship between the camera and the workpiece of the positioning device in the second embodiment, and FIG. 4(b)
) is a diagram of the pattern incident on the camera, Figure 5 is a configuration diagram of the controller, Figures 6 (a) to (d) are plan views of various templates stored in the image processing unit, and Figure 7 is the central control unit. FIG. 1...Robot (moving mechanism), 3...
Parts (moving object), 4...Light source, 6...
・Camera, 8... Work (target object),...
...missing part (shadow created by being blocked by a moving object), 10...shadow, 11...shadow, 13.
...Movement mechanism control section (control means), 14...
...Image processing unit (image processing means). Name of agent: Patent attorney Akira Okaji and two others (d) (b) (b> (a) tb) (C) (d)

Claims (4)

[Claims] (1) A device that aligns a moving object to a predetermined position on the target object by moving a moving mechanism, and projects a pattern onto the moving object and the target object at an angle to a vertical line drawn from the moving object to the target object. A light source that emits light, a camera that receives reflected light in a pattern that is projected onto a moving object and a target object at an angle different from the direction in which the light is emitted, and an input image that is received by the camera to determine the relative relationship between the moving object and the target object. The image processing means includes an image processing means for determining a positional relationship, and a control means for determining the moving direction and amount of movement of the moving object from the relative positional relationship and controlling the movement of the moving mechanism, and the image processing means generates a shape from the shape of the target object. position information obtained by recognizing the shadows generated when the light projected from the light source on the target object is blocked by a moving object, and the shadows generated when the reflected light from the target object is blocked by the moving object. A positioning device characterized in that the relative positional relationship between a moving object and a target object is determined based on the following. (2) Claim 1 characterized in that the light source and the camera are arranged so that the light emitting direction of the light source and the light receiving direction of the camera are opposite to each other.
The alignment device described. (3) Claim 1 characterized in that the light source and the camera are arranged so that the light emitting direction of the light source and the light receiving direction of the camera are orthogonal to each other.
The alignment device described. (4) The positioning device according to claim 1, wherein the light source and camera are attached to a moving mechanism so that the light source and camera can move together with the moving object.