JPH07286820A - Position measuring method using three-dimensional visual sensor, and positional deviation correcting method - Google Patents

Abstract

PURPOSE:To provide a position measuring method by which the position of an object to be measured can be measured with accuracy even when the distance between a camera and the object changes and a positional deviation correcting method by which the positional deviation between objects can be corrected based on the measured results of the position measuring method. CONSTITUTION:A robot controller 1 successively forms light belts 61 and 62 on the fitting surface of a mechanism section 50 by moving a robot 20 to a position A and turning on a light projector 11 through a sensor section controller 3. The picture of the light belts is taken with a CCD camera 12 and fetched to a picture processor 2. Then the picture of the light belts 61 and 62 is analyzed and the three-dimensional positions of points P1-Q2 on the outer edge of a surface 53 are found, and then, the position and direction of the surface 53 are found. The controller 1 calculates the robot position correcting amount to the position B facing the surface 53 at a fixed interval and moves the robot 20 to the position B. The camera 12 usually takes the picture of the light belts and the processor 2 fetches the video of the surface 53, finds the position of a hole 54, and calculates the positional deviation of the hole 54. Furthermore, the processor 2 determines the robot position correcting amount corresponding to the positional deviation of the hole 54 to the position of the robot 20 for assembling parts 6 and executes assembling work by moving the robot 20 to the corrected assembling position.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is a method for measuring the position or position and orientation of an object (hereinafter, both are simply referred to as "position" unless otherwise specified) using a three-dimensional visual sensor. In addition, the present invention relates to a method for correcting the positional relationship with the work-related object based on the information indicating the positional deviation amount of the object obtained by the measurement (hereinafter, referred to as “positional deviation correction method”). Fields of application of the present invention include, for example, assembling work and processing work in a manufacturing line of a factory.

[0002]

2. Description of the Related Art In assembling work and processing work in a factory production line, an automatic machine such as a robot and a visual sensor using a camera means such as a CCD camera are used in order to automate and save the work. A combined vision system is used. For example, one assembly work to be supplied to a predetermined position on the manufacturing line is imaged by a CCD camera, and an accurate position of the work is detected by a visual sensor by analyzing the obtained image. The positional deviation between the works is corrected by a method of correcting the position of the robot holding the other work to be assembled based on the above.

By the way, in order to accurately measure the position, orientation, dimensions, etc. of an object by ordinary image analysis, the following conditions must be satisfied in the positional relationship between the camera used and the object. There is.

(1) It is assumed that the displacement of the object occurs only in a plane perpendicular to the optical axis of the camera.

(2) The distance between the camera and the plane where the object is present does not change.

Therefore, when the vision system is used, it is necessary to design the manufacturing line so that the measurement target (work to be assembled or the like) can be supplied so as to satisfy the above conditions. In particular, when the positional deviation includes a component that changes the distance between the camera and the object, it is difficult to perform the positional deviation correction using the vision system.

In order to avoid such a restriction, a three-dimensional visual sensor is incorporated into the vision system. A typical three-dimensional visual sensor projects slit-shaped light onto a target object from a predetermined direction to form a light band of higher brightness than the surroundings on the target object, and this is a camera means such as a CCD camera. The three-dimensional measurement of the object is performed by observing.

However, with such a three-dimensional visual sensor, although three-dimensional position information about a characteristic portion such as a contour line of a workpiece that can be grasped as a bending point or an end point of an optical band can be obtained, a portion apart from the contour of the workpiece is obtained. It is not easy to accurately detect the position of a certain fine feature (for example, a screw hole). Therefore, the correction of the positional deviation of the measured object that is necessary for smoothly performing the assembly work, the processing work, etc. (relative to the work-related object depending on the change in the position of the measured object) It has been difficult to accurately adjust the positional relationship so that the relative positional relationship between the two and the planned positional relationship are matched) based on the measurement results.

[0009]

SUMMARY OF THE INVENTION An object of the present invention is to overcome the problems of a conventional position measuring method using a vision system or a positional deviation correcting method based on the method. That is, the present invention provides a position measurement method capable of performing accurate position measurement even when the distance between the camera means and the target object changes, and also based on the measurement result It is intended to provide a method for accurately correcting the positional deviation of the.

[0010]

SUMMARY OF THE INVENTION The present invention has a basic configuration of a method for measuring the position of an object, which solves various problems in the above-mentioned prior art, as follows: "Projecting light in a controlled direction A method of measuring the position of an object using a three-dimensional visual sensor including a light projecting unit that forms a light band on an object, a camera unit, and an image analysis unit that analyzes a video signal acquired by the camera unit. And a step of forming a light band on a predetermined surface of the object using the light projecting means, and the object using the camera means in a state where the light band is formed on the predetermined surface of the object. Is photographed, and an image including an image of a light band formed on the predetermined surface of the object is analyzed by the image analysis means to determine the three-dimensional position of the predetermined surface of the object. The step of obtaining and the three-dimensional of the obtained predetermined surface The camera means and the predetermined surface of the target object at a predetermined interval based on the position, and the camera means is used to photograph the target object under the facing condition. Analyzing the image including the image of the predetermined surface of the object using the image analysis means to obtain three-dimensional position information regarding a specific portion of the object within the predetermined surface. The above method "(claim 1) is proposed.

Further, in addition to the above configuration, the requirement that "the step of causing the camera means and the predetermined surface of the object to face each other at a predetermined interval is performed by a robot supporting the camera means" ( (2) or the requirement that "the step of causing the camera means and the predetermined surface of the object to face each other at a predetermined interval is performed by a robot that supports the object" (claim 3) It is proposed to use a robot for the above position measuring method.

Further, according to the present invention, the position measuring method according to the above-mentioned basic structure is linked to the correction of the deviation of the relative positional relationship expected between the work-related objects, and the "optical direction toward the controlled direction" is set. 3 including a light projecting unit for projecting light to form a light band on the first object, a camera unit, and an image analyzing unit for analyzing a video signal acquired by the camera unit.
The position of the first object is measured using a three-dimensional visual sensor, and the deviation from the expected relative positional relationship between the first object and the second object is measured based on the measurement result. A method for correcting the positional deviation of an object, the method comprising: forming a light band on a predetermined surface of the first object using the light projecting means; and on a predetermined surface of the first object. The first object is photographed with the camera means in a state where a light band is formed in the image, and the image including the image of the light band formed on the predetermined surface of the first object is taken as the image. By analyzing using the analyzing means, 3 of the predetermined surface of the first object can be obtained.
A step of obtaining a dimensional position, and 3 of the obtained predetermined surface
Causing the camera means and the predetermined surface of the first object to face each other at a predetermined interval based on a dimensional position; and using the camera means under the facing condition A specific part within the predetermined surface of the first object by capturing an image of the first object and analyzing an image including an image of the predetermined surface of the first object using the image analysis means. And a step of obtaining three-dimensional position information regarding the specific part, and based on the obtained three-dimensional position information regarding the specific part, a schedule is set between the first object and the second object. And a step of determining a deviation from the relative positional relationship and correcting the relative positional relationship between the first object and the second object so as to eliminate the deviation. ” (Claim 4) is proposed.

Further, according to the present invention, in order to utilize the robot for the positional deviation correction, "the step of causing the camera means and the predetermined surface of the first object to face each other at a predetermined interval is And a step of causing the camera means and the predetermined surface of the first object to face each other at a predetermined interval. Performed by a robot supporting an object "(claim 6), or" to eliminate a deviation from a predetermined relative positional relationship between the first object and the second object. The step of correcting the relative positional relationship between the first object and the second object is performed by a robot supporting the first object or the second object "(claim 7). The above-mentioned basic structure Each positional deviation correction method imposed is that proposed.

[0014]

In the present invention, in order to measure the position of the object,
Three-dimensional vision including a light projecting means for projecting light in a controlled direction to form a light band on an object, a camera means, and an image analyzing means for analyzing a video signal acquired by the camera means. The sensor is roughly divided into two types. First, the light projecting means projects the slit light toward the target object or projects the light in the spot beam scanning form to form an optical band on a predetermined surface. Camera means such as a CCD camera photographs an object having a light band formed on a predetermined surface, and a video signal representing a video including a video of the light band is acquired. This optical band formation and photographing / image signal acquisition are repeated as many times as necessary until a video signal for a required number of optical bands is obtained.

The image signal representing the light band is analyzed by the image analysis means, and from the three-dimensional position of the characteristic point (bending point, end point, etc.) of the light band, three-dimensional position information about a predetermined surface on which the light band is formed is obtained. Be won. In principle, if the three-dimensional positions of three or more characteristic points are determined, the equation representing the predetermined surface on which the light band is formed can be determined. The process itself performed so far by combining the respective elements including the light projecting means of the three-dimensional visual sensor is conventionally known.

Next, the relative positional relationship between the camera means and the object is adjusted on the basis of the obtained three-dimensional position of the predetermined surface, and the camera means and the predetermined surface which is the object of measurement are previously set. A state of facing each other is realized at a fixed interval. Therefore, the camera means or the object is moved. A robot is suitable as the moving means. The robot supports the camera means or the object, and moves to a position where the camera means and the predetermined surface to be measured face each other at a predetermined constant interval. The inclusion of such a relative position adjustment step is a basic and important feature of the present invention.

Then, under such a facing condition, the object is photographed again by using the camera means, and the image including the image of the predetermined surface of the object is analyzed by the image analyzing means. However, this time, instead of analyzing the image of the light band, a specific portion within a predetermined plane is grasped as light and dark information, and position information regarding the specific portion is obtained. The information obtained by the camera means is two-dimensional information, but since the three-dimensional position information of the predetermined surface where the specific part exists is already obtained, the three-dimensional position information of the specific part is eventually acquired. It becomes possible.

The above is the outline of the process which constitutes the position measuring method of the object according to the present invention. By adding a process for correcting the positional deviation to such a process, the process is operatively related. It is possible to correct the relative positional relationship between the target objects so that they match the planned relationship.

In assembling work, processing work, etc., 2
Often a specific relative positional relationship is planned between two objects. For example, when incorporating the other component into one component, it is necessary to align both components in advance. Further, in the case where some processing is performed on a specific position of the parts that are sequentially supplied, the position of each part and the relative position of the processing tool are accurately adjusted in order to avoid a decrease in processing accuracy due to variations in the position of supplying the parts. The need arises.

Therefore, in the position shift correcting method according to the present invention, one object (first object) related to the work is
To obtain three-dimensional position information of a specific portion (for example, the position of a screw hole), and to establish a specific relative positional relationship with the first object based on the information. Corrects the relative positional relationship with another scheduled object (second object) by a necessary amount. Relative position adjustment after this measurement, the first or second
It is carried out by moving the object.

A robot is suitable as the moving means. It is practical to use a camera means or a robot supporting the first object in this measurement process. When the robot supports the first or second object, it can be considered that the relative positional relationship planned between the two corresponds to the teaching trajectory of the robot. It is practical to implement it in a way that corrects the trajectory.

The moving means for directly facing the camera means and the object at the measuring stage and the moving means used for the positional deviation correction are not necessarily robots. That is, any automatic machine having the ability to move to the position designated by the control device while supporting the camera means or the object can be used for the position measuring method or the positional deviation correcting method of the present invention. Is.

As a light projecting means included in the three-dimensional visual sensor, a slit light projecting device is typical, but a spot light scanning device as disclosed in Japanese Patent Application No. 5-32407. It is also possible to use the floodlighting device.

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a schematic block diagram showing an example of a system configuration used when carrying out each method of the present invention. The entire system includes a robot controller 1, an image processing device 2, a sensor unit controller 3, a sensor unit 10, and a robot 20. The robot controller 1 is of a commonly used type and has a central processing unit (hereinafter referred to as "CPU") including a microprocessor.

The CPU has a ROM memory in which programs for controlling the entire system are stored, a program for controlling robot operations, command transmission to the image processing apparatus, reception of image analysis results from the image processing apparatus, and related programs. A RAM memory that stores set values and the like, is also used as a memory for temporary storage when the CPU executes calculations, a register area that is set as necessary, and each axis of the robot 20 is controlled via a servo circuit. Axis controller and image processing device 2
A general-purpose signal interface that functions as an input / output device for an off-line programming device, a control unit of a manufacturing line, and the like is connected via a bus (illustration of individual elements is omitted).

The image processing apparatus 2 is also of a type that is normally used, and has a CPU that controls the entire apparatus. This CPU has a ROM memory in which a program for image analysis, a program for controlling the sensor unit controller 3 in accordance with a command from the robot controller 10 and the like are stored, and an image captured by the camera 12 is converted into a gray scale. Frame memory for storing as a signal, image processing processor for image analysis, RA for storing various setting values and temporarily storing calculation data
CRT for M memory, sensor controller 3 and monitor
A general-purpose signal interface that serves as an input / output device for is connected via a bus (illustration of individual elements is omitted).

Reference numerals 21, 22, and 23 represent the arms of the robot 20, and the sensor unit 10 is the arm 2 at the end.
It is attached to 3. In addition to the sensor unit 10, the tip of the arm 23 is for work (for example, for holding an assembled component).
Is attached (not shown).

The projector 11 of the sensor section 10 converts the laser light into slit light 13 and projects it onto the object 4 to be measured. A light band 14 is formed on the object 4, and the reflected light 15 is observed by the camera 12 as a light band image. Floodlight 1
The projection direction of the slit light 13 by the image processing device 1 is
Is controlled by the control signal given from the sensor controller 3 to the deflection angle of the deflection mirror built in the projector 11. Therefore, the image processing device 2 can always recognize in which direction the light is projected, as well as the on / off state of the light projector 11.

When the slit light 13 is projected in the direction designated by the image processing apparatus 2 and the optical band 14 is formed on the predetermined surface of the object 4, the image processing apparatus 2 causes the sensor controller 3 to move. A shooting command is sent to the CCD camera 12 via the camera. The image of the light band 14 taken by the CCD camera 12 is sent to the image processing apparatus 2 via the sensor unit controller 3. In the image processing device 2, the image signal taken in through the general-purpose signal interface is converted into a light / dark signal based on a light and shade gray scale and then stored in the frame memory.

Such projection of slit light and photographing of a light band image / acquisition of an image signal are usually executed for a plurality of light bands by changing the slit light projection direction at appropriate intervals. The image processing device 2 analyzes the optical band image using the image processing processor, and obtains three-dimensional positions of the bending point, the end point, etc. of the optical band 14. The triangulation principle is used to determine the three-dimensional position of the bending point, the end point, or the like. Like this 3
Since the dimension position measuring method itself is already known, detailed description thereof will be omitted. When the three-dimensional positions of three or more bending points or end points that are not on a straight line are obtained, the CPU of the image processing apparatus 2 calculates the plane (direction and position) on which the three points are based on the data.

The camera 12 of the sensor section 10 turns off the projector 11 according to a command from the image processing apparatus 2 and photographs the object 4 under normal illumination light or natural light. The image obtained by photographing is converted into a light and dark signal by a gray scale, and is stored in the frame memory, like the light band image. This image contains detailed information that appears as light and dark on the surface of the object 4. In the present invention, such normal photographing is executed in a state in which the robot 20 is moved and the camera 12 is made to face the predetermined surface (the surface on which the slit light is projected) of the object 4 at regular intervals.

The CPU of the image processing apparatus 2 analyzes the image acquired in such a facing state, and calculates the position of a specific portion (for example, a hole) in the predetermined plane of the object 4.
Data representing the position of the specific part within the predetermined plane and data representing the direction of the predetermined plane previously obtained (for example,
Based on the vector data representing the direction and length of the perpendicular line drawn from the origin of the coordinate system set on the robot to the surface, three-dimensional position information (for example, the center position of the hole, the extending direction, Size, shape, etc.).

Robot controller 1 and image processing device 2
Are connected by a communication line via each general-purpose signal interface, and the operation command of the image processing apparatus 2 and the transmission / reception of the position data regarding the object 4 obtained by the image processing apparatus 2 are transmitted to the memory of the robot controller 1. It is designed to be executed according to the stored operation program.

FIG. 2 is a schematic diagram showing the entire arrangement when the method of the present invention is applied to the positional deviation correction in the assembly work of a small robot by using the system having the above-described configuration and function. is there. In the figure,
The elements designated by the same reference numerals as those in FIG. 1, that is, the configurations, connection relationships, and functions of the robot controller 1, the image processing device 2, the sensor unit controller 3, the sensor unit 10, and the robot 20 have already been described.

Objects corresponding to those shown as the object 4 to be measured in FIG. 1 are successively supplied into the working space of the robot 20 and are part 5 (hereinafter, referred to as "the small robot mechanism portion" during the assembly process. Mechanical unit 5 "). The mechanical unit 5 includes a base 50 and a rotation shaft 51 thereof.
It includes an arm 52 that rotates about. For this reason, the mechanism portion to which the position (including the posture) of the mounting surface 53 to which a new component such as a reduction gear and a motor is mounted and the hole 54 into which the convex portion (for example, a screw portion) of the mounting component is inserted is supplied every time. It is unavoidable that there is a deviation for every 5 points, and there is a displacement from the planned reference position.

In FIG. 2, new parts such as the speed reducer and the motor are represented by the part 6 held by the hand H of the robot 20, and details of the shape are omitted. The symbol T is
The tool tip point set at the tip of the convex portion of the component 6 is shown. Further, the two robot positions A and B shown in the drawing are teaching positions for measuring the position / direction of the mounting surface 53 which is executed by turning on the light emitters, and turn off the light emitters determined based on the teaching positions. Represents a normal shooting position to be executed. This normal photographing position B is the CCD of the sensor unit 10.
The optical axis C of the camera is chosen to be perpendicular to the mounting surface 53 (see Examples below).

With such an arrangement, in order to correct the positional deviation and execute the work of assembling the component 6, the robot controller 1 performs image processing by the robot 20 according to the processing procedure described in the flowchart shown in FIG. A program for controlling the entire system including the device 2 is stored. The outline will be described below. In addition, here, the projector 11
The deflection angle of the deflection mirror, the switching contents, the setting of the number of times of light projection / imaging, the setting of the data indicating the reference position / direction of the mounting surface 53 and the hole 54, the position / direction of the mounting surface 53 and the hole 54 are the reference position / It is assumed that the preparatory work including teaching of the optimum normal photographing robot position B (including the distance data at the time of direct facing), the position data of the assembly work execution position, and the like, which are optimal when the directions match, is completed.

When the CPU of the robot controller 1 receives an external signal indicating that the mechanical section 50 has been supplied to the assembly position, processing for one work cycle is started.
First, the operation program is read block by block, and the robot 20 is moved to a robot position A taught as a position for emitting slit light to project the optical band (step S1).

The robot 20 is temporarily stopped at the position A, a command is sent to the image processing apparatus 2, the projector 11 is turned on via the sensor section controller 3, and a light band 61 is formed on the mounting surface 53 of the mechanism section 50. (Step S2). In this state, a shooting command is further sent to the image processing apparatus 2, shooting is performed by the CCD camera 12 via the sensor unit controller 3, and an image including the optical band 61 is taken into the image processing apparatus 2 (step S3).

Steps 3 and S4 are repeated the set number of times while switching the light projecting direction according to the set contents (steps S2 to S4). Here, assuming that the setting is twice, an image including the light band 62 is captured in the second projection / light band photographing. When the photographing is completed, the projector 11 is turned off (step S5).

Next, the accumulated light band images are analyzed using an image processor, and the bending points P1 and P2 and end points Q1 and Q2 of the light bands 61 and 62 corresponding to various points on the outer edge of the mounting surface 53 are analyzed. The three-dimensional position is calculated, the position / direction of the mounting surface 53 is obtained based on the result (step S7), and the deviation from the reference position / direction is calculated (step S8).

This result is transmitted to the robot controller 1, and the normal photographing robot position B facing directly at the distance set on the mounting surface 53 in the robot controller 1.
The correction amount of the robot position that gives is obtained (step S9). The deviation amount obtained in step S8 can be expressed as a deviation amount from the reference value of the vector data representing the direction and length of the perpendicular line drawn from the origin of the coordinate system set on the robot to the surface, Using this value, the correction amount of the robot position is calculated in step S9.

The CPU of the robot controller 1 restarts the movement of the robot 20, and moves the robot 20 toward the position where the teaching position corresponding to the robot position B is corrected by the robot position correction amount obtained in step S9 ( Step S10).

When the robot 20 arrives at the position B, the robot 20 is stopped again, a photographing command is sent to the image processing device 2, and the CCD camera 12 is sent via the sensor unit controller 3.
Then, the normal image capturing is performed, and the image of the mounting surface 53 on which the light band is not formed is captured in the image processing apparatus 2 (step S11).

Next, the captured image is analyzed in the image processing apparatus 2 by using the image processor to obtain the position of the hole 54 (sensor data). The result is transmitted to the robot controller 1 and the three-dimensional position of the hole 54 is calculated (step S12). Since this calculation two-dimensionally determines the position of the hole 54 in the mounting surface 53 whose position / direction has already been obtained, it can be executed without any problem even if the position / direction of the mounting surface 53 varies. You can do it.

When the three-dimensional position of the hole 54 is obtained, the hole 5
A deviation amount from the reference position set for No. 4, that is, a "position deviation amount" is obtained (step S13). Furthermore, for one or more assembly work robot positions taught for assembling the component 6, a robot position correction amount corresponding to the position shift amount of the hole 54 is determined (step S14), and the robot movement is restarted. Then, the robot 20 is moved to the corrected assembly work position (if there is more than one, they are sequentially moved), and the taught assembly work is executed (step S15). When the assembling work is completed, the robot 20 is returned to the home position to complete one work cycle.

In this embodiment, the entire sensor unit 10 is mounted on the robot 20 together with the assembly parts 6 and the mounting surface 53 is attached.
Although the light was projected onto and photographed, the projector 11 was fixed at an appropriate position in the work space, and only the camera 12 was used by the robot 20.
It is also possible to adopt the arrangement mounted on the.

Further, there is no variation in the position of the assembly process product (corresponding to the mechanical unit 50 in FIG. 2) on the side not gripped by the robot, and there is variation in the gripping state of the assembly component (corresponding to component 6 in FIG. 2) by the robot. In the case where it is predicted, the positional deviation correction can be performed by the procedure according to the above-described embodiment by arranging the entire sensor unit (camera and light projector) at an appropriate position in the work space.

That is, the robot is stopped at an appropriate position (corresponding to the robot position A in the above-mentioned embodiment), the light band formation / light band image photographing is performed by the light projector and the camera which are fixedly arranged, and the part gripped by the robot. Then, the position / direction of the predetermined surface is determined, and the robot is moved accordingly to directly face the predetermined surface of the component with the camera at a constant interval (corresponding to the robot position B in the above embodiment). Then, by performing normal imaging and image analysis of the predetermined surface of the part gripped by the robot, the amount of deviation from the reference position (three-dimensional position) of the specific portion (hole, edge of the convex portion, etc.) in the specific surface. Ask for. If the subsequent assembly work robot position is corrected so as to cancel out the position shift amount, the position shift corrected robot work can be executed.

Further, in the measuring stage, the moving means for directly facing the camera means and the object and the moving means used for the positional deviation correction are the positions designated by the control device while supporting the camera means or the object. The machine need not necessarily be a robot as long as it is an automatic machine having the ability to move to the other side, and the light projection means included in the three-dimensional visual sensor is not limited to the slit light projection type light projection device, but is also a spot light scanning type light projection device. As described above, the optical device can be used.

[0051]

According to the present invention, accurate position measurement can be performed even when the distance between the camera means and the object changes. Further, it becomes possible to accurately correct the positional deviation between the objects based on the measurement result.

When the position deviation correcting method of the present invention is applied to assembly work, processing work, etc. by a robot, the robot work is effectively used for both the position measurement according to the present invention and the position deviation correction based on the position measurement. The accuracy of can be improved.

[Brief description of drawings]

FIG. 1 is a block diagram of a main part schematically showing an example of a system configuration used when implementing each method of the present invention.

FIG. 2 is a schematic diagram showing the overall arrangement when the method of the present invention is applied to the positional deviation correction in the assembly work of a small robot.

FIG. 3 is a flowchart showing an outline of a processing procedure for correcting a positional deviation and performing a component assembling work in the arrangement shown in FIG.

[Explanation of symbols]

 1 Robot Controller 2 Image Processing Device 3 Sensor Unit Controller 4 Object to be Measured 5 Mechanism Unit 6 Assembly Parts 10 Sensor Unit 11 Projector 12 CCD Camera 20 Robot 21-23 Robot Arm 50 Base 51 Rotation Axis 52 Arm 53 Attachment Surface 54 Hole A Optical band shooting robot position B Normal shooting robot position C Optical axis of CCD camera H Hand

Continuation of the front page (51) Int.Cl. ⁶ Identification code Office reference number FI Technical display location B25J 19/04 G01B 11/24 C G06T 7/00 7/60

Claims (7)

[Claims] 1. A light projecting means for projecting light in a controlled direction to form a light band on an object, a camera means, and an image analyzing means for analyzing a video signal acquired by the camera means. A method of measuring the position of an object using a three-dimensional visual sensor including: a step of forming an optical band on a predetermined surface of the object using the light projecting means; The object is photographed by using the camera means in a state where a light band is formed on, and an image including an image of the light band formed on the predetermined surface of the object is analyzed by the image analysis means. The step of obtaining a three-dimensional position of the predetermined surface of the object, and the camera means and the predetermined surface of the object based on the obtained three-dimensional position of the predetermined surface. A step of directly facing each other at a predetermined interval, Of the target object by photographing the target object using the camera means underneath and analyzing the image including the video of the predetermined surface of the target object using the image analysis means. The method, comprising the step of obtaining three-dimensional position information regarding a specific part. 2. The robot supporting the camera means, wherein the step of causing the camera means and the predetermined surface of the object to face each other at a predetermined interval is performed by a robot supporting the camera means. the method of. 3. The robot supporting the target object, wherein the step of causing the camera means and the predetermined surface of the target object to face each other at a predetermined interval is performed by a robot supporting the target object. the method of. 4. A light projecting means for projecting light in a controlled direction to form a light band on a first object, a camera means, and a video signal acquired by the camera means. The position of the first object is measured using a three-dimensional visual sensor including an image analysis means, and the relative distance between the first object and the second object is scheduled based on the measurement result. A method for correcting a positional deviation of an object for correcting a deviation from a physical positional relationship, the method comprising: forming a light band on a predetermined surface of the first object using the light projecting means; An image of the light band formed on the predetermined surface of the first object by photographing the first object using the camera means in a state where the light band is formed on the predetermined surface of the object By analyzing the image including the image using the image analysis means, the predetermined object of the first object is analyzed. Of the three-dimensional position of the first surface of the first object and the camera means and the first surface of the first object are directly opposed to each other at a predetermined distance based on the obtained three-dimensional position of the predetermined surface. The step of causing the first object to be photographed using the camera means under the facing condition, and using the image analysis means to obtain an image including an image of the predetermined surface of the first object. A step of obtaining three-dimensional position information regarding a specific portion of the first object within the predetermined plane by analyzing, and the step of obtaining the three-dimensional position information regarding the specific portion obtained based on the obtained three-dimensional position information. A deviation from a predetermined relative positional relationship between the first object and the second object is obtained, and the first object and the second object are arranged so as to eliminate the deviation. And a step of correcting a relative positional relationship. Law. 5. The robot supporting the camera means performs the step of causing the camera means and the predetermined surface of the first object to face each other at a predetermined interval. The method according to 4. 6. The robot supporting the first object is characterized in that the step of causing the camera means and the predetermined surface of the first object to face each other at a predetermined interval is performed. The method of claim 4, wherein 7. The first object and the second object so as to eliminate a deviation from a predetermined relative positional relationship between the first object and the second object. The step of correcting the relative positional relationship of
The method according to claim 4, wherein the method is performed by a robot that supports the object.