JP7011805B2 - Robot control device and robot control method for setting robot control points - Google Patents

Description

The present invention relates to a robot control device and a robot control method for controlling a robot capable of moving an object to a predetermined destination by displacement of a joint.

The present inventor has previously proposed a robot control device disclosed in Patent Document 1 below. This robot control device is a control device for a robot including an arm in which a link is driven via a joint by a driving means, and a reference point set on the arm and a target point to which the reference point is moved are imaged. The position deviation between the reference point and the target point is obtained based on the image pickup result, and the drive voltage for driving the drive means is pulse-width modulated based on the obtained position deviation to generate and generate a PWM signal. By outputting the PWM signal to the drive means, the reference point can be moved to the target point.

In addition, the present inventor has previously proposed a robot control system disclosed in Patent Document 2 below. This robot control system virtually adds the error between the position of the robot's hand detected by the camera and the target position of the hand detected by the camera to the position of the hand detected by the encoder after coordinate conversion. Set the target position and move the hand to the virtual target position.

Further, the present inventor has previously proposed a robot control device disclosed in Patent Document 3 below. This robot control device performs an integral calculation on the value corresponding to the position deviation between the reference point of the robot and its moving destination as the applied amount which is the force or torque applied to the joint to displace the joint of the robot. The first applied amount is calculated from the result of obtaining the integrated value for the reference point, and the displacement amount between the target displacement amount, which is the displacement amount of the joint when the reference point matches the movement destination, and the detected displacement amount of the encoder. The second applied amount is calculated by performing a proportional calculation on the value corresponding to the displacement. Then, the first application amount and the second application amount are added to the joint to move the reference point toward the movement destination.

Japanese Unexamined Patent Publication No. 2015-221474 Japanese Unexamined Patent Publication No. 2016-52699 International Publication No. WO2013 / 176212

However, conventionally, since the setting points of the reference point and the target point are not considered, for example, when moving a three-dimensional object to the moving destination, the object is moved to a predetermined position and in a predetermined direction at the moving destination. It may not be possible to place it. However, such a thing is desired in an actual work site using a robot.

The present invention has been made in view of such circumstances, and a main object thereof is to provide a robot control device and a robot control method capable of arranging an object at a predetermined position in a predetermined direction at a moving destination. To do.

The robot control device according to the present invention for achieving the above object is a robot control device that moves an object by displacement of a joint, and is used to determine the position and posture of the object at the destination. While the reference point is set for the object, the target point as the movement destination of the reference point is set, and the image pickup means for imaging the reference point and the target point, and the robot and the image pickup means. The object is first provided with a control means for controlling the displacement of the joint so that the reference point moves to the target point based on the image pickup result obtained by the image pickup means without performing calibration. It has a region, and the destination of the object has a second region having the same shape as the outer shape of the first region, and the object is moved so that the edge of the first region is the second region. A plurality of the reference points are set on the object, and a plurality of the reference points are set on the edge of the first region, or a plurality of the reference points are set in the control of contacting the edges of the object so as to match the outer shape. Of the above, at least one reference point is set at the edge of the first region, and the remaining reference points are set at a location other than the edge of the first region on the object, and among the plurality of reference points. When the at least one reference point is set at the edge of the first region and the remaining reference points are set at a location other than the edge of the first region on the object. When the destination of the object has a prismatic body having the second region on one end surface in the axial direction and the at least one reference point is one point, the one reference point and the remaining reference points are connected. When the remaining reference points are set on the object so that the reference line segment is perpendicular to one end surface in the axial direction of the square column at the time of contact, and there are a plurality of the at least one reference points, among them. The remaining reference points are set on the object so that the reference line segment connecting one reference point and the remaining reference points is perpendicular to one axial end surface of the square column at the time of contact. Two other target points are set on one side of the side segment between the axial one end surface and the axial end surface of the prismatic body so as to be separated from each other, and the control means is the first region. The displacement of the joint is controlled so that the reference point set on the edge of the object moves to the target point, and the direction of the target line segment connecting the two other target points and the direction of the reference line segment. It is characterized in that the displacement of the joint is controlled so that the two are the same .

The robot control device according to the present invention for achieving the above object is a robot control device that moves an object by displacement of a joint, and is an imaging unit that captures an image of the object and the destination of the object. On the image obtained by the imaging means, a reference point is set on the object and a target point which is the destination of the reference point is set on the destination of the object, and the position of the object and the position of the object at the destination and the target point. The reference point is set based on the setting means for setting the posture and the image in which the reference point and the target point are set by the setting means without calibrating the robot and the imaging means. A control means for controlling the displacement of the joint so as to move to the target point is provided, the object has a first region, and the destination of the object has the same shape as the outer shape of the first region. It has two regions, and in the control of moving the object so that the edge of the first region is brought into contact with the edge of the second region so that the outer shape matches the edge of the second region, a plurality of reference points are set on the object. The plurality of reference points are set at the edge of the first region, or at least one of the plurality of reference points is set at the edge of the first region, and the remaining reference points are set. A point is set on the object at a location other than the edge of the first region, at least one of the plurality of reference points is set on the edge of the first region, and the remaining reference points. When a point is set as one point and the object is set at a position other than the edge of the first region, the destination of the object has a prismatic body having the second region on one end surface in the axial direction. When the at least one reference point is one point, the reference line segment connecting the one reference point and the remaining reference points is perpendicular to one axial end surface of the square column at the time of contact. When the remaining reference points are set on the object and there are a plurality of the at least one reference points, the reference line segment connecting one of the reference points and the remaining reference points is the square column body at the time of the contact. The remaining reference points are set on the object so as to be perpendicular to one end surface in the axial direction, and one of the side portions between the one end surface in the axial direction and the other end surface in the axial direction of the square column. Two other target points are set on the sides apart from each other, and the control means displaces the joint so that the reference point set on the edge of the first region moves to the target point. It is characterized by controlling and controlling the displacement of the joint so that the direction of the target line segment connecting the two other target points and the direction of the reference line segment are the same .

The robot control device according to the present invention for achieving the above object extends the target point corresponding to the reference point, which is one end of the reference line segment, and the target point in the same direction as the target line segment. The other target point is set so that the length of the line segment connecting to another target point set on the extending line is the same as the length of the reference line segment, and the control means is the reference. It is characterized in that the displacement of the joint is controlled so that the remaining reference point moves to the other target point so that the direction of the line segment and the direction of the target line segment become the same.

In the robot control device according to the present invention for achieving the above object, the control means has the reference line segment so that the direction of the reference line segment and the direction of the target line segment are the same. A deviation corresponding to the length of the line segment drawn perpendicularly from the remaining reference points with respect to the extending line extending in the same direction as the target line segment from the target point corresponding to the reference point at one end of the above. It is characterized in that the displacement of the joint is controlled based on the deviation vector.

The robot control device according to the present invention for achieving the above object has an insertion portion in which the object has the first region at the tip, and the second region has the insertion portion of the object. It is a insertion portion to be inserted, and after the object is moved and the edge portion of the first region is brought into contact with the edge portion of the second region so that the outer shape matches the insertion portion, the insertion portion is brought into contact with the insertion portion. In the control of inserting into the insertion portion, the second region after the insertion portion is inserted into the insertion portion at the proximal end side of the reference point set in the first region. Another reference point corresponding to the target point set in is further set, and the control means sets the other reference point after the contact between the edge portion of the first region and the edge portion of the second region. It is characterized in that the displacement of the joint is controlled so as to move to the target point set in the second region.

According to the robot control device according to the present invention, in order to determine the position and posture of the object at the moving destination, a reference point is set for the object and a target point is set for the moving destination of the object. Then, the reference point and the target point are imaged, and the robot is controlled so that the reference point moves to the target point based on the imaged result. Therefore, when the robot moves the object, the object can be placed at a predetermined position and in a predetermined direction at the destination. At this time, since the robot and the image pickup means are not calibrated, the burden on the calibration can be reduced.

According to the robot control device according to the present invention, an object and its moving destination are imaged, a reference point is set for the object, and a target point is set for the moving destination of the object on the captured image. To. At this time, the reference point and the target point are set so that the position and posture of the object at the moving destination can be determined. Then, the robot is controlled so that the reference point moves to the target point based on the image in which the reference point and the target point are set. Therefore, when it is difficult to set a reference point directly on the object, or when it is difficult to set a target point directly on the destination of the object, when the robot moves the object, the object is moved. Can be placed in a predetermined position and in a predetermined orientation. At this time, since the robot and the image pickup means are not calibrated, the burden on the calibration can be reduced.

Further, according to the robot control device according to the present invention, a plurality of reference points set on the object are set at the edge of the first region, or at least one reference point among the plurality of reference points is set. It is set at the edge of the first region of the object, and the remaining reference points are set at the object other than the edge of the first region. Therefore, the robot can move the object so that the edge of the first region is brought into accurate contact with the edge of the second region so that the outer shape matches the edge of the second region.

Further, according to the robot control device according to the present invention, among the plurality of reference points set on the object, a plurality of reference points are set on the same straight line at the edge of the first region. Then, the remaining reference points are set at the edges of the first region at locations different from the plurality of reference points. At this time, the remaining reference points are set at positions deviating from the same straight line. Therefore, with a simple setting, the edge portion of the first region can be accurately contacted with the edge portion of the second region so that the outer shape matches.

Further, according to the robot control device according to the present invention, the displacement of the joint is controlled so that the reference point set at the edge of the first region of the object moves to the target point set at the edge of the second region. At the same time, the displacement of the joint is controlled so that the direction of the target line segment and the direction of the reference line segment are the same. Therefore, for example, when the thick portion of the plate material is the first region, the edge portion of the first region can be accurately contacted with the edge portion of the second region so that the outer shape matches.

Further, according to the robot control device according to the present invention, the displacement of the joint is controlled so that the reference point set at the edge of the first region of the object moves to the target point set at the edge of the second region. At the same time, the displacement of the joint is controlled so that the remaining reference point moves to another target point. Therefore, for example, when the thick portion of the plate material is the first region, the edge portion of the first region can be accurately contacted with the edge portion of the second region so that the outer shape matches the edge portion of the second region with a simple setting.

Further, according to the robot control device according to the present invention, the displacement of the joint is controlled so that the reference point set at the edge of the first region of the object moves to the target point set at the edge of the second region. At the same time, the displacement of the joint is controlled based on the deviation vector. Therefore, for example, when the thick portion of the plate material is the first region, the edge portion of the first region can be accurately contacted with the edge portion of the second region so that the outer shape matches the edge portion of the second region with a simple setting.

Further, according to the robot control device according to the present invention, after the contact between the edge portion of the first region and the edge portion of the second region, the robot control device is set closer to the base end side than the reference point set in the first region. The displacement of the joint is controlled so that the reference point of is moved to the target point set in the second region. Therefore, the robot can move the object and accurately insert the insertion portion of the object into the insertion portion to which the object is moved.

According to the robot control method according to the present invention, a reference point is set for the object and a target point is set for the moving destination of the object in order to determine the position and posture of the object at the moving destination. Then, the reference point and the target point are imaged, and the robot is controlled so that the reference point moves to the target point based on the imaged result. Therefore, when the robot moves the object, the object can be placed at a predetermined position and in a predetermined direction at the destination. At this time, since the robot and the image pickup means are not calibrated, the burden on the calibration can be reduced.

According to the robot control method according to the present invention, an object and its moving destination are imaged, a reference point is set for the object, and a target point is set for the moving destination of the object on the captured image. To. At this time, the reference point and the target point are set so that the position and posture of the object at the moving destination can be determined. Then, the robot is controlled so that the reference point moves to the target point based on the image in which the reference point and the target point are set. Therefore, when it is difficult to set a reference point directly on the object, or when it is difficult to set a target point directly on the destination of the object, when the robot moves the object, the object is moved. Can be placed in a predetermined position and in a predetermined orientation. At this time, since the robot and the image pickup means are not calibrated, the burden on the calibration can be reduced.

It is a schematic diagram which shows the use state 1 of one Embodiment of the robot control apparatus of this invention. It is a schematic block diagram which shows the robot control device of FIG. It is a flowchart which shows an example of the robot control method using the robot control device of FIG. It is a schematic diagram which shows the use state 1 of the robot control device of FIG. 1, and shows the state which moved the object from the state of FIG. It is a schematic diagram which shows the use state 2 of the robot control device of FIG. It is a schematic diagram which shows the use state 2 of the robot control device of FIG. 1, and shows the state which moved the object from the state of FIG. It is a schematic diagram which shows the use state 3 of the robot control device of FIG. It is a schematic diagram which shows the use state 3 of the robot control device of FIG. 1, and shows the process of moving an object from the state of FIG. It is a schematic diagram which shows the use state 3 of the robot control device of FIG. 1, and shows the state which moved the object from the state of FIG. 7 through the state of FIG. It is a schematic diagram which shows the use state 4 of the robot control device of FIG. It is a schematic diagram which shows the use state 4 of the robot control device of FIG. 1, and shows the state which moved the object from the state of FIG. It is a schematic diagram which shows the use state 5 of the robot control device of FIG. It is a schematic diagram which shows the use state 5 of the robot control device of FIG. 1, and shows the process of moving an object from the state of FIG. It is a schematic diagram which shows the use state 5 of the robot control device of FIG. 1, and shows the state which moved the object from the state of FIG. 12 through the state of FIG. It is a schematic diagram which shows the use state 6 of the robot control device of FIG. It is a schematic diagram which shows the use state 6 of the robot control device of FIG. 1, and shows the process of moving an object from the state of FIG. It is a schematic diagram which shows the use state 6 of the robot control device of FIG. 1, and shows the state which moved the object from the state of FIG. 15 through the state of FIG. It is a schematic block diagram which shows another embodiment of the robot control apparatus of this invention. It is a schematic diagram which shows another use state of the robot control device of FIG.

Hereinafter, specific embodiments of the present invention will be described in detail with reference to the drawings.

1 and 2 are diagrams showing an embodiment of the robot control device of the present invention, FIG. 1 is a schematic diagram showing a usage state 1, and FIG. 2 is a schematic block diagram. The robot control device 1 of this embodiment is a control device of the robot 4 that moves the object 2 to the moving destination by the displacement of the joint 3.

The robot 4 is an industrial robot used in a field such as a factory, and includes an arm 5 rotatably provided on a base and a hand portion 6 provided on the arm 5. The arm 5 has a plurality of link rods 7 and joints 3 that rotatably connect the adjacent link rods 7 and 7 to each other. The arm 5 of this embodiment has three link rods 7, and the adjacent link rods 7, 7 are rotatably connected to each other via a joint 3. As shown in FIG. 1, the arm 5 has a base end rotatably connected to a base via a joint 3 and extends outward, and a hand portion 6 for gripping an object 2 is provided at the tip thereof. .. A motor 8 is provided in each joint 3, and by driving the motor 8, the joint 3 can be rotated, and by extension, the link rod 7 can be rotated. The coordinate system XYZ shown in FIG. 1 is a three-dimensional robot coordinate system.

A reference point 9 is set in advance on the object 2 carried by the robot 4. On the other hand, the target point 10 as the destination of the reference point 9 is set in advance at the destination of the object 2. These reference points 9 and target points 10 are set to determine the position and posture of the object 2 at the destination. As the reference point 9 and the target point 10, a color marker, a two-dimensional bar code, or the like is used.

As shown in FIG. 2, the robot control device 1 of the present embodiment mainly includes an image pickup means 11 for capturing a reference point 9 and a target point 10 and a control means 12 for controlling the displacement of the joint 3 of the robot 4. Prepare as. In addition, the robot control device 1 includes a displacement amount detecting means 13 for detecting the displacement amount of the joint 3 of the robot 4.

The image pickup means 11 is a means for acquiring an image of the object 2 and the moving destination of the object 2, and is referred to as a camera in this embodiment. In this embodiment, two cameras 11 are used, and the cameras 11 are arranged so that the image planes IM1 and IM2 intersect each other. Typically, one camera 11 is arranged so that it can photograph the YZ plane of the robot coordinate system, and the other camera 11 is arranged so that it can photograph the ZX plane of the robot coordinate system. As shown in FIG. 1, one camera 11 is provided with a three-dimensional camera coordinate system including a U1 axis, a V1 axis, and a W1 axis that are orthogonal to each other, and the other camera 11 is orthogonal to each other. A three-dimensional camera coordinate system including a U2 axis, a V2 axis, and a W2 axis is provided. As described above, each camera 11 is arranged so that the image planes IM1 and IM2 intersect each other, and can photograph the object 2 and its moving destination.

The control means 12 is a means for controlling the displacement of the joint 3 so that the reference point 9 moves to the target point 10. The control means 12 controls the displacement of the joint 3 based on the image pickup result obtained by the image pickup means 11. Specifically, in this embodiment, the image pickup means 11 and the robot 4 are connected to the control means 12, and the control means 12 enables control of the robot described later. At this time, the camera 11 and the robot 4 are not calibrated.

Further, the displacement amount detecting means 13 is connected to the control means 12. The displacement amount detecting means 13 is an encoder in this embodiment and is provided in each joint 3. The encoder 13 of this embodiment detects the rotational position of the motor 8, and can obtain the position of the current hand portion 6 of the robot 4 and the position of the object 2 gripped by the hand portion 6 from the detection result. ..

Next, a robot control method using the robot control device 1 of this embodiment will be described. FIG. 3 is a flowchart showing an embodiment of the robot control method of the present invention. Further, FIG. 4 is a schematic diagram showing a usage state 1 of the robot control device of this embodiment, and shows a state in which an object is moved from the state of FIG.

As shown in FIG. 3, in this embodiment, the setting step S1, the imaging step S2, the detection step S3, the deviation calculation step S4, the position setting step S5, and the control step S6 are sequentially executed. The robot control method is not limited to this, and may be a conventionally known method for controlling the robot 4 based on an captured image and which does not require calibration of the camera 11 and the robot 4. Just do it.

First, the reference point 9 is set in the object 2, while the target point 10 is set in the movement destination of the object 2 (step S1). As described above, this setting is made by attaching a marker to the object 2 and the moving destination of the object 2. Then, the set reference point 9 and the target point 10 are imaged (step S2). This is done by two cameras 11. Next, the arm 5 of the robot 4 is measured, and the position of the hand portion 6 of the robot 4 and the position of the reference point 9 of the object 2 are obtained (step S3). This is done by the encoder 13 provided in the joint 3. At this time, the hand portion 6 of the robot 4 is in a state of sandwiching the object 2 at the position before the movement.

Then, the position deviation between the reference point 9 imaged by the camera 11 and the target point 10 imaged by the camera 11 is calculated, and this position deviation is converted from the camera coordinate system to the robot coordinate system (step). S4). The virtual target point of the object 2 is set by adding the position deviation converted from the camera coordinate system to the robot coordinate system to the position of the reference point 9 obtained by the encoder 13 (step S5). The arm 5 is controlled so that the reference point 9 moves to the set virtual target point (step S6). Specifically, each step is sequentially executed as shown below.

Here, a case where the object 2 has the first region 14 while the destination of the object 2 has the second region 15 will be described. The outer shape of the second region 15 is formed to have the same outer shape as the outer shape of the first region 14. In this case, the control is such that the object 2 is moved so that the edge portion of the first region 14 is brought into contact with the edge portion of the second region 15 so that the outer shape matches the outer shape. In this control, a plurality of reference points 9 are set on the object 2.

The plurality of reference points 9 are set at the edge of the first region 14 of the object 2. In this case, for example, among the plurality of reference points 9, a plurality of reference points 9 are set on the same straight line, and the remaining reference points 9 are set at positions deviating from the same straight line. The target point 10 is the destination of the reference point 9, and is set at a position corresponding to the reference point 9 at the destination so that the position and posture of the object 2 at the destination can be determined.

Typically, the examples shown in FIGS. 1 and 4 are used. In this example, the object 2 is a rectangular parallelepiped whose lower surface is the first region 14, and the second region 15 is a flat surface portion having the same shape as the lower surface of the rectangular parallelepiped. Here, reference points a, b, and c are set at the three lower vertices of the object 2, that is, the three corners of the lower surface of the object 2, and the three corners of the planar second region 15 are set. At the time of the contact, the target points a _d , b _d , and cd corresponding to the reference points a, b, and _c are set. Then, the three reference points a, b, c and the three target points a _d , b _d , and cd are imaged by the two cameras 11, and the three reference points a, b, and _c are imaged based on the imaging results. The displacement of the joint 3 of the robot 4 is controlled so as to move to the corresponding target points a _d , _{b d} , and cd, respectively. That is, the robot 4 is controlled so that the reference point a moves to the target point a _d , the reference point b moves to the target point b _d , and the reference point _c moves to the target point cd. At this time, either of the two cameras 11 is arranged at a position where the reference points a, b, _c and the target points a _d , b _d , and cd can be imaged.

A case where the reference point _b moves to the target point bd among the three reference points a, b, and c will be described. When viewed from the 11 side of the camera (referred to as camera C1) that captures the YZ plane, the reference point b is b = (b _1y , b _1z ), and the target point b _d is b _d = (b _1dy , b _1ds ). It is said that. Further, when viewed from the 11 side of the camera (referred to as camera C2) for photographing the ZX plane, the reference point b is b = (b _2x , b _2z ), and the target point b _d is b _d = (b _2dx , b _2dz). ). The numbers 1 and 2 in the coordinates indicate that they are viewed from the cameras C1 and C2, respectively.

Therefore, the deviation in the X-axis direction, the position deviation in the Y-axis direction, and the position deviation in the Z-axis direction between the reference point b and the target point b _d are expressed by the following equations, respectively. Since the Z-axis direction can be seen from the camera C1 and the camera C2, either the case seen from the camera C1 or the case seen from the camera C2 is used as described below.

This is regarded as the position deviation between the reference point b and the target point b _d , and the displacement of the joint 3 of the robot 4 is controlled based on this position deviation. Specifically, when the reference point 9 and the target point 10 are one point, the control is performed as shown below.

The control means 12 calculates a position deviation between the reference point 9 of the object 2 photographed by the camera 11 and the target point 10 photographed by the camera 11, and this position deviation is transferred from the camera coordinate system to the robot coordinate system. Be converted. Further, the control means 12 calculates a reference error between the true position of the reference point 9 of the object 2 and the position of the reference point 9 taken by the camera 11. The control means 12 is connected to two cameras 11, and can acquire the position information of the reference point 9 of the object 2 from these cameras 11. The control means 12 can calculate a reference error, which is an error between the acquired position of the reference point 9 and the true position of the reference point 9 of the object 2. Here, the reference error will be described. When the reference point 9 of the object 2 is captured from the camera 11, an error occurs between the camera 11 and the true position of the reference point 9 due to the distortion of the camera 11, which is the reference error. It is said that.

Further, the control means 12 can calculate a target error between the true position of the target point 10 set at the destination of the moving object 2 and the position of the target point 10 photographed by the camera 11. The control means 12 is connected to the camera 11 as described above, and can calculate a target error which is an error between the position of the target point 10 from the camera 11 and the true position of the target point 10. This target error is an error that occurs between the target point 10 and the true position, which is the actual position, when the target point 10 is captured from the camera 11 due to a cause such as distortion of the camera 11.

The true position of the reference point 9 is x = (x, y, z), the true position of the target point 10 is x _d = (x _d , _{yd, z d )} , and the reference point 9 taken by the camera 11 The position is x _c = (x _c , y _c , z _c ), and the position of the target point 10 taken by the camera 11 is x _cd = (x _cd , y _cd , z _cd ). The position of the reference point 9 taken by the camera 11 and the position of the target point 10 taken by the camera 11 include a calibration error. At this time, the error Δx _c due to the calibration of the camera 11, that is, the reference error and the target error, which are deviations from the true value, are expressed by the following equations. In the following equation, the "w" provided on the left shoulder indicates that it is viewed from the reference world coordinate system. In this embodiment, the robot coordinate system is the world coordinate system. Further, "c" provided on the left shoulder indicates that it is viewed from the camera coordinate system.

Here, the tilde shows the value including the error, and the others are as follows.

In numbers 2, 3, and 4, the reference and target errors are converted from the camera coordinate system to the robot coordinate system. On the other hand, assuming that the current position of the reference point 9 measured by the encoder 13 is x _m = (x _m , _{ym, z m )} , the error Δx _m due to the calibration of the robot 4, that is, the deviation from the true value is It is expressed by the following formula.

Here, f (q) indicates a function for converting each joint angle to the position of the reference point 9. If the geometric information of the robot 4 is accurate and each link rod 7 has sufficient rigidity, Δx _m approaches zero. Similarly, assuming that the position of the target point 10 measured by the encoder 13 is x _md = (x _md , y _md , z _md ), the error in the vicinity of the position of the target point 10, that is, the deviation from the true value is as follows. Defined by an expression.

In this way, between the position of the reference point 9 measured by the encoder 13 and the position of the target point 10 measured by the encoder 13 and their respective true values, the cause is distortion or slack of the link rod 7. There will be an error.

Next, the error that occurs when controlling the position and posture of the object 2 is defined by the following equation. This is an error due to the influence of static friction and gravity.

Here, it is as follows.

The error represented by the equation 9 has a small influence of static friction and the like, and when each joint angle of the robot 4 can be accurately controlled to the target point 10, Δx _m2 = 0. Here, since the target is an industrial robot, this value is set to zero. However, it will be described during the expansion of the expression. As a result, when the target point described later is given, the position of the object 2 after control becomes as follows from the equation 9.

Further, the control means 12 sets a virtual target point at the movement destination of the object 2 by adding the position deviation obtained by converting the coordinate system as described above to the position of the reference point 9 detected by the encoder 13. be able to. The control means 12 is connected to the encoder 13 and can acquire the position information of the reference point 9 from the encoder 13. Then, by adding the above-mentioned position deviation to the position of the reference point 9 from the encoder 13, a virtual target point as a movement destination of the reference point 9 can be set.

Specifically, it is as follows. The virtual target point x _vd is set as follows, and the virtual target point x _vd becomes as follows by the number 2, the number 3 and the number 5.

Further, the control means 12 controls the displacement of the joint 3 of the robot 4 so that the reference point 9 moves to the set virtual target point. The control means 12 is connected to the motor 8, and by controlling the motor 8, the joint 3 and thus the link rod 7 can be rotated so that the reference point 9 reaches the set virtual target point.

In this way, the robot 4 is controlled so that one of the three reference points 9 and the three target points 10 moves to the one target point 10. The remaining two points may be controlled in the same manner as the above-mentioned control. As a result, the reference points a, b, and _c can be made to reach the corresponding target points a _d , b _d , and cd, respectively.

In the case of this embodiment, as described above, the reference points 9 are set at the three corners of the lower surface of the rectangular parallelepiped object 2. That is, two of the three reference points 9 are set on the same straight line, and the remaining one point is set at a position deviating from the same straight line. For example, if the reference points a and b are arranged on the same straight line, the reference point c is set to a position deviated from the same straight line. Therefore, according to the present embodiment, with a simple setting, the edge portion of the first region 14 is brought into accurate contact with the edge portion of the second region 15 so that the outer shape matches the edge portion of the second region 15, and the object 2 is placed at a predetermined position at the destination. Can be placed in a predetermined orientation.

By the way, in this embodiment, the control is performed based on the position deviation between the reference point 9 and the target point 10, but it is also possible to control based on the posture angle deviation between the reference point 9 and the target point 10. Specifically, it is as follows.

Here, a case where the reference point _b moves to the target point bd among the three reference points a, b, and c will be described. When viewed from the 11 side of the camera (referred to as camera C1) that captures the YZ plane, the reference point a is a = (a _1y , a _1z ) and the reference point b is b = (b _1y , b _1z ). The point a _d is a _d = (a _1dy , a _1ds ), and the target point b _d is b _d = (b _1dy , b _1ds ). Further, when viewed from the 11 side of the camera (referred to as camera C2) for photographing the ZX plane, the reference point b is b = (b _2x , b _2z ) and the reference point c is c = (c _2x , c _2z ). The target point b _d is b _d = (b _2dx , b _2dz ) and the target point c _d is c _d = (c _2dx , c _2dz ).

By the way, as described above, since either of the two cameras 11 is arranged at a position where the reference points a, b, _c and the target points a _d , b _d , and cd can be imaged, in addition to the above, The reference point c is c = (c _1y , c _1z ) and the target point c _d is c _d = (c _1dy , c _1ds ), or the reference point a is a = (a _2x , a _2z ) and the target. The point a _d is a _d = (a _2dx , a _2dz ). The numbers 1 and 2 in the coordinates indicate that they are viewed from the cameras C1 and C2, respectively.

Therefore, the attitude angle deviation between the reference point b and the target point b _d is expressed by the following three equations. Regarding Δγ _d , the camera C1 is placed at a position where the reference points a, b, _c and the target points a _d , b _d , and cd can be imaged, or the camera C2 is located at the reference points a, b, c. And since it differs depending on the position where the target points a _d , _{b d} , and cd can be imaged, either one may be used.

Here, when −π / 4 <θ <π / 4, θ≈tanθ, so that the equation 13 is expressed by the following equation.

The displacement of the joint 3 of the robot 4 is controlled based on this posture angle deviation. Specifically, when the reference point 9 and the target point 10 are one point, the control is performed as shown below.

The true posture angle of the reference point 9 of the object 2 is R, the true posture angle of the target point 10 to which the object 2 is moved is R _d , and the posture angle of the reference point 9 captured by the camera 11 is R _ca. Let R _cd be the posture angle of the target point 10 captured by the camera 11. The posture angle of the reference point 9 captured by the camera 11 and the posture angle of the target point 10 captured by the camera 11 include a calibration error. At this time, the error ΔR _c due to the calibration of the camera 11, that is, the deviation from the true value is expressed by the following equation. In the following equation, the "T" provided on the right shoulder represents the transposed matrix of the matrix to which it is attached.

Here, the tilde shows the value including the error, and the others are as follows.

In numbers 15, 16, and 17, each error is converted from the camera coordinate system to the robot coordinate system. On the other hand, assuming that the current posture angle of the reference point 9 measured by the encoder 13 is R _m , the error ΔR _m due to the calibration of the robot 4, that is, the deviation from the true value is expressed by the following equation.

Here, g (q) shows a function for converting each joint angle into the posture angle of the object 2. If the geometric information of the robot 4 is accurate and each link rod 7 has sufficient rigidity, ΔR _m approaches I. Similarly, the posture angle of the target point 10 measured by the encoder 13 is R _md , and the error in the vicinity of the target posture, that is, the deviation from the true value is defined by the following equation.

In this way, the link rod 7 is distorted or loosened between the posture angle of the reference point 9 measured by the encoder 13 and the posture angle of the target point 10 measured by the encoder 13 and their respective true values. There is an error that is the cause.

Next, the error that occurs when controlling the position and posture of the object 2 is defined by the following equation. This is an error due to the influence of static friction and gravity.

Here, it is as follows.

The error shown by the number 22 is ΔR _m2 = I when the influence of static friction and the like is small and each joint angle of the robot 4 can be accurately controlled to the target posture. Here, since an industrial robot is targeted, this value is set to I. However, it will be described during the expansion of the expression.

Then, the virtual target attitude angle _Rvd is set as follows. The displacement of the joint 3 is controlled so as to have this virtual target posture angle R _vd .

According to this embodiment, the final error Δx _fp when each step is sequentially executed once is as follows.

Here, if Δx _m2 = 0 is taken into consideration, the error Δx _fp becomes as follows.

As described above, in this embodiment, the calibration error affects as the difference between the error before control and the error after control. This indicates that the closer the error amount at each position before and after control is, the smaller the final error Δx _fp is.

On the other hand, conventionally, the error due to one execution when a certain target position is given is different from that of the present embodiment. Specifically, it is as follows. Conventionally, the error of the object 2 is Δx _cd + Δx _m2 . Assuming that the final error in the conventional case is _Δxf , the target value is as follows because the target position measured by the camera 11 is used.

Here, by using the number 3, Δx _f becomes as follows.

Further, considering Δx _m2 = 0, Δx _f is as follows.

This indicates that both the calibration errors of the camera 11 and the robot 4 remain in their original form. That is, the calibration error of the camera 11 and the robot 4 directly affects the final error. Therefore, the robot 4 and the camera 11 had to be calibrated with high accuracy. However, in this embodiment, since it is expressed by an equation such as the above-mentioned number 26, the calibration error of the camera 11 and the robot 4 does not directly affect the time required for the calibration of the camera 11 and the robot 4. The cost can be reduced.

Further, according to the present embodiment, in the attitude angle control, the final error ΔR _fp when the attitude angle control is executed once is as follows, as in the case of the position control.

Here, if ΔR _m2 = I, the error ΔR _fp is as follows.

As described above, it can be seen that, as for the posture angle error, the phase amount between the errors before and after the execution remains as the final error, as in the case of the position.

On the other hand, in the past, the final error ΔR _f is as follows.

Here, as in the case of position control, if ΔR _m2 = I, the result is as follows.

As described above, conventionally, the calibration error of the camera 11 and the robot 4 remains as it is, as in the case of the position control.

Further, according to this embodiment, each step except the setting step S1 can be sequentially and repeatedly executed. That is, after the control step S6 is completed, the imaging step S2 is performed again in that state, and the remaining steps are sequentially executed. The number of repetitions can be changed as appropriate. Specifically, it is repeatedly executed when the following conditions are satisfied. Here, assuming that the position of the object 2 before the execution of the robot 4 is x _α and the position of the object 2 after the execution of the robot 4 is x _β , the conditions are as follows.

More accurate control can be performed by repeatedly executing when this condition is satisfied. That is, each time each step except the setting step S1 is repeated, the true target position is approached, and the accuracy can be improved. This indicates a case where the distance between the position of the object 2 and the target position before the execution of the robot 4 is larger than the measurement error of the sensor such as the camera 11 or the encoder 13 depending on the location. That is, it can be seen that it is possible to achieve the same accuracy as the non-linear component of the measurement error of the sensor by repeating the execution. Further, with respect to the non-linear component, the closer the position is before and after the control, the smaller the amount of error, and the accuracy can be improved each time it is repeated.

Further, according to the present embodiment, the posture angle control can be repeatedly executed as in the case of the position control. Specifically, if the posture angle of the object 2 before the execution of the robot 4 is R _α and the posture of the object 2 after the execution is R _β , the error is reduced by repeated execution and the target value is converged. The conditions for this are as follows.

When this condition is satisfied, it can be converged to the target value by repeatedly executing it. The nature of this condition is similar to the number 34, which is the convergence condition in the case of position control.

Further, according to the present embodiment, since the robot 4 can be controlled by the above-mentioned control, it can be easily applied to a controller (control means) mounted on an existing industrial robot while suppressing changes in software. be able to. That is, it can be directly applied to an existing robot 4 operating in a factory or the like while suppressing changes in hardware or software, and the range of use can be expanded.

5 and 6 are diagrams showing a usage state 2 of the robot control device of FIG. 1, FIG. 5 is a schematic diagram, and FIG. 6 shows a state in which an object is moved from the state of FIG. 5 to its moving destination. It is a schematic diagram. The robot control device 1 here is basically the same as the above-mentioned example. Therefore, in the following, the differences between the two will be mainly described, and the corresponding parts will be described with the same reference numerals.

In the above-mentioned example, a plurality of reference points 9 are set at the edge of the first region 14, but here, at least one of the plurality of reference points 9 is set at the edge of the first region 14. The remaining reference point 9 is set in the object 2 at a place other than the edge of the first region 14. For example, of the plurality of reference points 9, at least one reference point 9 is set at the edge of the first region 14, and the remaining reference points 9 are set as one point, and the edge of the first region 14 in the object 2 is set. It may be set in a place other than. In this case, the destination of the object 2 has a prism 16 having a second region 15 on one end surface in the axial direction.

When there are a plurality of at least one reference points 9, the remaining reference points 9 are set in the object 2 so that the reference line segment connecting one of the reference points 9 and the remaining reference points 9 is as follows. Will be done. The reference line segment is perpendicular to one axial end surface of the prism body 16 at the time of contact between the edge portion of the first region 14 and the edge portion of the second region 15. On the other hand, when the edge of the first region 14 and the edge of the second region 15 come into contact with each other, the target point 10 is set at a position corresponding to the reference point 9 set at the edge of the first region 14. .. Further, two other target points 18 and 18 are set on one side portion 17 of the side portions between the axial end surface and the axial end surface of the prism body 16 so as to be separated from each other.

In this example, the control means 12 controls the displacement of the joint 3 so that the reference point 9 set at the edge of the first region 14 moves to the target point 10, and the two other target points 18, 18 The displacement of the joint 3 is controlled so that the direction of the target line segment connecting the spaces and the direction of the reference line segment are the same. The movement of the reference point 9 set at the edge of the first region 14 to the target point 10 is controlled in the same manner as described above.

Typically, the examples shown in FIGS. 5 and 6 are used. In this example, the target point 10 which is one end of the above-mentioned reference line segment and corresponds to the reference point 9 set in the first region 14, and the extension extending from the target point 10 in the same direction as the target line segment. Another target point 20 is set so that the line segment connecting the other target point 20 set on the output line 19 is as follows. The length of the line segment connecting the target point 10 and another target point 20 is the same as the length of the reference line segment.

In this case, the control means 12 moves the remaining reference point 9 described above to another target point 20 so that the direction of the reference line segment and the direction of the target line segment are the same. Is controlled. That is, the control means 12 controls the displacement of the joint 3 so that the reference point 9 set at the edge of the first region 14 moves to the target point 10, and the remaining reference point 9 is another target point 20. The displacement of the joint 3 is controlled so as to move to.

In the illustrated example, the object 2 is formed in the shape of a quadrangular thin plate, and the lower surface thereof is the first region 14. The lower surface of the object 2 is not the plate surface of the plate material, but a thick portion of the plate material, and has a rectangular shape. The second region 15 is a flat surface portion formed on the upper surface of the square columnar prism body 16 in the axial direction, and is formed in the same shape as the first region 14. Here, three reference points d, e, and f are set for the object 2. Of the three reference points d, e, f, the two reference points d, e correspond to the two lower vertices of the object 2, that is, both ends of one long side of the lower surface of the object 2. Set at the corners. The remaining reference points f are the upper vertices of the object 2, and the reference points e are set to the vertices corresponding to the set vertices in the vertical direction. Therefore, the reference line segment connecting the reference point e and the reference point f is a side portion along the vertical direction of the plate-shaped object 2. As described above, since the object 2 is controlled so that the lower surface of the object 2 abuts on the upper surface of the prism body 16, the reference line segment is perpendicular to the upper surface of the prism body 16 at the time of abutting.

On the other hand, target points d _d and ed corresponding to the reference points _d and e are set at the two corners of the second region 15 at the time of the contact. Specifically, the target points d _d and _ed are set at two corners corresponding to both ends of one long side of the second region 15 formed in a rectangular plane. Further, of the four side portions along the axial direction of the prismatic prism 16, two other target points g and h are set on one side portion 17 so as to be separated from each other in the vertical direction. Since the one side portion 17 is a side portion of the prism body 16, it is perpendicular to the upper surface of the prism body 16. Therefore, the target line segment connecting the two other target points g and h is perpendicular to the upper surface of the prism body 16.

Then, the two cameras 11 image the three reference points d, e, f, the two target points d _d , _ed , and the two other target points g, h, and the robot 4 is based on the image capture results. Is controlled. Specifically, first, an extension line 19 extending from the target point _ed in the same direction as the target line segment is drawn. Here, the target point _ed is the lower end of the reference line segment and is a target point corresponding to the reference point e set in the first region 14. After the extension line 19 is formed, another target point _fd is set on the extension line 19. Another target point f _d is set so that the length of the line segment connecting it and the target point _ed is the same as the length of the reference line segment. Then, the displacement of the joint 3 is controlled so that the reference points _d and e move to the target points d _d and ed, respectively, and the reference point f is controlled to move to another target point f _d . The movement of each reference point to the target point is performed in the same manner as described above. As a result, the thin plate-shaped object 2 can be moved so that the first region 14 which is the lower surface of the object 2 is accurately brought into contact with the second region 15 of the prism body 16. At this time, the edge portion of the first region 14 and the edge portion of the second region 15 are brought into contact with each other so that the outer shapes match.

7 to 9 are diagrams showing the usage state 3 of the robot control device of FIG. 1, FIG. 7 is a schematic diagram, and FIG. 8 is a schematic diagram showing the process of moving an object from the state of FIG. 7 to its moving destination. FIG. 9 is a schematic diagram showing a state in which an object is moved from the state of FIG. 7 to the moving destination via the state of FIG. The robot control device 1 here is basically the same as the examples shown in FIGS. 5 and 6. Therefore, in the following, the differences between the two will be mainly described, and the corresponding parts will be described with the same reference numerals.

In the examples shown in FIGS. 5 and 6, the joint 3 is moved so that the remaining reference point 9 moves to another target point 20 so that the direction of the reference line segment and the direction of the target line segment are the same. The displacement was controlled, but here, the displacement of the joint 3 of the robot 4 is based on a deviation vector corresponding to the length of a line segment drawn perpendicular to the extension line 19 from the remaining reference point 9. Is controlled. In the illustrated example, the object 2 and the first region 14 formed therein, and the prism 16 and the second region 15 formed therein are the same as in FIGS. 5 and 6. Further, as in the case of FIGS. 5 and 6, three reference points d, e, f, two target points d _d , _ed , and two other target points g, h are set.

Then, the two cameras 11 image the three reference points d, e, f, the two target points d _d , _ed , and the two other target points g, h, and the robot 4 is based on the image capture results. Is controlled. Specifically, first, the extension line 19 described above is drawn from the target point _ed . After the extension line 19 is formed, the displacement of the joint 3 is controlled so that the reference points _d and e move to the target points d _d and ed, respectively, and the joint 3 is drawn perpendicularly to the extension line 19 from the reference point f. The displacement of the joint 3 is controlled based on the deviation vector which is a perpendicular line. Here, as a method of controlling the robot 4 based on the deviation vector, a conventionally known method is used. The movement of the reference points _d and e to the target points _d and ed is performed in the same manner as described above. As a result, the thin plate-shaped object 2 can be moved so that the first region 14 which is the lower surface of the object 2 is accurately brought into contact with the second region 15 of the prism 16 so that the edges of the prism 16 are aligned with each other. can.

10 and 11 are diagrams showing a usage state 4 of the robot control device of FIG. 1, FIG. 10 is a schematic diagram, and FIG. 11 is a schematic diagram showing a state in which an object is moved from the state of FIG. .. The robot control device 1 here is basically the same as the examples shown in FIGS. 5 and 6. Therefore, in the following, the differences between the two will be mainly described, and the corresponding parts will be described with the same reference numerals.

In the examples shown in FIGS. 5 and 6, the at least one reference point 9 is a plurality, but here, the at least one reference point 9 is a single point. That is, one reference point 9 is set at the edge of the first region 14, and the remaining reference point 9 of one point is set at a location other than the first region 14 with respect to the object 2. At this time, the reference line segment connecting one reference point 9 and the remaining reference point 9 of one point becomes perpendicular to one axial end surface of the prism 16 at the time of contact between the first region 14 and the second region 15. The remaining reference point 9 of one point is set as described above.

Typically, the examples shown in FIGS. 10 and 11 are used. That is, in the examples shown in FIGS. 5 and 6, the object 2 is formed in the shape of a quadrangular thin plate, but here, it is formed in a columnar shape. In the columnar object 2, the lower surface of the circular shape is the first region 14. On the other hand, the second region 15 is a circular flat surface portion formed on the upper surface of the square columnar prism body 16 in the axial direction, and is formed in the same shape as the first region 14.

Here, two reference points i and j are set for the object 2. Of the two reference points i and j, the reference point i is set on the outer edge of the lower surface of the object 2, and the remaining reference point j is set on the outer wall surface of the cylindrical object 2. At this time, the remaining reference point j is set on a perpendicular line 21 drawn perpendicularly to the lower surface of the object 2 from the reference point i. Therefore, the reference line segment connecting the reference point i and the reference point j is controlled so that the lower surface of the object 2 abuts on the upper surface of the prism body 16 as described above, so that the reference line segment is a reference at the time of abutting. The line segment is perpendicular to the upper surface of the prism body 16. On the other hand, a target point id corresponding to the reference point i at the time of _contact is set on the outer edge portion of the circular second region 15. Further, as described above, among the four side portions along the axial direction of the square columnar prism body 16, on one side portion 17, two other target points g, h separated from each other in the vertical direction are formed. Is set.

Then, the two reference points i and j, the target point _id and the two other target points g and h are imaged by the two cameras 11, and the robot 4 is controlled based on the imaged image results. Specifically, first, an extension line 19 extending from the target point id in the same direction as the target line segment connecting the two other target points _g and h is drawn. Here, the target point _id is the lower end of the reference line segment and is the target point corresponding to the reference point i set in the first region 14. After the extension line 19 is formed, another target point _jd is set on the extension line 19. Another target point _jd is set so that the length of the line segment connecting it and the target point _id is the same as the length of the reference line segment. Then, the displacement of the joint 3 is controlled so that the reference point i moves to the target point _{id, and the reference point j is controlled to move to another target point j d .} The movement of each reference point to the target point is performed in the same manner as described above. As a result, the columnar object 2 can be moved so that the first region 14 which is the lower surface of the object 2 is accurately brought into contact with the second region 15 of the prism 16 so that the edges of the prism are aligned with each other. ..

12 to 14 are diagrams showing the usage state 5 of the robot control device of FIG. 1, FIG. 12 is a schematic diagram, and FIG. 13 is a schematic diagram showing the process of moving an object from the state of FIG. 12, FIG. 14; Is a schematic diagram showing a state in which an object is moved from the state of FIG. 12 to the state of FIG. 13. The robot control device 1 here is basically the same as the example shown in FIGS. 7 to 9. Therefore, in the following, the differences between the two will be mainly described, and the corresponding parts will be described with the same reference numerals.

In the examples shown in FIGS. 7 to 9, the object 2 is formed in the shape of a quadrangular thin plate, but here, it is formed in a columnar shape. In the columnar object 2, the lower surface of the circular shape is the first region 14. On the other hand, the second region 15 is a circular flat surface portion formed on the upper surface of the square columnar prism body 16 in the axial direction, and is formed in the same shape as the first region 14.

Here, two reference points i and j are set for the object 2. Of the two reference points i and j, the reference point i is set on the outer edge of the lower surface of the object 2, and the remaining reference point j is set on the outer wall surface of the cylindrical object 2. At this time, the remaining reference point j is set on a perpendicular line 21 drawn perpendicularly to the lower surface of the object 2 from the reference point i. Therefore, the reference line segment connecting the reference point i and the reference point j is controlled so that the lower surface of the object 2 abuts on the upper surface of the prism body 16 as described above, so that the reference line segment is a reference at the time of abutting. The line segment is perpendicular to the upper surface of the prism body 16. On the other hand, a target point id corresponding to the reference point i at the time of _contact is set on the outer edge portion of the circular second region 15. Further, as described above, among the four side portions along the axial direction of the square columnar prism body 16, on one side portion 17, two other target points g, h separated from each other in the vertical direction are formed. Is set. Since the one side portion 17 is a side portion of the prism body 16, it is perpendicular to the upper surface of the prism body 16. Therefore, the target line segment connecting the two other target points g and h is perpendicular to the upper surface of the prism body 16.

Then, the two reference points i and j, the target point _id and the two other target points g and h are imaged by the two cameras 11, and the robot 4 is controlled based on the imaged image results. Specifically, first, an extension line 19 extending from the target point _id in the same direction as the target line segment is drawn. Here, the target point _id is the lower end of the reference line segment and is the target point corresponding to the reference point i set in the first region 14. After the extension line 19 is formed, the displacement of the joint 3 is controlled so that the reference point i _moves to the target point id, and the deviation vector is a perpendicular line drawn perpendicularly to the extension line 19 from the reference point j. The displacement of the joint 3 is controlled based on the above. Here, as a method of controlling the robot 4 based on the deviation vector, a conventionally known method is used. The movement of the reference point _i to the target point id is performed in the same manner as described above. As a result, the columnar object 2 can be moved so that the first region 14 which is the lower surface of the object 2 is accurately brought into contact with the second region 15 of the prism 16 so that the edges of the prisms 2 are aligned with each other. can.

15 to 17 are diagrams showing a usage state 6 of the robot control device of FIG. 1, FIG. 15 is a schematic diagram, and FIG. 16 is a schematic diagram showing a process of moving an object from the state of FIG. 15, FIG. 17 Is a schematic diagram showing a state in which an object is moved from the state of FIG. 15 through the state of FIG. The robot control device 1 here is basically the same as the example shown in FIGS. 1 and 4. Therefore, in the following, the differences between the two will be mainly described, and the corresponding parts will be described with the same reference numerals. That is, in the examples shown in FIGS. 1 and 4, the edge of the first region 14 of the object 2 is brought into contact with the edge of the second region 15 to which the object 2 is moved so as to match the outer shape. , The object 2 is inserted into the destination of the object 2.

The object 2 has an insertion portion 22 having a first region 14 at the tip thereof. On the other hand, the destination of the object 2 has a second region 15, and the second region 15 is a insertion portion into which the insertion portion 22 of the object 2 is inserted. In this case, the object 2 is moved so that the edge of the first region 14 of the object 2 is brought into contact with the edge of the second region 15 so that the outer shape matches, and then the insertion portion 22 of the object 2 is covered. It is controlled to be inserted into the insertion unit 15. In this control, a plurality of reference points 9 are set on the object 2.

The plurality of reference points 9 are set at the edge of the first region 14 of the object 2. On the other hand, the target point 10 is the destination of the reference point 9, and is set at a position corresponding to the reference point 9 at the destination so that the position and posture of the object 2 at the destination can be determined. Here, another reference point 23 is further set in the object 2 at a position different from the reference point 9. The other reference points 23 are located on the proximal end side of the reference point 9 set in the first region 14 into the insertion portion 22 of the object 2 and after the insertion portion 22 is inserted into the insertion portion 15. The position corresponding to the target point 10 set in the two regions 15 is set. Then, after the contact between the edge of the first region 14 of the object 2 and the edge of the second region 15 by the control means 12, another reference point 23 moves to the target point 10 set in the second region 15. The displacement of the joint 3 is controlled so as to do so. The contact between the edge portion of the first region 14 and the edge portion of the second region 15 is performed in the same manner as described above.

Typically, it is an example shown in FIGS. 15 to 17. In this example, the object 2 is a rectangular parallelepiped having the lower surface as the first region 14 as described above, and the lower half is the insertion portion 22. On the other hand, the second region 15 is a recess that opens upward and is formed on the upper surface of the prism 16. The recess is formed in a shape into which the lower half of the object 2 can be inserted. Here, as described above, reference points a, b, and c are set at the three lower vertices of the object 2, that is, the three corners of the lower surface of the object 2, and the planar second region 15 is formed. Target points a _d , b _d , and cd corresponding to reference points a, b, and _c are set at the three corners at the time of contact.

Further, in the insertion portion 22 of the object 2, other reference points k, l, m are set on the base end side of the reference points a, b, and c, that is, on the upper end portion of the insertion portion 22. Specifically, the other reference point k is set at the center of the side portion connecting the apex where the reference point a is set and the apex above the apex. The other reference point l is set at the center of the side portion connecting the apex where the reference point b is set and the apex above the apex. The other reference point m is set at the center of the side portion connecting the apex where the reference point c is set and the apex above the apex.

Then, the three reference points a, b, c, the other three reference points k, l, m, and the three target points a _d , _{b d} , and cd are imaged by the two cameras 11, and the imaging results thereof. The robot 4 is controlled based on the above. Specifically, after the displacement of the joint 3 of the robot 4 is controlled so that the three reference points a, b, and _c move to the corresponding target points a _d , b _d , and cd, respectively, the other three reference points are used. The displacement of the joint 3 is controlled so that the points k, l, and m move to the target points a _d , _{b d} , and cd. As a result, the lower half of the object 2 can be accurately inserted into the recess to which the object 2 is moved. The movement of the three reference points a, b, and _c to the target points a _d , b _d , and cd is performed in the same manner as described above, and the other three reference points k, l, and m are targeted points. The movement to a _d , b _d , and cd is performed in the same manner as the movement of the three reference points a, b, and _c to the target points a _d , _{b d ,} and cd.

The robot control device and the robot control method of the present invention are not limited to the configurations shown in the above illustrated examples, and can be appropriately changed. For example, in the illustrated example, the reference point 9 and the target point 10 are set in the corresponding points in advance before the image pickup by the image pickup means 11, but each of these points is set on the image captured by the image pickup means 11. You may. FIG. 18 is a schematic block diagram showing another embodiment of the robot control device of the present invention. Here, the robot control device 1 includes an image pickup means 11, a setting means 24, and a control means 12, which will be described later. Since other configurations are the same as those shown in the above illustrated example, the description thereof will be omitted.

The image pickup means 11 is a means for capturing an image of the object 2 and the moving destination of the object 2, and is referred to as a camera in this embodiment. In this embodiment, two cameras 11 are arranged as described above. It should be noted that the reference point 9 and the target point 10 are not set for the object 2 photographed by the camera 11 and its moving destination.

The setting means 24 is a means for setting a reference point 9 or the like on the object 2 and setting a target point 10 or the like on the moving destination of the object 2 on the image obtained by the image pickup means 11. When setting the reference point 9, the target point 10, and the like, each of these points is set so that the position and posture of the object 2 at the destination can be determined. Typically, the reference point 9, the target point 10, and the like are set at the positions shown in the above-described illustrated examples on the image. As a method for setting each point on the image, a conventionally known method is used. For example, a reference point 9 and a target point 10 are set on the image based on a template image stored in advance in the control means 12 or the like.

The control means 12 is a means for controlling the displacement of the joint 3 so that the reference point 9 moves to the target point 10. The control means 12 controls the displacement of the joint 3 based on an image in which the reference point 9 and the target point 10 are set at predetermined positions by the setting means 24. In this embodiment, the control means 12 has the setting means 24 described above.

Next, a robot control method using the robot control device 1 of this embodiment will be described. In this embodiment, the imaging step, the setting step, and the control step are sequentially executed.

First, the object 2 and its moving destination are imaged. This is done by two cameras 11. Then, on the captured image, the reference point 9 is set for the object 2, while the target point 10 is set for the movement destination of the object 2. Then, the robot 4 is controlled based on the image in which the reference point 9 and the target point 10 are set. For this control, a conventionally known control method for controlling the robot 4 based on an image is used.

Further, in the examples shown in FIGS. 1 and 4, three reference points 9 and three target points 10 are set, but the present invention is not limited to these, and four reference points 9 and four target points 10 are set. It may be set in a predetermined place. FIG. 19 is a schematic diagram showing another usage state of the robot control device of FIG.

Here, two reference points 9 and 9 are set on one long side of the lower surface of the object 2, and two reference points 9 and 9 are set on one short side of the lower surface of the object 2. To. On the other hand, in the second region 15, four target points 10 corresponding to each of the four reference points 9 are set at the time of contact. Then, the control means 12 controls the displacement of the joint 3 of the robot 4 so that the four reference points 9 move to the corresponding four target points 10.

Further, the control method is not limited to the method described above, and may be, for example, a method disclosed in International Publication No. WO2013 / 176212. Further, the shapes of the object 2, the first region 14 and the second region 15 are not limited to the above illustrated example, and can be appropriately changed. At this time, the positions and numbers of the reference point 9 and the target point 10 are appropriately set.

The present invention can be suitably used for a robot that carries an object to its destination.

1 Robot control device 2 Object 3 Joint 4 Robot 9 Reference point 10 Target point 11 Camera (imaging means)
12 Control means 14 First area 15 Second area 16 Prism 17 Sides 18 Other target points 19 Extension line 20 Another target point 22 Insertion part 23 Other reference points 24 Setting means

Claims (5)

A control device for a robot that moves an object by displacement of a joint, and a reference point is set for the object in order to determine the position and posture of the object at the destination, while the reference point of the reference point is set. A target point as a moving destination is set, and the image pickup result obtained by the image pickup means without calibrating the reference point and the image pickup means for imaging the target point, and the robot and the image pickup means. Based on this, the control means for controlling the displacement of the joint so that the reference point moves to the target point is provided.
The object has a first region, the destination of the object has a second region having the same shape as the outer shape of the first region, and the object is moved to the edge of the first region. In the control of bringing the portion into contact with the edge portion of the second region so that the outer shape matches, a plurality of the reference points are set on the object, and the plurality of reference points are set on the edge portion of the first region. Alternatively, at least one of the plurality of reference points is set at the edge of the first region, and the remaining reference points are set at the object other than the edge of the first region.
Of the plurality of reference points, at least one reference point is set at the edge of the first region, and the remaining reference points are set as one point, and the object has a location other than the edge of the first region. When set to, when the destination of the object has a prism having the second region on one end surface in the axial direction and the at least one reference point is one point, the one reference point and the said The remaining reference points are set on the object so that the reference line segment connecting the remaining reference points is perpendicular to one axial end surface of the prism at the time of contact, and the at least one reference point is set. In the case of a plurality, the remaining reference points are said to be perpendicular to one axial end surface of the prism at the time of contact. Set on the object,
Two other target points are set on one side of the side between the axial end surface and the axial end surface of the square column, and the control means is the first. The displacement of the joint is controlled so that the reference point set at the edge of the region moves to the target point, and the direction of the target line segment connecting the two other target points and the direction of the reference line segment. A robot control device for setting control points of a robot, characterized in that the displacement of the joint is controlled so as to be the same as the above . It is a control device of a robot that moves an object by displacement of a joint, and is an image pickup means for capturing an image of the object and the moving destination of the object, and on an image obtained by the image pickup means, the object is attached to the object. A setting means for setting a reference point and a target point to which the reference point is moved to the destination of the object so that the position and posture of the object at the destination can be determined, the robot, and the image pickup. A control means for controlling the displacement of the joint so that the reference point moves to the target point based on the image in which the reference point and the target point are set by the setting means without calibrating the means. , Equipped with
The object has a first region, the destination of the object has a second region having the same shape as the outer shape of the first region, and the object is moved to the edge of the first region. In the control of bringing the portion into contact with the edge portion of the second region so that the outer shape matches, a plurality of the reference points are set on the object, and the plurality of reference points are set on the edge portion of the first region. Alternatively, at least one of the plurality of reference points is set at the edge of the first region, and the remaining reference points are set at the object other than the edge of the first region.
Of the plurality of reference points, at least one reference point is set at the edge of the first region, and the remaining reference points are set as one point, and the object has a location other than the edge of the first region. When set to, when the destination of the object has a prism having the second region on one end surface in the axial direction and the at least one reference point is one point, the one reference point and the said The remaining reference points are set on the object so that the reference line segment connecting the remaining reference points is perpendicular to one axial end surface of the prism at the time of contact, and the at least one reference point is set. In the case of a plurality, the remaining reference points are said to be perpendicular to one axial end surface of the prism at the time of contact. Set on the object,
Two other target points are set on one side of the side between the axial end surface and the axial end surface of the prism, and the control means is the first. The displacement of the joint is controlled so that the reference point set at the edge of the region moves to the target point, and the direction of the target line segment connecting the two other target points and the direction of the reference line segment. Controlling the displacement of the joint so that
A robot control device that sets control points for robots. A line connecting the target point corresponding to the reference point, which is one end of the reference line segment, and another target point set on an extension line extending from the target point in the same direction as the target line segment. The other target point is set so that the length of the minute is the same as the length of the reference line segment.
The control means controls the displacement of the joint so that the remaining reference points move to the other target points so that the direction of the reference line segment and the direction of the target line segment are the same. matter,
The robot control device for which the control point of the robot according to claim 1 or 2 is set. The control means has the target line from the target point corresponding to the reference point, which is one end of the reference line segment, so that the direction of the reference line segment and the direction of the target line segment are the same. Controlling the displacement of the joint based on a deviation vector, which is a deviation corresponding to the length of the line segment drawn perpendicular to the remaining reference point with respect to the extending line extending in the same direction as the minute.
The robot control device for which the control point of the robot according to claim 1 or 2 is set. The object has an insertion portion having the first region at the tip thereof, and the second region is an insertion portion into which the insertion portion of the object is inserted, and the object is moved. In the control of inserting the insertion portion into the insertion portion after the edge portion of the first region is brought into contact with the edge portion of the second region so as to match the outer shape.
At the base end side of the reference point set in the first region, at the insertion portion, and at the target point set in the second region after the insertion portion is inserted into the insertion portion. Other corresponding reference points are set further and
After the contact between the edge of the first region and the edge of the second region, the control means displaces the joint so that the other reference point moves to the target point set in the second region. Control
The robot control device for which the control point of the robot according to any one of claims 1 to 4 is set.

Images