Control method and control device for robot

Abstract

To reduce the time to complete the movement between access points, of a robot comprising a rotatable swing arm, a common arm installed at the swing arm and a hand installed at both ends of the common arm through a tip arm.SOLUTION: A control method for a robot according to the present invention performs control so as to: when the robot moves so as to rotate a swing arm 21 around an axis T from a starting point access point where hands 25, 26 overlap each other and are oriented in a first direction to an end point access point where the hands 25, 26 overlap each other and are oriented in a second direction, move the hand 26 by a continuous path (CP) operation in which a tip of the hand 26 is located on a path L and a change in the orientation of the hand 26 is designated, and a tip of the other one hand 25 is located on the path L.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to the control of an industrial robot (hereinafter referred to as a robot) suitable for transporting a workpiece, and more particularly to a control method and a control device capable of operating the robot at high speed while preventing interference with a wall surface or the like disposed around the robot.

Background Art

[0002] In the manufacture of liquid crystal display panels and organic EL (electroluminescence) display panels, it is necessary to transport a glass substrate as a workpiece via a transfer chamber between process chambers that execute respective processes. For a robot used for such a purpose, as described in Patent Document 1, for example, there is a robot having a swing arm whose one end is connected to a main body (i.e., a base) and swings in a horizontal plane, an arm group attached to the other end of the swing arm, i.e., the tip of the swing arm, and a hand attached to the tip of each arm for holding a workpiece. In particular, Patent Documents 2 and 3 disclose a robot in which a common arm bent in an L shape or a V shape is attached to the tip of the swing arm at its central portion, and hands are attached to both ends of the common arm via tip arms. The workpiece is placed on the hand and transported between process chambers. Further, these robots are configured such that the hand moves on a straight line connecting the center of the hand and the tip of the swing arm regardless of the direction of the swing arm. Therefore, by driving the arm group, the hand expands and contracts so as to move away from or approach the position of the tip of the swing arm with respect to the position of the tip of the swing arm. The transfer chamber is adjacent to the process chamber such that the loading / unloading port of the process chamber opens into the transfer chamber. The robot first moves the hand to an access point which is a position in front of the process chamber, and then the hand is inserted into and removed from the loading / unloading port of the process chamber from the access point, whereby the workpiece can be placed in the process chamber or taken out from the process chamber.

Prior Art Documents

Patent Document

[0003]

Patent Document 1

Patent Document 2

Patent Document 3

Summary of the Invention

Problems to be Solved by the Invention

[0004] When transferring a workpiece between process chambers using the robots for conveyance described in Patent Documents 1 - 3, the swing arm of the robot also rotates according to the positions of the respective process chambers that are the starting point and the ending point. On the other hand, in order to reduce the moving distance when inserting and removing the hand with respect to the process chamber, the size of the transfer chamber is made as small as possible. Therefore, when moving the robot from one posture to another, there is a risk that the hand or the workpiece placed on the hand may collide with the wall surface of the transfer chamber. Conventionally, in order to prevent the hand or the workpiece from colliding with or interfering with the surrounding objects of the robot, such as the wall surface, during the movement of the robot, when moving the robot, first, while fixing the swing arm, the common arm is rotated to change the position and orientation of the hand so that the hand does not protrude greatly in the outer direction of the rotation of the swing arm. Then, while keeping the orientation of the hand with respect to the surrounding space, that is, the transfer chamber unchanged, the swing arm is rotated by a desired angle, and after the rotation of the swing arm is completed, the robot is controlled so that the position and orientation of the hand become the desired ones. However, when controlling the robot in this way, since the operation of changing the orientation of the hand relative to the transfer chamber and the operation of rotating the swing arm itself are performed in separate periods, as a whole, there is a problem that the time until the movement is completed becomes long.

[0005] An object of the present invention is to provide a robot control method and a robot control device that can shorten the time until the completion of the movement of the robot between access points by simultaneously performing an operation of rotating the swing arm itself while preventing collisions and an operation of relatively moving the orientation of the hand with respect to the surrounding space.

Means for Solving the Problems

[0006] The robot targeted by the robot control method and control device according to the present invention includes, for example, a swing arm having a base end connected to a first axis and rotatable around the first axis, and a common arm connected to the tip of the swing arm via a second axis and rotatable around the second axis. A first tip arm having a base end connected to one tip of the common arm via a third axis and rotatable around the third axis, and a base end connected to the other tip of the common arm via a fourth axis and rotatable around the fourth axis. A second tip arm, a first hand having a base end connected to the tip of the first tip arm via a fifth axis and rotating around the fifth axis in conjunction with the rotation of the first tip arm with respect to the common arm around the third axis, and a base end connected to the tip of the second tip arm via a sixth axis. And a second hand that rotates around the sixth axis in conjunction with the rotation of the second tip arm with respect to the common arm around the fourth axis, and the first to sixth axes are parallel to each other.

[0007] In the robot control method of the first aspect, when moving the robot by rotating the swing arm around the first axis from a starting access point where the first hand and the second hand overlap each other and face the first direction to an ending access point where the first hand and the second hand overlap each other and face the second direction, the control target point of the first hand is on a predetermined path and the change in the orientation of the first hand is specified. Control is performed to move the first hand by a continuous path (CP) operation, and at the same time, control is performed so that the control target point of the second hand is on the above-mentioned predetermined path.

[0008] By performing such control, since the first hand and the second hand move along a predetermined path during the rotation of the swing arm, it is possible to prevent the hands, the work on the hands, etc. from colliding with objects such as the surrounding wall surface, while quickly moving the entire robot from the starting access point to the ending access point. When the first direction and the second direction are the same direction, it is preferable not to change the orientation of the first hand in the CP operation. When the first direction and the second direction are different, in the CP operation, the orientation of the first hand may be changed from the first direction to the second direction at a uniform speed, or assuming that the swing arm is rotated at a uniform speed, the orientation of the first hand may also be changed according to the change in the rotation angle of the swing arm.

[0009] In the robot control method of the second aspect, when moving the robot by rotating the swing arm around the first axis from the starting access point where the first hand and the second hand overlap each other and face the first direction to the ending access point where the first hand and the second hand overlap each other and face the second direction, one or more waypoints are defined so that the control target point of the first hand and the control target point of the second hand exist on a predetermined path, and the robot is moved by a point-to-point (PTP) operation from the starting access point to the ending access point via the one or more waypoints.

[0010] When performing such control, since the robot is moved by a PTP operation using, as waypoints, one or more points known not to cause collisions, it is possible to prevent the hands and the work from colliding with the wall surface etc. around the robot, while quickly moving the robot from the starting access point to the ending access point.

[0011] When performing the PTP operation, in accordance with the angular difference in the direction of the common arm between the starting access point and the ending access point, the common arm rotates relative to the swing arm around the second axis in proportion to the rotation amount of the swing arm around the first axis, under the first constraint condition, and under the second constraint condition that the change in the rotation angle of the swing arm around the first axis, the change in the rotation angle of the first tip arm around the third axis, and the change in the rotation angle of the second tip arm around the fourth axis are interlocked. It is preferable to execute the PTP operation. By introducing such constraint conditions, the degree of freedom in control is reduced, so it is more reliably prevented that the robot 10 moves along a trajectory that may cause a collision.

[0012] In the robot control method of each of the above-described aspects, for each of the first hand and the second hand, a control target point is defined as a point on the hand or on an object mounted on the hand that may potentially collide with an object around the robot as the robot moves from the starting access point to the ending access point, and it is preferable to define a predetermined path as a path along which the control target point does not collide with an object around the robot. By defining the control target point and the predetermined path in this way, the occurrence of a collision in the robot can be more reliably prevented. Also, when the first direction and the second direction are in the same direction, it is preferable to use a straight line as the predetermined path, and when the second direction is orthogonal to the first direction, it is preferable to use a curve such as an arc as the predetermined path. By using such a predetermined path, the amount of computation required for controlling the robot can be reduced.

[0013] In the robot to which the robot control method of each of the above-described aspects is applied, the common arm is, for example, an arm bent at a position held by the swing arm via the second axis. Also, in this robot, the distance between the second axis and the third axis is equal to the distance between the third axis and the fifth axis, and the distance between the second axis and the fourth axis is equal to the distance between the fourth axis and the sixth axis. Such a robot is suitable for transporting workpieces in a relatively narrow transfer chamber. However, by applying the robot control method of each of the above-described aspects, it becomes possible to make the transfer chamber narrower while avoiding the occurrence of collisions.

[0014] The robot control device of the first aspect includes a storage unit that stores parameters necessary for controlling the robot, and from a starting access point where the first hand and the second hand overlap each other and face the first direction, to an ending access point where the first hand and the second hand overlap each other and face the second direction, when the robot is moved by rotating the swing arm around the first axis, by referring to the parameters in the storage unit and performing calculations, the control target point of the first hand is on a predetermined path and the first hand moves by a CP operation in which the change in the direction of the first hand is specified, and at the same time, the first to fourth axes are controlled so that the control target point of the second hand is on the aforementioned predetermined path. By using such a control device, since the first hand and the second hand move along a predetermined path during the rotation of the swing arm in the robot, it is possible to prevent the hand, the workpiece on the hand, etc. from colliding with an object such as the surrounding wall surface, and at the same time, the robot can be quickly moved from the starting access point to the ending access point as a whole.

[0015] The robot control device according to the second aspect includes a storage unit that stores parameters necessary for controlling the robot, and from a starting access point where the first hand and the second hand overlap each other and face the first direction, to an ending access point where the first hand and the second hand overlap each other and face the second direction, when rotating the swing arm around the first axis to move the robot, by performing calculations with reference to the parameters in the storage unit, one or more waypoints are determined so that the control target points of the first hand and the second hand exist on a predetermined path, and the first to fourth axes are controlled so that the robot moves in a PTP operation from the starting access point to the ending access point via the one or more waypoints, and a calculation unit.

Advantages of the Invention

[0016] According to the present invention, when moving the robot between access points, it becomes possible to simultaneously execute an operation of rotating the swing arm itself and an operation of relatively moving the hand with respect to the surrounding space while preventing collisions and the like, so that the time until the completion of the movement of the robot can be shortened.

Brief Description of the Drawings

[0017]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Embodiments for Carrying Out the Invention

[0018] Next, embodiments for carrying out the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing an example of a robot to which the control method according to an embodiment of the present invention is applied, where (a) is a front view, (b) is a plan view, and (c) is a schematic perspective view. In FIG. 1, (a), (b), and (c) correspond to different postures of the robot 10. The arrows indicated by X, Y, and Z in the figure indicate the directions of the X-axis, Y-axis, and Z-axis in a three-dimensional orthogonal coordinate system defined with respect to the robot 10 as a world coordinate system. The Z-axis is a vertical axis. Therefore, the XY plane is a horizontal plane.

[0019] The robot 10 shown in FIG. 1 is a horizontal articulated robot also called a translation type, and is the same as the robots described in Patent Documents 2 and 3. This robot 10 is used for transporting plate-shaped workpieces such as glass substrates. The robot 10 includes a base 20, a swing arm 21 having a base end connected to the base 20, a common arm 22 connected to the tip of the swing arm 21, two tip arms 23 and 24 respectively connected to the common arm 22, and two hands 25 and 26 respectively connected to the tip arms 23 and 24. The hands 25 and 26 are elongated in shape. The swing arm 21 can move up and down in the Z-axis direction, that is, in the height direction, by an elevating mechanism 27 provided in the base 20, and is driven by a motor (not shown) provided in the elevating mechanism 27 to rotate within a horizontal plane about the base end of the swing arm 21, that is, the position where the swing arm 21 is connected to the base 20. The center of rotation of the swing arm 21 within the horizontal plane is defined as the T-axis. The T-axis corresponds to the first axis.

[0020] The common arm 22 is an arm bent in an L shape or a V shape (i.e., a boomerang shape), which is connected to the other end of the swing arm 21 at the bent position, and is driven by a motor (not shown) built into the swing arm 21 to be rotatable in a horizontal plane. The axis of this rotation is defined as the TH axis. The TH axis corresponds to the second axis. The L-shaped common arm 22 has two ends, with the tip arm 23 connected to one end and the tip arm 24 connected to the other end.

[0021] One tip arm 23 is connected to the common arm 22 at its proximal end side and is driven by a motor (not shown) built into the common arm 22 to be rotatable in a horizontal plane. The axis of this rotation of the tip arm 23 is defined as the RR axis. One hand 25 is connected to the tip of one tip arm 23 and is driven via a link mechanism (not shown) by a motor (not shown) that rotationally drives the tip arm 23 to be rotatable in a horizontal plane. The link mechanism includes, for example, a pair of pulleys and a belt stretched between these pulleys. When the tip arm 23 rotates by an angle θ with respect to the common arm 22, the hand 25 is configured to rotate by an angle 2θ with respect to the tip arm 23. The axis of this rotation of one hand 25 is defined as the AR axis. The distance between the TH axis and the RR axis and the distance between the RR axis and the AR axis (i.e., the length of one tip arm 23) in the horizontal plane are equal, forming an isosceles triangle. Since the relationship between the rotation angle of the tip arm 23 with respect to the common arm 22 and the rotation angle of the hand 25 with respect to the tip arm 23 is defined as described above and an isosceles triangle is formed, a certain reference line fixed to the hand 25 will always pass through the TH axis regardless of the rotation angle of the tip arm 23 during its rotational drive. In a normal case, this reference line is the longitudinal center line of the hand 25. That is, the hand 25 will expand and contract while satisfying the condition that its longitudinal center line passes through the TH axis. If such movement of the hand 25 can be realized, the hand 25 may be rotationally driven using a motor (not shown) provided inside the tip arm 23 instead of the link mechanism.

[0022] Similarly, the other tip arm 24 is connected to the common arm 22 at its base end side and is driven by a motor (not shown) built in the common arm 22 to be rotatable around the RL axis in the horizontal plane. The other hand 26 is connected to the tip of the other tip arm 24 and is driven via a link mechanism (not shown) by a motor (not shown) that rotationally drives the tip arm 24 to be rotatable around the AL axis in the horizontal plane. The same link mechanism as described above is used, and a motor (not shown) that rotationally drives the hand 26 may be provided in the tip arm 24 instead of the link mechanism. In the horizontal plane, the distance between the TH axis and the RL axis is equal to the distance between the RL axis and the AL axis, forming an isosceles triangle. However, the length of the other tip arm 24, that is, the distance between the RL axis and the AL axis, may be different from the length of the one tip arm 23, that is, the distance between the RR axis and the AR axis. Regarding the other hand 26 as well, a certain reference line fixed to this hand 26 will always pass through the TH axis regardless of the rotation drive angle of the tip arm 24. In the normal case, this reference line is the longitudinal center line of the hand 26. That is, the other hand 26 will also expand and contract while satisfying the condition that its longitudinal center line passes through the TH axis. One of the RR axis and the RL axis corresponds to the third axis, and the other corresponds to the fourth axis. Similarly, one of the AR axis and the AL axis corresponds to the fifth axis, and the other corresponds to the sixth axis.

[0023] In the robot 10 shown in FIG. 1, in order to prevent interference between the tip arms 23 and 24 and interference between the hands 25 and 26, the rotation planes formed by the rotation of each of the tip arms 23 and 24 and the hands 25 and 26 are arranged in the order of the rotation plane of one tip arm 23, the rotation plane of one hand 25, the rotation plane of the other tip arm 24, and the rotation plane of the other hand 26 from the lower side in the height direction. These rotation planes are all horizontal. In this robot 10, with the tip arms 23 and 24 and the hands 25 and 26 folded, the swing arm 21 extends in the Y direction, both hands 25 and 26 extend in the Y direction so as to overlap on the swing arm 21, and the posture in which the AR axis and the AL axis are near the T axis on the straight line connecting the T axis and the TH axis is the posture of the origin position. Hereinafter, the direction in which the hands 25 and 26 extend from the AR axis and the AL axis will be simply referred to as the direction or orientation of the hands 25 and 26.

[0024] As shown in FIG. 1(a), a control device (robot controller) 50 for controlling and driving the motors of each axis of the robot 10 is connected to the robot 10. The control device 50 includes a storage unit 51 that stores various parameters necessary for controlling the robot 10, and an arithmetic unit 52 that performs arithmetic operations necessary for controlling each motor while referring to the parameters stored in the storage unit 51. In particular, the arithmetic unit 52 drives the motors of the T axis, the TH axis, the RR axis, and the RL axis so as to perform the control described later by performing arithmetic operations with reference to the parameters in the storage unit 51.

[0025] The robot 10 shown in Fig. 1 is arranged in the transfer chamber and used for transferring workpieces. Figs. 2(a) to (c) are schematic plan views showing examples of the postures of the robot 10, and show the arrangement of the robot 10 in the translation chamber, the positions of the access points, the access to the process chamber, etc. Consider a case where the robot 10 is arranged in a transfer chamber, which is a space surrounded by a pair of wall surfaces extending in the X direction, and the robot 10 is used to transfer workpieces between two process chambers. In the robot 10 of the present embodiment, the TH axis exists on the longitudinal centerlines of the hands 25 and 26. However, in Fig. 2, for the sake of explanation, the hands 25 and 26 are drawn so that the longitudinal centerlines of the hands 25 and 26 are displaced from the position of the TH axis.

[0026] As shown in FIG. 2, on one wall surface 41 extending in the X direction of the transfer chamber, loading / unloading ports 42 and 43 for two process chambers are respectively provided. When the robot 10 accesses each process chamber, it moves to an access point that is directly opposite the loading / unloading ports 42 and 43 corresponding to that process chamber, and from the position of that access point, one of the hands 25 and 26 is linearly advanced toward the loading / unloading ports 42 and 43. FIG. 2(a) shows the state where the robot 10 has moved to one access point, and FIG. 2(b) shows the state where it has moved to the other access point. As shown in these figures, in this robot 10, as an example, in the state of moving to one access point, the angle α formed by the extending direction of the swing arm 21 and the Y direction is +45°, and in the state of moving to the other access point, the angle α is -45°. At the access point, so that either of the hands 25 and 26 can enter the process chamber through the loading / unloading ports 42 and 43, the two hands 25 and 26 are both arranged to extend parallel to the entering direction into the process chamber (in this example, the +Y direction) through the loading / unloading ports 42 and 43 and to overlap each other. Note that since the common arm 22 needs to rotate around the TH axis for the hands 25 and 26 to enter the process chamber, both hands 25 and 26 do not extend simultaneously toward the inside of the process chamber. FIG. 2(c) shows the state where the hand 25 is entering the process chamber through the loading / unloading port 42. At this time, the other hand 26 is in a folded state.

[0027] Next, the control of the robot 10 in the present embodiment will be described. The robot control method according to the present invention moves the robot 10 between two access points while preventing the hands 25 and 26 of the robot 10 from colliding with surrounding objects, such as the wall surface 41, and moves the robot 10 quickly. FIGS. 3(a) to 3(c) are schematic plan views for explaining the control of the robot 10 in the present embodiment, and show the state of movement of the robot when the angle α formed by the extending direction of the swing arm 21 and the Y direction is changed. In FIG. 3, the change in the posture of the robot 10 accompanying the change in the angle α is shown while being shifted in the drawing, but it should be noted that the position itself of the T axis in the horizontal plane within the transfer chamber is actually unchanged. Also in FIG. 3, for the sake of explanation, the hands 25 and 26 are drawn so that the center lines in the longitudinal directions of the hands 25 and 26 are displaced from the position of the TH axis. Here, considering the case where the robot 10 is moved from the access point shown in FIG. 2(a) as the starting access point to the access point shown in FIG. 2(b) as the ending access point.

[0028] Fig. 3(a) shows the change in the posture of the robot 10 when the robot is moved while maintaining the orientations of the hands 25 and 26 at the starting access point, that is, while maintaining the posture in which the hands 25 and 26 are oriented in the Y direction. In this case, in the process in which the swing arm 21 rotates around the T axis so that the angle α changes from +45° to -45°, as depicted in the figure such that the tips of the hands 25 and 26 protrude into the wall surface 41, the tip portions of the hands 25 and 26 collide with the wall surface 41 of the transfer chamber. That is, it is not possible to move from the starting access point to the ending access point while maintaining the directions of both hands 25 and 26. Conventionally, for example, the direction of the hands 25 and 26 is set to the -X direction by rotating the common arm 22 around the TH axis, and then the swing arm 21 is moved around the T axis while keeping the hands 25 and 26 oriented in the -X direction. Since the directions of the hands 25 and 26 do not change, at this time, the common arm 22 also rotates around the TH axis with respect to the swing arm 21. When the swing arm 21 has rotated to the desired angle, finally, the common arm 22 is rotated around the TH axis so that the directions of the hands 25 and 26 become the Y direction. As a result, the movement from the starting access point to the ending access point is completed without colliding with the wall surface 41. However, since the operation of changing the directions of the hands 25 and 26 and the operation of rotating the swing arm 21 are performed in separate periods, the time required to complete the movement becomes longer.

[0029] Figure 3(b) shows the change in the posture of the robot when the robot 10 is controlled by the control method according to the present invention. Generally, when operating the robot 10 by specifying a target position, as is well known, there are two common operating modes: PTP (point-to-point) operation and CP (continuous path) operation. The PTP operation generally involves moving the tool or hand by specifying only the start and end points of the trajectory that the tip of the tool or hand attached to the robot should follow. The CP operation generally involves specifying a path that is a straight line or curve in three-dimensional space and moving the tip of the tool or hand along that path. In the PTP operation, although the start and end points are specified, the trajectory of the robot between the start and end points is not specified. Particularly for a robot having two or more axes, after determining how much each axis should move between the start and end points for each axis, each axis is moved independently according to the movement amount of each axis. In the control by the PTP operation, an intermediate point can also be defined between the start and end points. If an intermediate point is defined, the robot can be moved from the start point to the intermediate point by the PTP operation, and the robot can also be moved from the intermediate point to the end point by the PTP operation. If multiple intermediate points are set, the robot is also moved between the intermediate points by the PTP operation. In contrast, the CP operation controls each axis at each instant so as not to deviate from the specified path. For example, it is used when interpolating the movement between the teaching points shown in the teaching data with a straight line, an arc, or the like. It is known that the PTP operation can move the robot faster than the CP operation. In the example shown in Figure 3(b), by controlling the movement of the hand 26 from the start access point to the end access point by the CP operation, while changing the directions of the hands 25 and 26 to avoid collision with the wall surface 41, the swing arm 21 is rotated around the T axis at the same time, aiming to shorten the time until the movement between the access points is completed.

[0030] The goal of control during the CP operation is to prevent the tip of the hand 26 from colliding with the wall surface 41. Therefore, control by the CP operation is executed with the tip of the hand 26 as the control target point. Considering the degrees of freedom in the mechanism of the robot 10, the control parameters when the hand 26 performs the CP operation include not only the coordinates of the tip of the hand 26, that is, the coordinates of the control target point, but also the orientation of the hand 26. In the example shown here, since the direction of the hand 26 is the Y direction at both access points, it is assumed that the direction of the hand 26 during movement also remains in the Y direction. In order to avoid collision with the wall surface 41, a straight path L is set parallel to the wall surface 41, slightly separated from the wall surface 41. Then, the robot 10 is driven and controlled so that the tip of the hand 26, that is, the control target point, moves on the path L. Once the position (XY coordinate values) of the tip of the hand 26 and the extending direction of the hand 26 are determined, the rotation angle of the tip arm 24 around the RL axis, the rotation angle of the common arm 22 around the TH axis, and the rotation angle of the swing arm 21 around the T axis (angle α) are determined. Therefore, in the control device 50, the arithmetic unit 52 determines the driving amount of the motors of each axis by referring to the values of each parameter stored in the storage unit 51. As a result, the robot 10 is controlled so that the hand 26 moves by the CP operation. Since the position of the RR axis on the common arm 22 is also determined by the control of the hand 26 during the CP operation, at this time, only the degree of freedom of the rotation angle around the RR axis remains in the robot 10. That is, only one degree of freedom regarding the other hand 25 remains. Therefore, regarding the hand 25, control is performed such that its tip, that is, the control target point of the hand 25, is on the path L, that is, control that specifies only the Y coordinate of the tip of the hand 25. Note that when the hand 26 always faces the Y direction, since the relationship between the X direction position of the hand 26 controlled by the CP operation and the angle α of the swing arm 21 is determined one-to-one, it is also possible to perform control such that the swing arm 21 rotates at a constant rotational speed around the T axis when moving the robot 10 by the CP operation.

[0031] If the control described above is performed, as shown in Fig. 3(b), while the direction of the hand 26 remains in the Y direction and the hand 25 rotates within the XY plane, it is possible to quickly move the robot 10 from the starting access point to the ending access point while avoiding the hands 25 and 26 from colliding with the wall surface 41. Also, depending on the configuration of the process chamber, it may be necessary to enter the process chamber at an inclined angle from the Y direction. That is, although the two access points are arranged along the X direction, the directions in which the hands 25 and 26 should face at the access points are shifted from the Y direction, and it may be necessary to change the directions of the hands 25 and 26 between the access points. In such a case, for the hand 26, control by a CP operation may be performed such that the tip of the hand moves along the path L from the starting access point to the ending access point and the direction of the hand 26 gradually changes. Even when such control is performed for the hand 26, the direction of the other hand 25 also coincides with the direction of the hand 26 at both access points.

[0032] The control described above moves the hand 26 by a CP operation along the path L and moves the other hand 25 in a natural manner such that its tip is on the path L. However, the roles of the hands 25 and 26 may be exchanged, and control may be performed such that the hand 25 is moved by a CP operation along the path L and the hand 26 is moved in a natural manner.

[0033] The movement of the robot 10 from the starting access point to the ending access point can also be controlled by a PTP operation. Fig. 3(c) shows the change in the posture of the robot 10 when it is controlled by a PTP operation. When performing a PTP operation by specifying only the starting access point and the ending access point, since the trajectory in the middle is not specified, it is impossible to eliminate the possibility that the hands 25 and 26 collide with the wall surface 41. Therefore, waypoints are defined for collision avoidance. The waypoints are defined as positions where no collision occurs in the robot 10. Since the swing arm 21 gets closest to the wall surface 41 when α = 0°, it is preferable to define that the tips of the hands 25 and 26, that is, the control target points of the hands 25 and 26, are on the above-mentioned path L when α = 0° as the waypoint. In addition to the waypoint where α = 0°, additional waypoints may be defined.

[0034] Furthermore, in the present embodiment, in order to reduce the degree of freedom in control and make collision avoidance more reliable, constraint conditions are added to perform control by a PTP operation. The first constraint condition is that, according to the angular difference between the direction of the common arm 22 at the starting access point and the direction of the common arm 22 at the ending access point in the XY plane which is the world coordinate, the direction of the common arm 22 in the XY plane also changes in proportion to the rotation amount of the swing arm 21 around the T axis. The direction of the common arm 22 can be defined, for example, as the angle formed by the direction from the TH axis to the RR axis and the X direction in the XY plane. In the example shown in Fig. 3(c), since the direction of the common arm 22 in the XY plane is the same at the starting access point and the ending access point and the angular difference between the two is 0, the direction of the common arm 22 in the XY plane does not change even during the movement of the robot 10. Also, a second constraint condition is added that the change in the rotation angle of the swing arm 21 around the T axis is linked to the change in the respective rotation angles of the tip arms 23 and 24 around the RR axis and the RL axis. Note that the control by the first constraint condition and the second constraint condition is not essential.

[0035] According to the first constraint condition, the positions of the RR axis and the RL axis are farther from the wall surface 41 than the position of the TH axis, and the Y coordinate value of the RR axis is always equal to the Y coordinate value of the RL axis. The second constraint condition is that the angle formed by the common arm 22 and the tip arm 23 across the RR axis and the angle formed by the common arm 22 and the tip arm 24 across the RL axis both become smaller if the absolute value |α| of the angle formed by the swing arm 21 and the Y direction is larger, and become larger if the absolute value |α| becomes smaller. Since the swing arm 21 is closest to the wall surface 41 when α = 0°, it is preferable to define that, as a passing point, the tips of the hands 25 and 26, that is, the control target points of the hands 25 and 26, are on the above-mentioned path L when α = 0°. Additional passing points may be defined in addition to the passing point where α = 0°. Then, the robot 10 is controlled to move in a PTP motion so as to pass through each passing point from the start access point to the end access point under the first and second constraint conditions.

[0036] When controlled by such a PTP motion, as shown in Fig. 3(c), the hands 25 and 26 move symmetrically with respect to a straight line passing through the TH axis and extending in the Y direction. Then, the robot 10 moves from the start access point to the end access point such that the tips of the hands 25 and 26 are generally on the path L.

[0037] As described above, for the case of moving the robot 10 by a CP motion or a PTP motion between the start access point and the end access point shown in Figs. 2(a) and 2(b), the control to prevent the tips of the hands 25 and 26 from colliding with the wall surface 41 has been explained. By executing these controls, it is possible to avoid the occurrence of collisions at the tips of the hands 25 and 26 and shorten the time until the movement of the robot 10 from the start access point to the end access point is completed.

[0038] Next, the control when the workpieces 31, 32 are mounted on the hands 25, 26 will be described. In reality, the robot 10 is a transfer robot, and in many cases, a plate-shaped workpiece is placed on at least one of the hands 25, 26. Therefore, when the robot 10 is carrying a workpiece, it is necessary to control the robot so that the workpiece does not collide with the wall surface 41. The control of the robot 10 when carrying a workpiece will be described with reference to FIGS. 4(a) and 4(b).

[0039] Fig. 4(a) shows the change in the posture of the robot 10 when the robot 10 is moved from the starting access point to the ending access point by controlling the CP operation in the same manner as in Fig. 3(b). However, it is different from that shown in Fig. 3(b) in that the workpieces 31 and 32 are placed on the hands 25 and 26, respectively. Regarding the hand 26 which is actually the object of the CP operation, since it moves while maintaining the state where its direction is the Y direction, the workpiece 32 on the hand 26 also moves accordingly. As a result, it is not necessary to consider the collision of the workpiece 32 with the wall surface 41. The same control as that described with reference to Fig. 3(b) may be performed for the hand 26. On the other hand, since the direction of the hand 25 changes as the robot 10 moves, there is a possibility that the workpiece 31 placed on the hand 25 may collide with the wall surface 41 even if the tip of the hand 25 does not collide with the wall surface 41. The vertex S closer to the ending access point among the vertices at both ends of the side on the tip side of the workpiece 31 is the one where the workpiece 31 may collide with the wall surface 41. Therefore, regarding the hand 25, control may be performed such that either the tip of the hand 25 or the vertex S is on the path L under the condition that neither the tip of the hand 25 nor the vertex S approaches the wall surface 41 more than the path L. That is, control may be performed such that the one closer to the wall surface 41 among the tip of the hand 25 and the vertex S is set as the control target point and the control target point is on the path L. The tip of the hand 25 and the vertex S can be collectively referred to as the potential collision position with respect to the hand 25. Using the term potential collision position, the control condition in the CP operation can be rephrased as that at least one potential collision position is on the path L under the condition that no potential collision position approaches the wall surface 41 more than the path L. Note that regarding the hand 26 which is always moved facing the Y direction, its tip position is the only potential collision position and the control target point.

[0040] In the example shown in Fig. 4(a), the tip of the hand 25 moves on the path L for a while after starting from the starting access point. Among them, since the tip of the hand 25 and the vertex S are simultaneously located on the path L, then the vertex S moves on the path L. At this time, as shown when α = 0°, the tip of the hand 25 is farther from the wall surface 41 than the path L. After that, since the tip of the hand 25 and the vertex S are simultaneously located on the path L, then the tip of the hand 25 moves on the path L and reaches the end access point. When the directions of the hands 25 and 26 are also inclined with respect to the Y direction at each access point, regarding the hand 26 whose position and orientation are controlled by the CP operation, control may be performed so that at least one potential collision position is on the path L under the condition that none of the potential collision positions approaches the wall surface 41 more than the path L.

[0041] Fig. 4(b) shows the change in the posture of the robot 10 when the robot 10 is moved from the starting access point to the end access point by controlling the PTP operation in the same manner as in Fig. 3(c). However, it is different from that shown in Fig. 3(c) in that the workpieces 31 and 32 are placed on the hands 25 and 26, respectively. When controlling the PTP operation, while satisfying each of the above-mentioned restraint conditions, the waypoints are determined so that at least one potential collision position is on the path L under the condition that none of the potential collision positions approaches the wall surface 41 more than the path L, and the robot 10 is moved by the PTP operation so as to move from the starting access point through each waypoint to the end access point. In the example shown in Fig. 4(b), α = 0° is used as a waypoint. At this time, the vertex S of the workpiece 31 is on the path L, and the same applies to the workpiece 32.

[0042] When shown in FIGS. 4(a) and 4(b), if the protruding amounts of the hands 25 and 26 from the workpieces 31 and 32 are small and the collision between the workpieces 31 and 32 and the wall surface 41 is avoided, and if the possibility of collision between the hands 25 and 26 and the wall surface 41 can be substantially ignored, control may be performed from the beginning with the vertex of the workpiece 31, for example, the vertex closer to the end access point among the vertices at both ends of the side on the tip side of the workpieces 31 and 32 as the control target point. Further, when the possibility of collision between the workpieces 31 and 32 and the wall surface 41 can be ignored due to the dimensions of the workpieces 31 and 32 or the placement form of the workpieces 31 and 32 on the hands 25 and 26, the tip position of the hands 25 and 26 may be set as the control target point.

[0043] According to the control described with reference to FIGS. 4(a) and 4(b), it is possible to avoid the occurrence of collisions with the hands 25 and 26 and the workpieces 31 and 32, and shorten the time until the movement of the robot 10 from the start access point to the end access point is completed.

[0044] Depending on the arrangement of the robot 10 in the transfer chamber, the base ends of the hands 25 and 26, that is, the positions of the AR axis and the AL axis, may collide with the wall surface 41. FIGS. 5(a) to 5(c) show the movement states of the robot when the angle α formed by the extending direction of the swing arm 21 and the Y direction is changed, similar to FIGS. 3(a) to 3(c). Here, at the start access point, it is assumed that α = +45° and the directions of the hands 25 and 26 are -Y directions. That is, at the start access point, the common arm 22 rotates 180° around the TH axis without changing the relative positional relationship of the tip arms 23 and 24 and the hands 25 and 26 with respect to the common arm 22 from the state shown in FIG. 2(a). Similarly, at the end access point, it is assumed that α = -45° and the directions of the hands 25 and 26 are -Y directions.

[0045] Fig. 5(a) shows the change in the posture of the robot 10 when the robot 10 is moved while maintaining the orientations of the hands 25 and 26 at the starting access point, that is, while maintaining the posture in which the hands 25 and 26 are facing the -Y direction. In this case, the proximal ends of the hands 25 and 26 collide with the wall surface 41 of the transfer chamber during the movement. On the other hand, Fig. 5(b) shows the change in the posture of the robot 10 when the path L is set in the same manner as that shown in Fig. 3(b) and the robot 10 is moved by the CP operation along the path L. In the example shown here, the hand 25 is controlled by the CP operation so that its proximal end moves along the path L while facing the -Y direction. In this case, the proximal end of the hand 25 is the only potential collision position regarding this hand 25 and is called the control target point. The proximal end of the other hand 26 also moves along the path L as the control target point. Since no workpiece is mounted on the proximal ends of the hands 25 and 26, it is not necessary to consider the collision between the workpiece and the wall surface 41 in the cases shown in Figs. 5(a) to (c).

[0046] Fig. 5(c) shows the change in the posture of the robot 10 when the path L is set in the same manner as that shown in Fig. 3(c) and the robot 10 is moved by the PTP operation. The constraint conditions when performing the PTP operation are the same as those described with reference to Fig. 3(c). The robot 10 moves from the starting access point to the end access point via each waypoint in the same manner as described with reference to Fig. 3(c), except that the waypoints at which the proximal ends of the hands 25 and 26 are on the path L are used as waypoints.

[0047] According to the control described with reference to Figs. 5(a) to (c), it is possible to shorten the time until the movement of the robot 10 from the starting access point to the end access point is completed while avoiding the occurrence of collisions at the proximal ends of the hands 25 and 26.

[0048] As described above, the control of the movement of the robot 10 when the directions of the hands 25 and 26 at the start access point (i.e., the first direction) and the directions of the hands 25 and 26 at the end access point (i.e., the second direction) are the same has been explained. However, the control method of the present invention can also be applied when the orientations of the hands 25 and 26 are different between the start access point and the end access point. Here, consider the case where at the start access point, the hands 25 and 26 face the -X direction and the angle α representing the direction of the swing arm 21 is +45°, and at the end access point, the hands 25 and 26 face the Y direction and the angle α of the swing arm 21 is -45°. The start access point in this case is obtained by rotating the common arm 22, the tip arms 23 and 24, and the hands 25 and 26 by +90° around the TH axis at the start access point shown in Fig. 2(a). The end access point is the same as the end access point shown in Fig. 2(b). Figs. 6(a) to (c) are schematic plan views for explaining the control of the robot 10 when the access points are defined in this way. In particular, Figs. 6(a) and (c) show the movement of the robot when the angle α formed by the extending direction of the swing arm 21 and the Y direction is changed. Workpieces 31 and 32 are placed on the hands 25 and 26, respectively.

[0049] At both the starting access point and the ending access point, the hands 25 and 26 face the same direction and overlap each other. Also, between the starting access point and the ending access point, the swing arm 21 rotates -90° around the T axis, and the directions of the hands 25 and 26 also rotate -90°. Therefore, if the swing arm 21 is rotated -90° while keeping the common arm 22, the tip arms 23 and 24, and the hands 25 and 26 relatively fixed with respect to the swing arm 21, the robot 10 should move from the starting access point to the ending access point. However, in this case, as shown in Fig. 6(a), the tips of the hands 25 and 26 and the workpieces 31 and 32 collide with the wall surface 41, so the movement from the starting access point to the ending access point cannot be completed. Thus, similar to the above, it is considered to control the robot 10 by the CP operation or the PTP operation and move the robot 10 from the starting access point to the ending access point while avoiding collisions.

[0050] When performing a CP operation, it is necessary to define the path that the control target point should follow. Also, when performing a PTP operation, a path is required to determine the intermediate points. In this example, since the orientations of the hands 25 and 26 are different between the access points and the Y coordinates of the tips of the hands 25 and 26 are also different, it is not appropriate to use a straight path. Instead, a curved path C is set. Assuming that the protrusion amounts of the tips of the hands 25 and 26 from the workpieces 31 and 32 in the horizontal plane are negligible, the vertices on the front side with respect to the rotation direction of the swing arm 21 among the two end vertices of the sides on the tip side of the workpieces 31 and 32 can be considered as the potential collision positions for the hands 25 and 26 respectively, and this point is taken as the control target point. Then, consider a path C with the control target point when the hands 25 and 26 are at the starting access point as the starting point P and the control target point when the hands 25 and 26 are at the ending access point as the ending point Q. The path C is, for example, an arc or an elliptical arc, but it may also be a spline curve or a Bézier curve. Since the XY coordinate values of the T-axis position in the world coordinate system are invariant during the movement of the robot 10, Fig. 6(b) shows the superposition of the postures of the robot 10 at the starting access point and the ending access point so that the T-axis positions coincide in the XY plane, and shows the path C in the XY plane. In the example described here, the path C is an arc.

[0051] Fig. 6(c) shows the change in the posture of the robot 10 when the robot 10 is moved from the starting access point to the ending access point by controlling the CP operation. Since the orientations of the hands 25 and 26 differ by -90° between the starting access point and the ending access point, in the example shown here, the control target point for the hand 25 moves from position P to position Q, and the orientation of the hand 25 (or the angle α of the swing arm 21) changes at a constant speed according to the moving distance of the control target point, and the control for the hand 25 in the CP operation is performed on this basis. For the other hand 26, the control is performed on the condition that the control target point is on the path C.

[0052] When the robot 10 starts moving from the starting access point, the hand 25 starts moving along the path C towards the ending access point, while the hand 26 starts moving towards the ending access point ahead of the hand 25, that is, at a speed greater than that of the hand 25. Then, when the hand 26 approaches the ending access point, its speed becomes smaller than that of the hand 25, and finally, the hands 25 and 26 reach the ending access point simultaneously. When approaching the ending access point, among the vertices at both ends of the side on the tip side of the work 32, the vertex on the rear side with respect to the rotation direction of the swing arm 21 slightly protrudes outward from the path C, but this vertex also protrudes from the path C when the hand 26 is at the ending access point, and the work 32 does not collide with surrounding objects due to such protrusion of the vertex from the path C. In order to mitigate the rapid movement of the hand 26 when starting to move from the starting access point, it is conceivable to adopt, for example, a curve other than an arc as the path C. Alternatively, the hand 26 instead of the hand 25 can be moved by CP operation, and only the control to follow the path C can be performed for the hand 25.

[0053] When performing control by PTP operation, the orientation of the common arm 22 in the XY plane of the world coordinate system is different between the starting access point and the ending access point, and it rotates by -90° just like the swing arm 21. Therefore, the first constraint condition is that when the swing arm 21 rotates by -90°, the common arm 22 also rotates by -90° in the XY plane. In this case, no rotational difference occurs between the swing arm 21 and the common arm 22, so the common arm 22 does not move relative to the swing arm 21. The second constraint condition is the same as that used when explaining with reference to Fig. 3(c). Then, via points that satisfy the condition that the control target points for each of the hands 25 and 26 are on the path C are set, and the robot 10 can be controlled to move from the starting access point through the via points to the ending access point.

[0054] According to the control described with reference to FIGS. 6(a) to 6(c), even when the orientations of the hands 25 and 26 at the start access point are different from the orientations of the hands 25 and 26 at the end access point, it is possible to avoid collisions with the hands 25 and 26 and the workpieces 31 and 32, and to shorten the time until the movement of the robot 10 from the start access point to the end access point is completed.

Explanation of Signs

[0055] 10... robot; 20... base; 21... swing arm; 22... common arm; 23, 24... tip arms; 25, 26... hands; 31, 32... workpieces; 41... wall surface; 42, 43... loading / unloading ports; 50... control device; 51... storage unit; 52... arithmetic unit.

Claims

1. A swing arm having a proximal end connected to a first axis and rotatable about the first axis, a common arm connected to the distal end of the swing arm via a second axis and rotatable about the second axis, a first distal arm having a proximal end connected to one distal end of the common arm via a third axis and rotatable about the third axis, a second distal arm having a proximal end connected to the other distal end of the common arm via a fourth axis and rotatable about the fourth axis, a first hand having a proximal end connected to the distal end of the first distal arm via a fifth axis and rotatable about the fifth axis in conjunction with the rotation of the first distal arm about the third axis with respect to the common arm, and a second hand having a proximal end connected to the distal end of the second distal arm via a sixth axis and rotatable about the sixth axis in conjunction with the rotation of the second distal arm about the fourth axis with respect to the common arm, and a robot control method for controlling a robot in which the first to sixth axes are parallel to each other, wherein when moving the robot by rotating the swing arm about the first axis from a starting access point where the first hand and the second hand overlap each other and face a first direction to an ending access point where the first hand and the second hand overlap each other and face a second direction, control is performed to move the first hand by a continuous path operation in which a control target point of the first hand is on a predetermined path and a change in the direction of the first hand is specified, and at the same time control is performed so that a control target point of the second hand is on the predetermined path. A robot control method.

2. A swing arm having a proximal end connected to a first axis and rotatable about the first axis, a common arm connected to the distal end of the swing arm via a second axis and rotatable about the second axis, a first distal arm having a proximal end connected to one distal end of the common arm via a third axis and rotatable about the third axis, a second distal arm having a proximal end connected to the other distal end of the common arm via a fourth axis and rotatable about the fourth axis, a first hand having a proximal end connected to the distal end of the first distal arm via a fifth axis and rotating about the fifth axis in conjunction with the rotation of the first distal arm relative to the common arm about the third axis, and a second hand having a proximal end connected to the distal end of the second distal arm via a sixth axis and rotating about the sixth axis in conjunction with the rotation of the second distal arm relative to the common arm about the fourth axis, and a robot control method for controlling a robot in which the first to sixth axes are parallel to each other, When moving the robot by rotating the swing arm about the first axis from a starting access point where the first hand and the second hand overlap each other and face a first direction to an ending access point where the first hand and the second hand overlap each other and face a second direction, One or more waypoints are defined so that a control target point of the first hand and a control target point of the second hand exist on a predetermined path, and the robot is moved in a point-to-point operation from the starting access point to the ending access point via the one or more waypoints. A robot control method.

3. Under a first constraint condition that the common arm rotates relative to the swing arm about the second axis in proportion to the rotation amount of the swing arm about the first axis according to the angular difference in the orientation of the common arm between the starting access point and the ending access point, and a second constraint condition that the change in the rotation angle of the swing arm about the first axis, the change in the rotation angle of the first distal arm about the third axis, and the change in the rotation angle of the second distal arm about the fourth axis are interlocked, the point-to-point operation is executed. The robot control method according to claim 2.

4. The control target point is, for each of the first hand and the second hand, a point on the hand or on an object mounted on the hand, which is defined as a point that may potentially collide with an object around the robot as the robot moves from the start access point to the end access point. The robot control method according to any one of claims 1 to 3, wherein the predetermined path is defined as a path along which the control target point does not collide with an object around the robot.

5. The robot control method according to claim 4, wherein the first direction and the second direction are the same direction and the predetermined path is a straight line.

6. The robot control method according to claim 4, wherein the second direction is orthogonal to the first direction and the predetermined path is a curve.

7. In the robot, the common arm is an arm bent at a position held by the swing arm via the second axis, and the distance between the second axis and the third axis is equal to the distance between the second axis and the fifth axis, and the distance between the second axis and the fourth axis is equal to the distance between the fourth axis and the sixth axis. The robot control method according to any one of claims 1 to 3.

8. A robot control device for controlling a robot having a swing arm whose proximal end is connected to a first axis and which is rotatable around the first axis, a common arm whose proximal end is connected to the distal end of the swing arm via a second axis and which is rotatable around the second axis, a first distal arm whose proximal end is connected to one distal end of the common arm via a third axis and which is rotatable around the third axis, a second distal arm whose proximal end is connected to the other distal end of the common arm via a fourth axis and which is rotatable around the fourth axis, a first hand whose proximal end is connected to the distal end of the first distal arm via a fifth axis and which rotates around the fifth axis in conjunction with the rotation of the first distal arm with respect to the common arm around the third axis, and a second hand whose proximal end is connected to the distal end of the second distal arm via a sixth axis and which rotates around the sixth axis in conjunction with the rotation of the second distal arm with respect to the common arm around the fourth axis, wherein the first to sixth axes are parallel to each other. A storage unit that stores parameters necessary for controlling the robot. When the robot is moved by rotating the swing arm around the first axis from a starting access point where the first hand and the second hand overlap each other and face in a first direction to an ending access point where the first hand and the second hand overlap each other and face in a second direction, by performing calculations with reference to the parameters in the storage unit, the control target point of the first hand moves by a continuous path operation in which the control target point of the first hand is on a predetermined path and the change in the orientation of the first hand is specified, and at the same time, the control target point of the second hand is on the predetermined path, an arithmetic unit that controls the first to fourth axes; A robot control device having the same.

9. A swing arm having a proximal end connected to a first axis and rotatable around the first axis, a common arm connected to the distal end of the swing arm via a second axis and rotatable around the second axis, a first distal arm having a proximal end connected to one distal end of the common arm via a third axis and rotatable around the third axis, a second distal arm having a proximal end connected to the other distal end of the common arm via a fourth axis and rotatable around the fourth axis, a first hand having a proximal end connected to the distal end of the first distal arm via a fifth axis and rotating around the fifth axis in conjunction with the rotation of the first distal arm with respect to the common arm around the third axis, and a second hand having a proximal end connected to the distal end of the second distal arm via a sixth axis and rotating around the sixth axis in conjunction with the rotation of the second distal arm with respect to the common arm around the fourth axis, the robot control device controlling a robot in which the first to sixth axes are parallel to each other, A storage unit that stores parameters necessary for controlling the robot; When the robot is moved by rotating the swing arm around the first axis from a starting access point where the first hand and the second hand overlap each other and face in a first direction to an ending access point where the first hand and the second hand overlap each other and face in a second direction, by performing calculations with reference to the parameters in the storage unit, one or more waypoints are determined so that the control target points of the first hand and the second hand exist on a predetermined path, and the first to fourth axes are controlled so that the robot moves in a point-to-point operation from the starting access point to the ending access point via the one or more waypoints. An arithmetic unit that controls the robot to move in a point-to-point operation from the starting access point to the ending access point via the one or more waypoints. A robot control device having the above.

Images