JPWO2009096408A1 - Multi-joint structure teaching device - Google Patents

Abstract

An object of the present invention is to provide a multi-joint structure teaching apparatus that teaches the movement of a robot in a manner that is intuitive and does not impose a burden on the human body while observing the motion of the robot following the teaching content in real time. By measuring the amount of change in angle and / or axial rotation between adjacent joints accompanying the movement of the tip joint of the multi-joint structure, and adding up the amount of change between all joints accompanying the movement in the tip joint, Calculates the three-dimensional relative position and orientation after movement relative to the three-dimensional relative position and orientation before movement of the tip joint of the multi-joint structure, and adds the calculated position and orientation information to the calculated position and orientation information A multi-joint structure teaching apparatus characterized by outputting a corrected position / orientation information created by adding information and teaching a robot to follow and move the three-dimensional relative position / orientation of the end effector with respect to a reference origin Is provided.

Description

  The present invention relates to a technique for teaching the position and orientation of an end effector of an industrial robot, and more particularly to a multi-joint structure teaching apparatus that provides appropriate teaching while observing the motion of an industrial robot that follows the teaching content in real time. is there.

  Robot teaching methods include a direct teach method in which an operator moves the robot end effector directly to teach, a teach method using a dummy such as a manipulator, a remote teach method that teaches using an operation panel such as a teach pendant, etc. There is.

  The direct teach method has an advantage that the end effector can be directly and accurately guided to the working position. However, it is difficult to move a manipulator with a long reach. There is also a problem that the body is exposed to danger when an abnormal operation of the robot occurs.

  The teaching method using the manipulator dummy has the drawback that the movement of the operator's arm is restricted by the structure of the manipulator and the operation becomes heavy due to friction of the operating device, etc., but there is also a proposal to improve this ( Patent Document 1). This method uses a gripping part instead of a dummy to detect the movement of the gripping part using magnetic force and operates the manipulator. It takes a long time to perform the operation for a long time, and it is difficult to make a fine adjustment from the lower back to the lower part of the human body. There is also a problem that the movement of the gripping part and the movement of the manipulator are fixed one-to-one.

  The remote teaching method using a teach pendant (a portable device used for teaching the operation of a robot) or the like is a very common method. This method has the advantage that teaching can be performed while looking at the state of the robot, and teaching can be performed in a safe place while easily dealing with the length of reach. However, this method has a problem that an operator cannot be intuitively aware of the expected motion of the robot, so that skill is required for efficient operation.

  In order to solve such a problem, a method has been proposed in which a graphic image of a robot is displayed on the operation surface of a teach pendant, and touch keys for instructing movement in that direction are displayed around the robot image (Patent Literature). 2). However, instructing a three-dimensional movement by displaying the position on a plane and pressing the touch key instead of grasping the wrist of the manipulator does not easily perform an intuitive operation. Requires skill.

  In the remote teaching method, a system is proposed in which the operator has an operation grip with a sensor such as a digitizer in his hand and moves the wrist of the robot based on the relative position and orientation data of the operation grip with respect to the coordinate origin fixed on the waist. There is also (patent document 3). This system has the advantage that it can be operated intuitively while observing the movement of the robot, and can be operated while moving, which improves the proposal of Patent Document 1. However, problems remain in that the operation grip is taught while being held in the air, and there is a restriction of downward movement.

  On the other hand, a proposal using a multi-joint structure is related to a self-position identification technique for a traveling robot (Patent Document 4). However, this proposal is intended primarily for disaster relief, and does not provide a specific proposal for teaching industrial robots.

Japanese Patent Laid-Open No. 1-153288 Japanese Patent Laid-Open No. 10-146782 JP-A-9-216183 JP 2006-343114 A

  In order to solve these problems, the present invention teaches multi-joint structure teaching that teaches robot movement in a manner that is intuitive and does not impose a burden on the human body while observing the movement of the robot following the teaching contents in real time. An object is to provide an apparatus. It is another object of the present invention to provide more appropriate teaching by correcting or editing original data based on an action indicated by an operator.

  In order to solve the above-described problems, a multi-joint structure teaching apparatus according to the present invention is a teaching apparatus that teaches a three-dimensional position and orientation of an end effector of a robot, and includes a joint having a change amount detection sensor in a movable part. A multi-joint structure composed of a group and a processing unit for processing sensor signals and a communication unit for transmitting processed data, an arithmetic editing unit for calculating and editing the transmitted data, and an input for inputting additional instruction information And an output unit for outputting the position and orientation information calculated by the calculation / editing unit, and the angle and / or axis rotation between adjacent joints accompanying the movement of one or a plurality of tip joints of the multi-joint structure And the total amount of change between all joints associated with the movement of the one or more tip joints is summed or summed for each tip joint. Move 3D relative position and orientation after movement relative to the three-dimensional relative position and orientation of the calculated position and orientation information calculated or the corrected position created by adding additional instruction information to the calculated position and orientation information By outputting the posture information, the robot is instructed to follow the three-dimensional relative position and posture of the end effector with respect to the reference origin, and is moved accordingly.

The joint constituting the multi-joint structure has a movable portion in which the formation of the inclination angle with the adjacent joint and / or the rotation of the shaft is variable. Each joint movable part is provided with a sensor such as a potentiometer for measuring the amount of change. When teaching the robot where to move the end effector, first, the operator sets the tip joint of the multi-joint structure and the end effector of the robot to be taught to the start position. Next, the tip joint of the multi-joint structure is moved in the vicinity of a point that seems to correspond to a point where the end effector of the robot should reach.
At this time, the change amount of the three-dimensional position of the tip joint is calculated and processed by summing the inclination angle and the rotation angle of all the joints measured by the sensor, and the calculated three-dimensional position and orientation information is It is transmitted from the output unit to the control unit of the robot. Based on the transmitted information, an instruction to move the end effector is issued from the control unit of the robot, and the end effector follows the tip joint of the multi-joint structure almost in real time. The operator observes the movement of the end effector and further moves the tip joint while correcting the necessary direction and the like.
By repeating such operations, the end effector of the robot can be guided to a precise point to be reached by moving the tip joint.

  Note that the end effector generally means a part having a function of a robot directly acting on a work target, and corresponds to a gripping tool, a fastener, a welding gun, and the like. Further, the position / posture includes a concept of one or both of a three-dimensional position and an inclination of the intermediate portion including the joint and / or the end effector. For the position, the difference between the 3D position after the movement of the tip joint and the 3D position before the movement is output and reflected in the operation of the robot. ) And the tilt (orientation) before the movement is output and reflected in the robot. The movement of the end effector of the robot is performed by driving a servo motor or the like by a driver output based on a command from a robot control unit (controller), but is not limited thereto.

Here, with regard to the multi-joint structure teaching apparatus of the present invention, the three-dimensional taught by the movement of the tip joint is provided by attaching an attachment capable of giving an instruction corresponding to the function of the end effector of the robot to the tip joint. It is a preferable aspect that the end effector of the robot follows and moves based on the relative position and posture and moves and / or operates the end effector according to an instruction by the attachment.
This is because the movement of the tip joint is traced by the end effector of the robot, and operations such as opening / closing, gripping and releasing can be performed according to the instructions. Note that the shape of the attachment is the same as that of the end effector of the robot, so that the three-dimensional position and orientation information can be taught more accurately, but is not limited thereto. As long as it can instruct the end effector to follow the opening / closing, gripping, releasing, and the like, those having similar shapes or considerably different shapes are also included.

Further, the articulated structure teaching apparatus of the present invention fixes the reference origin of the robot, teaches the three-dimensional relative position and orientation of the end effector relative to the reference origin fixed to the robot, and follows and moves and / or operates. It is a preferred embodiment.
By matching the three-dimensional position of the tip joint of the multi-joint structure before starting teaching to the fixed reference origin, the three-dimensional position and orientation of the end effector being moved by teaching with respect to the fixed reference origin can be changed. This is because the absolute position version always matches the three-dimensional position / posture with respect to the three-dimensional position / posture before starting the teaching of the tip joint of the multi-joint structure. In addition, such an aspect has an advantage that the shape of the entire multi-joint structure and the shape of the robot arm are similar, and intuitive teaching is easy. In particular, when the reference origin is set at the base of the base of the robot arm and the 3D position before the start of teaching of the tip joint of the multi-joint structure is set as the position near the rear end joint, the multi-joint structure and the robot arm Intuitive teaching is easy because the moving shapes of are very similar. The reference origin refers to an initial position that serves as a reference for the end effector of the robot.

In addition, regarding the articulated structure teaching apparatus of the present invention, the reference origin of the robot is changed to an arbitrary position, and the robot follows the movement by teaching the robot the three-dimensional relative position and orientation of the end effector with respect to the changed reference origin. And / or operating is a preferred embodiment.
This is because the three-dimensional position of the multi-joint structure before the start of teaching of the tip joint is associated with the changed reference origin. For this reason, the three-dimensional position / posture of the end effector being moved by teaching with respect to the changed reference origin coincides with the three-dimensional position / posture of the tip joint of the moving multi-joint structure before starting teaching. The relative position version can be taught. In such a mode, the end effector can be placed at an arbitrary position and teaching can be started from there, so that there is an advantage that the alignment before the start is simple.

In the multi-joint structure teaching apparatus of the present invention, in the above-described series of movements of the tip joint, the fixed reference origin of the robot is combined and / or changed, and the three-dimensional relative position of the end effector with respect to the fixed reference origin It is a preferred aspect that a hybrid multi-joint structure teaching apparatus is provided that performs teaching that combines teaching of posture and teaching of the three-dimensional relative position and orientation of the end effector with respect to the variation reference origin.
According to this aspect, it is possible to switch between the variable type and the fixed type as appropriate at the teaching stage.

In addition, the articulated structure teaching apparatus of the present invention includes a fine moving means for finely correcting the position of the articulated structure tip, thereby moving the articulated structure tip in a stepwise manner. Is a preferred embodiment.
According to such an embodiment, the robot end effector can be made finer by causing the robot end effector to move to the vicinity of the arrival point by the output based on the movement of the tip of the multi-joint structure, and then outputting the additional information by the input means of another system. It is possible to reach an accurate target point by performing a simple movement.

In addition, the articulated structure teaching apparatus according to the present invention preferably includes means for correcting the initial movement direction of the end effector of the robot and moves the end effector of the robot in the corrected initial movement direction. is there.
There are various methods for installing the robot, such as a floor-standing type, a wall-mounted type, and a ceiling hanging type. Depending on the type, the way the arm bends before operation is different and the direction of the end effector is also different. Therefore, when the robot control unit moves the end effector based on information from the output unit of the multi-joint structure teaching device, the end effector moves smoothly when the direction of the end effector is opposite to the direction of the destination. Cannot start.
In such a case, according to this aspect, the operator can smoothly move the end effector by visually providing correction information on the moving direction as additional information.

The multi-joint structure teaching apparatus of the present invention further includes means for converting the three-dimensional relative position and orientation information of the tip of the multi-joint structure, and outputs an end effector of the robot by outputting the converted position and orientation information. It is a preferable aspect to extend or shorten the moving distance.
According to this aspect, the calculated three-dimensional relative position / posture information can be corrected, and the three-dimensional relative position of the end effector to be reached can be approached or moved away.

Further, the articulated structure teaching apparatus of the present invention has means for sequentially changing the three-dimensional relative position and orientation information of the tip of the articulated structure, and the three-dimensional by moving the tip of the articulated structure. In a preferred embodiment, continuous movement and / or movement of a robot end effector with respect to a plurality of objects is performed by teaching the relative position / orientation information and the number of objects and three-dimensional relative position / orientation information that sequentially changes. is there.
According to this aspect, the corrected position information obtained by sequentially adding the additional information with the coordinate value increased or decreased with respect to the basic teaching based on the movement of the multi-joint structure is output each time the object is changed, so that the objects are adjacent to each other. It can be moved to the position one after another. As additional information, the position on the X axis, the Y axis, and the Z axis, the inclination angle, and the combination thereof can be changed.

In addition, regarding the articulated structure teaching apparatus of the present invention, the difference between the two three-dimensional relative position and orientation information at the obliquely inclined corners of the rectangular region with respect to the tip of the articulated structure is equalized in an arbitrary number. Means for calculating each allocated three-dimensional relative position and orientation information, and the three-dimensional relative position and orientation information of each of the first object and the last object existing in the oblique corner and the object It is a preferable aspect that the movement and / or movement of the robot end effector with respect to all the objects is sequentially and continuously executed by teaching the number of objects in between.
The method of sequentially changing the three-dimensional relative position / orientation information has been described above. In the present invention, first, position / orientation information at both ends is given, and an intermediate space is divided from the information by the number of objects to be arranged. The space per object is determined, and the numerical value on the calculated coordinates is accumulated and added every time the target changes, and the target is arranged. This can be done in a rectangular area.

Here, by having a discontinuous output and / or output timing deferring means of the three-dimensional relative position and orientation information, the movement trajectory and / or movement speed of the end effector of the robot following the movement of the tip of the multi-joint structure is detected. It is preferable to make adjustments. If the number of joints constituting the multi-joint structure is large, if the end effector is made to follow the same movement locus as the movement locus of the tip joint, the amount of information becomes excessive and the tracking may be delayed.
Further, in the case where the operator accurately grasps the point corresponding to the arrival point of the end effector, it is not necessary to sequentially follow. In such a case, only information at regular time intervals or in the vicinity of the target point may be output. On the other hand, it is also possible to slow down the operation of the end effector by delaying and outputting information.

  Further, it is preferable to have means for instructing movement stop and movement resumption of the end effector of the robot. Movement of the end effector by inputting movement stop information when unnecessary movements in the movement of the end effector have increased or are expected to increase, or when approaching the prohibited area for robot work Is stopped and a movement resumption instruction can be issued when the original operation or the safe area is restored. Smoothing is preferably performed for the movement from the movement prohibited point to the movement resuming point.

Further, the calculated and / or corrected information is accumulated in the editing operation unit by the above-described articulated structure teaching apparatus, and editing information obtained by adding additional instruction information to the accumulated information is created or accumulated. 1 edit information selected and / or combined from a plurality of information is created, and the created edit information is output to teach the 3D relative position and posture and move the robot end effector. And / or more preferably it is operated.
Select teaching of the shortest time among multiple teachings, add safety mapping information by separately mapping information that prohibits entry to the taught information, or combine multiple teaching contents into one teaching This is because information can be created.

  In particular, the multi-joint structure described above can be used to move the machining tip axis of the NC machine. NC machine programming (teaching) is also designed to prevent interference when the tip shaft to be processed moves to the processing position of the processing target so that other protrusions of the processing target do not collide when the shaft moves. Although it is necessary, the conventional interactive teaching takes time and results in an increase in man-hours and machining costs. Therefore, in the NC machine programming (teaching), if the apparatus of the present invention is used, the tip shaft for appropriately machining the arm can be moved to the vicinity of machining while visually confirming the interference point. Thereafter, if a moving processing instruction to a processing point based on the conventional method is given, the processing operation can be performed quickly. In other words, this device is applied to parts that require less precision and quicker work when programming NC machines, and conventional methods such as numerical input are applied to parts that require high-precision position movement as before. By adopting such a hybrid configuration, it is possible to greatly reduce the program work time in the case where only the conventional method is performed.

  In addition, the articulated structure teaching apparatus described above is attached to a moving body including a human, and the movement of the articulated structure distal end portion due to the movement of the attachment portion of the articulated structure distal end portion of the moving body and / or the end effector It is preferable to teach the three-dimensional relative position and orientation and movement of the end effector of the robot by an instruction corresponding to the function. By mounting the multi-joint structure teaching apparatus, it is possible to move teaching points and output teaching information. Further, it is not necessary to hold it in the hand, and the mounting site can be selected.

  In addition, a method of teaching the three-dimensional relative position / posture and motion by the multi-joint structure teaching apparatus described above is preferable.

  In addition, the teaching apparatus teaches a three-dimensional position and orientation of a movement destination for one or a plurality of parts constituting a robot, and processes a joint group including a joint having a change amount detection sensor in a movable portion and a sensor signal. An articulated structure composed of a processing unit and a communication unit that transmits processed data, a calculation editing unit that calculates and edits the transmitted data, and an output unit that outputs position information calculated by the calculation editing unit And measuring the amount of change in angle and / or axial rotation between adjacent joints accompanying the movement of one or more joints constituting the multi-joint structure, and adding the amount of change, A three-dimensional position / posture after movement with respect to a three-dimensional position / posture before the movement of one or a plurality of joints of the multi-joint structure is calculated, and one or more parts of the robot are calculated through output of the calculated position / posture information. 3D before moving It is preferable to follow the movement teaches a three-dimensional position and orientation after movement relative postures.

  There are cases where it is necessary to teach the three-dimensional position and orientation of a portion other than the end effector of the robot. Even if the three-dimensional position of the end effector is the same, the three-dimensional position and orientation of the joint part of the robot arm may not be uniquely determined, and a plurality of positions may be possible. In these cases, the movement information of one or a plurality of joints not limited to the tip joint of the multi-joint structure is used as movement command information for a part constituting the robot, such as a joint part of the corresponding robot arm. A joint part etc. can be moved following. The three-dimensional position / orientation of the robot is not limited to the position / orientation of the robot arm part, but includes the position / orientation of the part constituting the whole robot based on the shape / motion of the whole robot including the robot leg and head. . Furthermore, position and orientation information such as the walking distance of the robot is also included. The position and orientation is a concept including one or both of the three-dimensional position and the inclination as described above.

  In addition, it has an input unit for inputting additional instruction information, and outputs corrected position / orientation information generated by adding the additional instruction information to the calculated position / orientation information. It is preferable to follow the three-dimensional position / posture before movement by teaching the post-movement three-dimensional position / posture. It is possible to make corrections such as extending or shortening the travel distance, deleting part of the travel information output (cutting out), and adjusting the deferred output timing.

Further, the position and orientation information calculated by the multi-joint structure teaching apparatus described above is stored and stored in the editing operation unit, and one editing information selected and / or combined from a plurality of stored information is created and created. By outputting the edited information, it is preferable to teach the robot the three-dimensional position / posture after movement with respect to the three-dimensional position / posture before the movement of one or more parts.

  The multi-joint structure is preferably attached to one or both human arms. Since the human body can move the arm portion most quickly and within a free range, it can teach a more effective movement to a portion such as an end effector or a joint of the robot by attaching it to the arm. On the other hand, it is preferable to hold the arithmetic editing unit, the output unit, etc. on the shoulder, back, or waist. In addition, mounting | wearing with both arms means mounting | wearing with one multi-joint structure body in each arm on either side.

  Further, it is preferable that the multi-joint structure is mounted on one or both arms of a human and the input unit is disposed in the vicinity of one hand. It is possible to arrange the calculation editing unit, the output unit, etc. in one hand integrally with the input unit, or to separate the input unit and arrange the calculation editing unit, the output unit, etc. on the shoulder, back or waist. It is particularly preferable from the viewpoint of operability to separate and reduce the weight of the input unit. In that case, the connection with the operation editing unit and the output unit of the input unit can be either wired or wireless.

  It is also preferable to have a teaching option switching unit that switches between execution and stop of the three-dimensional position and orientation teaching of one or more specific parts of the robot. It is possible to perform the operation by appropriately switching between the normal operation set by the robot and the operation based on the teaching of the articulated structure teaching apparatus.

  In addition, a detection unit that detects a workpiece and / or surrounding information recognized by the robot is provided, and based on the detected workpiece and / or surrounding information, one or a plurality of parts of the robot before the movement of the three-dimensional position and orientation are detected. It is preferable to teach a three-dimensional position and orientation after movement. Bidirectional teaching can be performed in which work information such as an image of the robot and surrounding information are detected and reflected in the movement operation of the multi-joint structure.

  In addition, a method of teaching the three-dimensional position / posture after movement with respect to the three-dimensional position / posture before movement of one or more parts of the robot using the multi-joint structure teaching apparatus described above is preferable.

  According to the present invention, the multi-joint structure teaching apparatus teaches the three-dimensional position and orientation of the end effector of the robot in real time. As a result, the end effector of the robot follows the movement of the tip joint of the multi-joint structure, and the operator who observes the following motion moves the tip joint of the multi-joint structure so that it matches the movement of the end effector. A collaborative and efficient teaching is provided that moves and the end effector follows further. Since the multi-joint structure has a shape and bending function similar to those of the robot arm, the operator can teach the three-dimensional position and orientation of the end effector through intuitive actions. If an attachment that can give an instruction corresponding to a function of the end effector is attached to the multi-joint structure, the end effector performs a motion intended by the operator. Depending on the shape of the end effector, by making the shape of the attachment the same shape, it becomes possible to perform movement and movement that are much more similar.

In addition, if the reference origin of the robot is fixed, the digitizer mode is set, and if the multi-joint structure is moved downward, the end effector also moves downward, making the follower movement excellent in intuition. Therefore, there is a merit that accurate work is easy. On the other hand, if it is changed, it becomes a three-dimensional mouse mode, and the operation is simple without considering the initial position of the robot.
In addition, the movement of the multi-joint structure exceeding human reach is possible by returning the mouse to the original position and moving it again. Combining digitizer mode and mouse mode, switch to the mouse mode when you want to return to origin or extend the travel distance, and switch to the digitizer mode when you want accurate teaching. It is possible to make teaching compatible.

  Further, by classifying the end effector into two stages of movement, ie, rough movement and detailed positioning, by the fine movement means, efficient and accurate teaching can be realized (implemented) as a whole. Smooth movement can be realized by correcting the initial moving direction of the end effector of the robot. Further, by converting the three-dimensional position and orientation information, it is possible to move or deform the expanded or contracted information.

  Further, when moving a large number of objects or performing operations on a large number of objects, it is complicated to teach all the objects one by one. In such a case, if an interval between coordinates such as the X axis and the Y axis is given as additional information and the interval is cumulatively increased, the objects are arranged at a constant interval. In this way, repeat movement and operation are possible without taking time and effort. Similarly, when storing an object in a rectangular box, etc., only the teaching on the first object (for example, the lower left corner) and the last object (for example, the upper right corner) is performed, and intermediate objects are equally distributed. By adding additional information by assignment, repeat movement and operation are possible without teaching each time.

  Further, the output information selection and / or output timing adjustment means can perform appropriate teaching while saving the amount of information, or can slow down the operation of the end effector of the robot and improve safety. In addition, the three-dimensional relative position / orientation information obtained by the movement of the tip of the multi-joint structure can be expanded and adapted to the teaching of a large robot. Furthermore, the operation of the end effector can be accelerated. Moreover, the movement of the end effector can be temporarily stopped by the means for instructing the movement stop, or the entry to the work prohibited area such as a passage can be blocked.

  By editing the teaching contents accumulated in the editing operation unit, it is possible to create and determine teaching contents that are evaluated as optimum, such as more accurate teaching and safer teaching. In addition, by mounting the multi-joint structure teaching apparatus, it is possible to perform teaching by a natural operation with less burden without gripping the apparatus. An attachment site can also be selected. For example, when wearing from the shoulder as a starting point, the downward movement can be performed smoothly like the upward movement. By installing a plurality of multi-joint structure teaching devices, it is possible to teach the end effectors of a plurality of robots. If the movement of the articulated structure teaching apparatus itself and the movement of the entire robot are linked, it is possible to teach a walking robot.

  Further, a specific part such as an intermediate joint of the robot can be moved by movement of one or a plurality of joints of the multi-joint structure. The movement of the entire robot including the intermediate part or only the movement of the intermediate part is not limited to the movement of the end effector, and various movements such as waving and dancing can be taught. In addition, by changing the position of the intermediate joint without changing the position of the tip joint of the multi-joint structure, it is possible to prevent the intermediate joint of the robot from coming into contact with the obstacle and to properly operate the end effector such as workpiece gripping. Can be done.

  Further, the movement of one or more parts of the robot based on the movement information of the multi-joint structure can be adjusted by inputting the additional instruction information. It is possible to expand / compress the moving distance of one or more parts of the robot. In addition, when the multi-joint structure follows a complicated movement route, it is possible to trace an efficient movement route that is shortcut to one or more parts of the robot by outputting only a part of the movement information. is there.

  By storing and accumulating teaching information from the movement of multi-joint structures, one or more parts of the robot can be repeatedly moved using the same teaching information, or more appropriate series of teaching information can be constructed by combining multiple teaching information It becomes possible to do.

  In addition, by attaching to the human body and attaching the multi-joint structure to the arm, timely teaching can be performed while moving. By teaching with one hand and performing an emergency stop, etc. with the other hand, quicker and safer teaching can be performed. By attaching two multi-joint structures to both arms, the robot can be operated on the entire robot. It becomes possible to teach. Intuitive operation that directly transmits and teaches the wearer's movements to the robot becomes possible. Note that intuitive teaching can be similarly performed for a whole-body type robot by attaching the multi-joint structure to the arm, leg, and head. Furthermore, if it is used as a measuring device that measures the movement of the human body, the measurement is performed while following the movement of the human body, so that it is possible to perform a wide range of measurement without being restricted by the limitation of the measurement range.

  By making it possible to always switch between execution and stop of teaching, only a moving part that requires teaching based on movement information of the multi-joint structure is selected and taught among a series of movements of one or more parts of the robot. Therefore, quick teaching can be performed.

  Further, by using the workpiece and surrounding information acquired by the robot for teaching, it becomes possible to perform teaching more quickly and accurately based on bidirectional information. For example, when the robot acquires a work image using a camera, the work image is transferred to the detection unit, and the movement of the multi-joint structure is adjusted with reference to the transferred work image. Or the movement of several parts can be taught.

  Hereinafter, an articulated structure teaching apparatus according to the best embodiment of the present invention will be described in detail with reference to the drawings.

  FIG. 1A shows a floor-mounted multi-joint structure teaching apparatus 1. The articulated structure teaching apparatus 1 includes an operation table 2, an articulated structure 3, and an attachment 4. The operation table 2 incorporates a touch panel 5, operation keys 6, a moving direction switching lever 7 and an emergency stop button 8 as an input interface. Of course, the input interface is not limited to this embodiment. For example, the operation table may be reduced in size so that no operation buttons or the like are provided, and the operation table may be connected to a personal computer and input from the personal computer side. In addition, when the multi-joint structure teaching apparatus is mounted on a human body, it is possible to teach while repeatedly moving.

  The multi-joint structure 3 is composed of a large number of joints, and an attachment 4 is attached to the distal end portion 9 of the joint. Of course, it is possible to extend the tip joint to the tip of the attachment. In addition, a joystick 10 having the same function as the input interface of the operation table 2 and a button (not shown) are attached to the attachment 4 for hand operation. After the teaching to the robot is started, it is more convenient to input additional information using the joystick 10 that is substantially in the same position as the tip joint 9.

  FIG. 1B is an enlarged view of the touch panel 5 of the operation table. The touch panel 5 includes a fixed / variable switching display, a distance / time switching display, a stop / restart display, a fine adjustment display, an output switching display, and an editing display. The fixed / variable switching display is a mode switching display indicating whether the reference origin position of the teaching target robot is fixed or variable. The distance switching display is a display for switching the ratio of the movement distance of the tip of the multi-joint structure that is normally set to 1: 1 and the movement distance of the end effector of the robot. The stop / restart display is a display for instructing a restart after the movement or stop of the end effector of the robot is stopped. When the movement is resumed, the movement from the stop position to the resume position is smoothed. The fine adjustment display is used when more precise movement or the like is taught with respect to the teaching by the movement of the normal articulated structure tip. The specific position is adjusted using the right operation key. The output switching display is a selection / non-selection display of an output adjustment mode in which only selected data of all the movements of the tip of the multi-joint structure is output or output with a time shift. The edit display is a display when processing and combining data calculated for movement or data corrected by adding input information.

  FIG. 2 is an example of a multi-joint structure. This multi-joint structure joint is composed of a front end joint 9, a rear end joint, a bending joint 12, an axis rotation joint 13, and a dummy joint. A movable part 11 is mounted between all the joints. The movable part is formed of a low friction resin so as to move smoothly, but a bearing or the like may be used. Each of the movable parts has one degree of freedom, but since the bending joint and the shaft rotation joint are arranged in a well-balanced manner, the entire multi-joint structure can easily advance to an arbitrary three-dimensional position. Of course, it may be more than two degrees of freedom. Each movable part is provided with a potentiometer 14, and a CPU 15 is disposed at some joints.

Next, calculation and transmission of the amount of change associated with movement of the multi-joint structure will be described. When the tip 9 of the multi-joint structure moves, the angle between the joints changes with the movement. On the other hand, the length between each joint does not change. Therefore, if the change in angle is summed, the three-dimensional position / posture after movement with respect to the three-dimensional position / posture before movement and the like can be determined. The change in the angle between the joints is measured by the potentiometer 14, and the measurement data is sent to the CPU 15. Since the CPU has a CAN communication function and one CAN network is constructed in the entire multi-joint structure, the CPU is directly wirelessly transmitted to the calculation editing unit in the operation table and calculated. When the tip joint 9 is branched, the angle change accompanying the movement is measured for each tip joint. Therefore, an operation such as opening / closing of the branched tip portion 9 is also measured as a change in the three-dimensional position. The three-dimensional position / orientation information of the distal joint calculated by the operation editing unit in the operation table is output and transmitted from the output unit in the operation table 2 to the robot control unit.
For example, in order to realize the three-dimensional position and orientation information of the instructed robot end effector, the robot control unit performs a process such as determining a joint angle by inverse kinematics, and moves and deforms the end effector. Next, the end effector will be described with reference to FIG.

  FIG. 3 shows the end effector 17 of the robot that follows the movement of the tip joint of the multi-joint structure. The operator intuitively manipulates the articulated structure as if it were the robot arm 16. The moving object 18 is a shampoo, the moving source is a belt conveyor 19, and the moving destination is a cardboard case 26 for 20 products. The end effector 17 follows the movement of the multi-joint structure of the operator. However, since delicate positioning is necessary at the stage of gripping the shampoo, the movement of the tip of the operator swings and the end effector 17 follows. It will move very much. The same applies when putting shampoo into the box. In such a case, an input stop button provided on the joystick 10 is pressed near the end of the smoothly moving region. When the stop button is pressed, the previous three-dimensional position / orientation information is recorded in the memory 1 and the following movement of the end effector 17 is stopped. Next, when it is determined that the tip portion has come to a position corresponding to the position directly above the shampoo (or just above the cardboard box), the movement of the end effector 17 is restarted by pressing the restart button. Through this series of operations, the end effector 17 can be taught only to the smoothly moving part. In addition, it can solve easily by performing a smoothing process for the connection of a stop location and a restart location.

  FIG. 4 shows an enlarged view of the attachment of the tip joint of the multi-joint structure. A is a joystick type. Using the joystick on the front side and the button on the back side, input information is instructed. B is a grip sensor type. The object grip / release information can be output to the end effector. C is an electronic globe type. The tip joint of the multi-joint structure is extended to the end of each finger, and accurate follow-up movement and operation instructions can be given to the multi-finger type robot arm. Further, as will be described later, it is possible to attach an attachment having a special shape similar to the shape of the end effector.

  FIG. 5 is a conceptual diagram relating to input means for correcting a problem of initial movement due to a difference in the initial position of the robot arm 16. The left side of FIG. 6 shows the initial positions of the multi-joint structure, the ceiling-suspended robot arm 16, and the floor-standing robot arm 16. When the tip joint of the multi-joint structure starts to move and starts teaching, the ceiling-suspended robot arm 16 is set so that the direction of the arm radiates from the base portion as in the multi-joint structure. Accordingly, it is possible to smoothly follow the movement of the joint at the tip of the multi-joint structure. On the other hand, the floor-standing robot arm 16 is set so as to approach the base. In this case, some robots may try to move downward. Therefore, by moving and correcting the direction with the movement direction switching lever 7 of the display table or the button of the joystick 10, the movement or the like can be performed smoothly.

  FIG. 6A is a conceptual diagram when the single bottle 22 is stored in a case, and FIG. 6B is a conceptual diagram when the shampoo 18 is stored in a square box. In FIG. 6A, the first movement of the first bottle 22 is taught. Next, new three-dimensional position / orientation information is added by adding data of an interval shifted to the right so that the three-dimensional position / orientation information of the multi-joint structure joint 9 is not in contact with the first bottle 22 stored earlier. Further, information with the same interval added is created and applied to the next bottle, and then the next bottle. By this series of operations, the next bottle is placed in the space 1 and the next bottle is placed in the space 2. The repeat operation after teaching once can be automated.

  FIG. 6B is a box containing 10 shampoos. In the teaching such as movement of the shampoo container, the first shampoo container 24 is stored in the lower left and the last shampoo container 25 is stored in the upper right. After that, there are 8 spaces available. In this case, without performing the same teaching on the remaining eight, the calculation is performed by dividing the lower four spaces and the upper four spaces by 8, and 1, 2, 3 of the space per calculation result. The data shifted to the right side for each of these is used as the three-dimensional position / orientation data of the container center point. Similarly, the remaining eight containers can be automatically moved by adding the position information of the upper part.

  FIG. 7 is a conceptual diagram of fixed type teaching of the shampoo conveyance system. In this figure, a human wears an articulated structure device. The attachment of the multi-joint structure is an electronic glove 21. The robot is arranged in the mirror image position opposite to the left and right, and the shampoo container 18 placed on the belt conveyor is moving between the human and the robot. Although not shown, the operation table 2 is a touch panel display.

  First, the operator turns on the power of the operation unit. Next, the origin position (arrangement position) of the robot arm 16 is selected. By this operation, the robot arm connection is started, and the robot arm 16 moves to the designated position. Here, the operator selects the fixed mode with the joystick 10 (shown by A in FIG. 4) of the multi-joint structure. Then, based on the three-dimensional position information of the multi-joint structure, it is confirmed whether or not the robot arm 16 follows.

Next, start teaching. The multi-joint structure 3 is operated to move the robot arm 16 to the vicinity of the reference origin. Teaching starts here. Press the memory 1 of the joystick 10 (see A in FIG. 4). The robot arm 16 moves to the position of the destination belt conveyor 19 by operating the multi-joint structure 3. Here, the memory 2 is pushed. Press the fine adjustment switching button to enter the fine adjustment mode. Fine adjustment is performed with the buttons of the joystick 10 (see A in FIG. 4). When the shampoo container 18 is gripped, the memory 3 is pushed. The multi-joint structure 3 is operated to move the robot arm 16 to the destination packing box and push the memory 4.
Next, fine adjustment is performed by pressing the fine adjustment switching button again. The operation is the same as above. Then, the shampoo container 18 is released and stored in the packing container 26. Here, the memory 5 is pushed. Thereafter, when teaching is continued, the robot arm 16 is moved to the original position, and when returning to the origin, it is moved to the vicinity of the origin. When a series of teaching is completed, teaching completion and storage processing are performed by editing and displaying the operation table 2. Thereafter, the stored teaching content is executed, and the robot arm 16 is automatically moved to check its operation. Correct any malfunctions.

FIG. 8 is a conceptual diagram of variation type teaching of the shampoo delivery system. The multi-joint structure attachment 4 is an electronic glove 21 and the operation unit (not shown) is a personal computer.
In the preparation stage, it is almost the same as that in FIG. 7, but here, the variation mode is selected unlike FIG. Next, the teaching begins. The robot arm 16 is moved to the origin by the operation unit. The teaching information is output to the robot only when the button in the electronic glove 21 of the multi-joint structure 3 is kept pressed. Unlike the case of the fixed mode, there is no correlation between the position of the arm of the multi-joint structure 3 operator and the position of the robot.
The multi-joint structure 3 is operated to move the robot arm 16 to the position of the belt conveyor 19 to be moved, and the memory 1 is pushed. Switch to fine adjustment and use the buttons in the hand glove. When holding the shampoo, press the memory 2. The multi-joint structure 3 is operated to move to the destination packing box 26 and the memory 3 is pushed. Switch to fine adjustment and operate with the hand. The shampoo container 18 is released and stored in the package, and the memory 4 is pushed. The subsequent steps are the same as in the example of FIG.

  FIG. 9 shows the teaching of the painting robot. You are operating from behind the robot. The multi-joint structure attachment 4 is obtained by connecting a paint gun type attachment 27 to an electronic glove 21. For fixed and variable selection, the hybrid mode is selected and the advantages of both are combined.

  The preparation stage is the same as in FIGS. 7 and 8 except for the hybrid mode selection. When teaching is started, the mode is variable. The memory 1 is pushed only when the robot is moved to the start position of the painting object without pushing the memory immediately at the start of the operation. Next, the mode is switched to the fixed mode. The articulated structure 3 is moved and deformed while controlling the coating amount using the joystick 10 or the analog amount detection switch, and the robot arm 16 is moved following the coating, and the coating gun 28 is used for coating. After painting is complete, move to the starting position of the painting object or return to the origin.

  In addition, when painting one item uniformly, the robot is taught by the multi-joint structure device to perform painting each time. This is because, if performed directly by human beings, dangerous gas may be generated by using an air gun for painting depending on products such as arts. In such a case, it is preferable to use the multi-joint structure apparatus according to the present invention. In consideration of safety, in this embodiment, the operator performs a remote operation from the back away from the robot.

  A human body-mounted multi-joint structure teaching apparatus will be described with reference to FIGS. 10, 11, 12, and 13. FIG.

  FIG. 10 is a conceptual diagram related to the human body-mounted multi-joint structure teaching device 29. It is possible to teach movement to a double-armed robot by arranging the multi-joint structure 3 along both arms, but currently only the multi-joint structure placed on the left arm is functioning, And an attachment 4 simulating a paint gun is attached to the tip. A portable control unit 30 is disposed on the back of the human body so as to be connected to the multi-joint structure 3. FIG. 11 is a conceptual diagram projected from the front side of the human body in the same embodiment as FIG. A portable input unit 31 for inputting additional information and for emergency stop is attached to the tip of the multi-joint structure 3 disposed on the right arm. As for the attachment of the multi-joint structure to the human body, the multi-joint structure faithfully traces the action of the human body (operator) to be attached if it is disposed in close contact with only a movable part of the human body such as a joint. be able to. It is not necessary to tightly fix all the joints constituting the multi-joint structure to the human body.

  FIG. 12 is a conceptual diagram of the internal configuration of the portable control unit 30, and FIG. The portable control unit 30 includes a calculation editing CPU 32 and an output unit 33. The output unit 33 is provided with an output module 33a, a wireless output connector 33b, and a wired output connector 33c, and output can be selected by wireless LAN or wired LAN.

  There are three types of joints constituting the multi-joint structure: a bending type and a shaft rotation type having a movable part, and a connection type having no movable part. In the bending type, the angle of a line connecting adjacent joints can be inclined. There are two types of bendable joints: one that can be folded only up and down and the other that can be folded only left and right. In both cases, the bendable range is ± 45 degrees with one degree of freedom. The shaft rotation type joint can rotate in the direction perpendicular to the axis passing through the joint without changing the line connecting the adjacent joints. The shaft rotation type also has a stop point with one degree of freedom and a rotatable range of 0 to 120 degrees. The connection type is arranged at both ends of the multi-joint structure 3 connected to the portable control unit 30 and the attachment 4. The possible range of the type, arrangement, bending, etc. of the joints constituting the multi-joint structure is not limited to this embodiment.

  A potentiometer 14 is disposed in each of the types of joints having a movable part, and a CPU 15 is provided for each unit composed of several joints, and CAN communication is wired. The amount of change between joints measured by the potentiometer due to bending or shaft rotation is transmitted to the CPU by wire, and the amount of change data between multiple joints calculated by the CPU is computed and edited by the portable control unit 30 by CAN communication (wired). Is transmitted to the CPU 32. The calculation editing CPU 32 calculates the transmitted measurement value, converts it into position / orientation information data corresponding to the configuration of the robot, and transmits the data from the output unit 33 to the robot controller in a wired or wireless manner. The sensor for measuring the amount of change between joints, the transmission to the CPU, and the CAN communication method are not limited to this embodiment, and the portable controller transmits position and orientation information data corrected by adding additional information to the amount of change data to the robot controller. It is also possible to do.

  An example of the operation of the robot based on the position and orientation information transmitted from the output unit will be described. Based on the position and orientation information transmitted to the robot controller of the articulated structure teaching apparatus according to the present invention, a motor drive pattern is created by the positioning control unit of the robot controller, and a drive command based on a pulse signal is issued. Based on the command from the controller, the driver outputs the modulated power to the motor. The motor rotates the rotor based on the command pulse, and the end effector moves toward the object to be painted by the rotation of the robot arm through the speed reducer 37.

  The articulated structure teaching apparatus according to the present invention can be widely used in factory assembly, transfer lines, etc., and has great industrial applicability.

Conceptual diagram of multi-joint structure teaching device Conceptual diagram of articulated structure Conceptual diagram of robot Enlarged view of the attachment Conceptual diagram of operation comparison by robot arm arrangement Container insertion concept Conceptual diagram of teaching with fixed multi-joint structure teaching device Conceptual diagram of teaching with variable multi-joint structure teaching device Conceptual drawing of painting with a hybrid articulated structure teaching device Human body-mounted multi-joint structure teaching device (back side of human body) Human body-mounted multi-joint structure teaching device (human abdomen side) Multi-joint structure teaching device control unit internal structure Multi-joint structure teaching device controller outline

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Multi-joint structure teaching apparatus 2 Operation table 3 Multi-joint structure 4 Attachment 5 Touch panel 6 Operation key 7 Movement direction switching lever 8 Emergency stop button 9 Tip joint 10 Joystick 11 Movable part 12 Bending joint 13 Shaft rotary joint 14 Potentiometer 15 CPU
16 Robot Arm 17 End Effector 18 Object 19 Belt Conveyor 20 Grip Sensor Type Attachment 21 Electronic Glove Type Attachment 22 First Bottle 23 Space 24 First Shampoo Container 25 Last Shampoo Container 26 Packing Box 27 Paint Gun Type Attachment 28 Paint Gun 29 Human Body Attachment Type multi-joint structure teaching device 30 Portable control unit 31 Portable input unit 32 CPU for arithmetic editing
33 Output unit 33a Output module 33b Wireless output connector 33c Wired output connector

Claims (24)

  A teaching apparatus for teaching a three-dimensional position and orientation of an end effector of a robot, a joint group including a joint having a change amount detection sensor in a movable part, a processing unit for processing sensor signals, and a communication unit for transmitting processed data A multi-joint structure comprising: a calculation editing unit for calculating and editing transmitted data; an input unit for inputting additional instruction information; and an output unit for outputting position and orientation information calculated by the calculation editing unit. And measuring the amount of change in angle and / or axial rotation between adjacent joints accompanying the movement of one or more tip joints of the multi-joint structure, and accompanying the movement in the one or more tip joints By adding the amount of change between all joints or summing each tip joint, the three-dimensional relative position after movement relative to the three-dimensional relative position and posture before the movement of one or more tip joints of the multi-joint structure Place By calculating the orientation and outputting the calculated position and orientation information or the corrected position and orientation information created by adding additional instruction information to the calculated position and orientation information, the robot can output the three-dimensional end effector to the reference origin. A multi-joint structure teaching apparatus characterized by teaching a relative position and posture to move following.   By attaching an attachment that can give an instruction corresponding to the function of the end effector of the robot to the tip joint, the end effector of the robot follows and moves based on the three-dimensional relative position and orientation taught by the movement of the tip joint. The multi-joint structure teaching apparatus according to claim 1, wherein the end effector is moved and / or operated in accordance with an instruction by an attachment.   The reference origin of the robot is fixed, and the three-dimensional relative position and orientation of the end effector with respect to the reference origin fixed to the robot is taught to follow and move and / or operate. Fixed multi-joint structure teaching device.   The reference origin of the robot is changed to an arbitrary position, and the three-dimensional relative position and orientation of the end effector with respect to the changed reference origin is taught to the robot, and the follow movement and / or operation is performed. 2. The variable multi-joint structure teaching apparatus according to 2.   In the course of a series of movements of the tip joint, the fixed and / or variation of the reference origin of the robot is combined to teach the three-dimensional relative position and orientation of the end effector relative to the fixed reference origin and the three-dimensional end effector relative to the variation reference origin 3. The hybrid articulated structure teaching apparatus according to claim 1, wherein teaching is performed in combination with teaching of relative position and orientation.   6. The movement of the tip of the multi-joint structure is performed stepwise by providing fine movement means for finely correcting the position of the tip of the multi-joint structure. Multi-joint structure teaching device.   7. The multi-joint structure according to claim 1, further comprising means for correcting an initial movement direction of the end effector of the robot, and moving the end effector of the robot in the corrected initial movement direction. Body teaching device.   The apparatus has means for converting the three-dimensional relative position and orientation information of the tip of the multi-joint structure, and the moving distance of the end effector of the robot is extended or shortened by outputting the converted position and orientation information. Item 8. The articulated structure teaching apparatus according to any one of Items 1 to 7.   Means for sequentially changing the three-dimensional relative position and orientation information of the tip of the multi-joint structure, and three-dimensional to sequentially change the three-dimensional relative position and orientation information by the movement of the tip of the multi-joint structure and the number of objects The articulated structure according to any one of claims 1 to 8, wherein continuous movement and / or movement of a plurality of objects of an end effector of a robot is executed by teaching relative position and orientation information. Body teaching device.   Each three-dimensional relative position / orientation information is calculated by equally assigning an arbitrary number of differences between the two three-dimensional relative position / orientation information at the oblique corners of the rectangular region with respect to the tip of the multi-joint structure. A robot end effector by teaching the three-dimensional relative position and orientation information of each of the first object and the last object in the oblique corner and the number of objects between the objects. The multi-joint structure teaching apparatus according to any one of claims 1 to 9, wherein the movement and / or operation for all the objects is sequentially and continuously executed.   By having discontinuous output and / or output timing deferring means of the three-dimensional relative position and orientation information, the movement trajectory and / or movement speed of the end effector of the robot that follows the movement of the tip of the multi-joint structure is obtained. The articulated structure teaching apparatus according to any one of claims 1 to 10, wherein adjustment is performed.   12. The articulated structure teaching apparatus according to claim 1, further comprising means for instructing movement stop and movement resumption of the end effector of the robot.   Information calculated and / or corrected by the articulated structure teaching apparatus according to any one of claims 1 to 12 is accumulated in an editing operation unit, and editing information obtained by adding additional instruction information to the accumulated information is stored. One edit information selected and / or combined from a plurality of information created or stored is created, and the created edit information is output to teach the three-dimensional relative position and orientation and motion. A multi-joint structure teaching apparatus characterized by moving and / or moving the end effector.   An articulated structure teaching apparatus using the articulated structure according to any one of claims 1 to 12 for moving a machining tip axis of an NC machine.   The multi-joint structure teaching apparatus according to any one of claims 1 to 12 is attached to a mobile body including a human and the tip of the multi-joint structure body is moved by the movement of the attachment portion of the multi-joint structure front-end part of the mobile body. A multi-joint structure teaching apparatus that teaches the three-dimensional relative position and orientation and movement of an end effector of a robot by instructions corresponding to movement and functions of the end effector.   Teaching the three-dimensional relative position and orientation of an end effector of a robot, wherein the multi-joint structure teaching apparatus according to any one of claims 1 to 15 is used to teach a three-dimensional relative position and orientation and movement. Method.   A teaching device that teaches a three-dimensional position and orientation of a movement destination for one or a plurality of parts constituting a robot, and a processing unit that processes a sensor group and a joint group including a joint having a change amount detection sensor in a movable part A multi-joint structure comprising a communication unit that transmits processed data, a calculation editing unit that calculates and edits the transmitted data, and an output unit that outputs position and orientation information calculated by the calculation editing unit By measuring the amount of change in angle and / or axial rotation between adjacent joints accompanying the movement of one or more joints constituting the multi-joint structure, and adding the amount of change. The movement of one or more parts of the robot is calculated by calculating a three-dimensional position / posture after movement relative to the three-dimensional position / posture before the movement of one or more joints of the joint structure, and outputting the calculated position / posture information. Previous three-dimensional Multi-joint structure teaching device, characterized in that to follow movement teaches a three-dimensional position and orientation after movement relative postures.   An input unit for inputting additional instruction information, and adding the additional instruction information to the calculated position / orientation information, and outputting corrected position / orientation information, thereby moving one or more parts to the robot 18. The articulated structure teaching apparatus according to claim 17, wherein the three-dimensional position and orientation after movement with respect to the previous three-dimensional position and orientation is taught and moved following.   The position / orientation information calculated by the multi-joint structure teaching apparatus according to claim 17 or 18 is stored and accumulated in an editing operation unit, and one edit selected and / or combined from a plurality of accumulated information By creating information and outputting the created editing information, the robot is instructed to follow the three-dimensional position after the movement relative to the three-dimensional position and orientation before the movement of one or more parts, Multi-joint structure teaching device.   The multi-joint structure teaching apparatus according to claim 15 or 17 to 19, wherein the multi-joint structure is mounted on one or both human arms.   The multi-joint structure according to claim 15 or 17 to 19, wherein the multi-joint structure is attached to one or both arms of a human and the input unit is arranged in the vicinity of one hand. Body teaching device.   The multi-joint structure teaching according to claim 15 or 17 to 21, further comprising a teaching option switching unit that switches between execution and stop of three-dimensional position and orientation teaching of a movement destination of one or more parts of a robot. apparatus. A detection unit that detects a workpiece and / or surrounding information recognized by the robot, and based on the detected workpiece and / or surrounding information, moves one or more parts of the robot relative to the three-dimensional position and orientation before moving. 23. The multi-joint structure teaching apparatus according to claim 15 or 17, wherein the following three-dimensional position is taught to be moved following.   The multi-joint structure teaching apparatus according to any one of claims 17 to 23 is used to teach and follow a three-dimensional position after movement of one or more parts of the robot with respect to a three-dimensional position and orientation before movement. A method for teaching a three-dimensional position of a robot.