JP2015058492A - Control device, robot system, robot, robot operation information generation method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a control device, a robot system, a robot, a robot operation information generation method, a program, and the like for generating robot operation information at high speed by input easy for a user by performing processing using local areas on the basis of layout information including an initial position and a final position of a workpiece.SOLUTION: A control device 100 includes: an information acquisition unit 110 that acquires layout information including a first position and a second position of a workpiece in a work space; and a processing unit 120 that generates robot operation information on the basis of the layout information. The processing unit 120 generates workpiece moving information that represents moving of the workpiece using a plurality of local areas by setting the local areas on a predetermined flat surface in the work space and determining a correspondence between the first and second positions of the workpiece and the local areas, and generates the robot operation information represented using the local areas on the basis of the workpiece moving information.

Description

  The present invention relates to a control device, a robot system, a robot, a robot motion information generation method, a program, and the like.

  In order to cause a robot to perform a desired operation, a method using teaching by a user is widely known. Specifically, direct teaching that teaches robot operation by a user (teacher) moving a robot arm or the like by hand, teaching playback that teaches robot operation by operating an operation unit such as a teaching pendant, Alternatively, off-line teaching or the like in which teaching is performed using a simulation on a computer without using a real robot is known. Patent Document 1 discloses a technique for performing offline teaching with an operation feeling close to teaching using an actual machine by operating a robot simulator using a teaching pendant.

  In addition, when a given condition is given by the user's instruction, the robot action according to the condition is automatically generated instead of designating the robot action one by one using a teaching pendant etc. Is also known to automatically generate a teaching program. For example, Patent Document 2 discloses a method of automatically generating a teaching program adapted to an actual environment based on the contact state of components when an assembly operation by contact is targeted.

JP 2003-236784 A JP 2000-267719 A

  In the teaching using the teaching pendant as in Patent Document 1, the user who performs the teaching has knowledge about the robot and knowledge about the work to be performed by the robot (for example, if welding is desired to be performed, welding Knowledge).

  Japanese Patent Application Laid-Open No. 2004-228561 describes that it can be adapted to changes in the actual environment, but it is a human task to teach the posture acquired by the range finder in accordance with the appearance (aspect). Therefore, when the layout is changed, the system of Patent Document 2 cannot automatically associate the range data corresponding to the layout with the aspect. As a result, in the conventional method such as Patent Document 2, when the layout is changed, the work by the user is required, so that the burden on the user is large.

  One aspect of the present invention is an information acquisition unit that acquires layout information including a first position of a work in a work space of a robot and a second position of the work different from the first position, and the acquired information A processing unit that generates robot motion information based on layout information, wherein the processing unit sets a plurality of local regions on a predetermined plane in the work space, and the first position of the workpiece And the correspondence between the local region and the correspondence between the second position of the workpiece and the local region, the movement of the workpiece from the first position to the second position is The present invention relates to a control device that generates work movement information expressed using a local area, and generates the robot motion information expressed using the local area based on the work movement information.

  In one aspect of the present invention, layout information including a first position and a second position of a workpiece is acquired, and robot motion information is generated using a local region. Therefore, if the user inputs layout information, a movement path between the first position and the second position, a robot operation that realizes movement of the workpiece along the movement path, and the like are automatically generated. An easy-to-use system can be realized. Also, by setting the processing unit as a local area set in a plane, the processing load of the robot motion information generation process can be reduced, and the processing can be executed at high speed.

  In one embodiment of the present invention, the robot includes a plurality of arms, and the robot operation information is obtained by using any one of the plurality of arms to move the workpiece from the first position to the first position. It may be information indicating whether to move to position 2.

  Thereby, it is possible to generate information as to which robot is used to move the workpiece as robot operation information.

  In one embodiment of the present invention, the processing unit obtains a correspondence relationship between the movable range of the arm and the local region for each arm of the plurality of arms, thereby using the local region to move the movable range. The robot motion information may be generated based on the movable range information and the workpiece movement information.

  As a result, it is possible to generate movable range information representing the movable range of the arm using a local region, and to generate robot motion information using the movable range information.

  In the aspect of the invention, the layout information further includes a third position of the workpiece different from the first position and the second position, and the processing unit includes the third position of the workpiece. By determining the correspondence between the position of the workpiece and the local region, the workpiece moves from the first position to the third position, and the workpiece moves from the third position to the second position. The workpiece movement information representing movement using the local area may be generated, and the robot motion information represented using the local area may be generated based on the workpiece movement information.

  This makes it possible to include the third position of the work in the layout information.

  In one aspect of the present invention, the processing unit specifies size information of the work that is a work target based on the layout information, and sets the size of the local region based on the size information. Also good.

  Thereby, it is possible to set the size of the local area based on the work size.

  In one embodiment of the present invention, when a plurality of the robot motion information is generated, the processing unit obtains a work evaluation value for each of the plurality of robot motion information, and based on the obtained work evaluation value. One determined robot operation information may be output.

  This makes it possible to appropriately determine the robot operation information to be output using the work evaluation value.

  Another aspect of the present invention relates to a robot system including any one of the control devices described above and the robot.

  This makes it possible to realize a robot system that generates the above-described robot motion information and controls the robot using the generated robot motion information.

  According to another aspect of the present invention, an information acquisition unit that acquires layout information including a first position of a workpiece in a predetermined work space and a second position of the workpiece different from the first position is acquired. A processing unit that generates robot motion information based on the layout information, wherein the processing unit sets a plurality of local regions on a predetermined plane in the work space, and the first of the workpiece By determining the correspondence between the position and the local region and the correspondence between the second position of the workpiece and the local region, the workpiece is moved from the first position to the second position. The present invention relates to a robot that generates work movement information represented using the local area and generates the robot motion information represented using the local area based on the work movement information.

  Thus, it is possible to realize a robot that generates the above-described robot operation information and performs an operation based on the generated robot operation information.

  According to another aspect of the present invention, a process of acquiring layout information including a first position of a work in a work space of a robot and a second position of the work different from the first position is performed. Based on the layout information, a robot motion information generation process for generating robot motion information is performed, and as the robot motion information generation process, a plurality of local regions are set on a predetermined plane in the work space, and the workpiece By obtaining the correspondence between the first position and the local region and the correspondence between the second position of the workpiece and the local region, from the first position to the second position of the workpiece. Generating movement information representing the movement of the robot using the local area, and generating the robot motion information represented using the local area based on the workpiece movement information. Relating to the robot operation information generation method for performing.

  Thereby, layout information including the initial position and the final position of the workpiece is acquired, and robot operation information is generated using the local region. Therefore, if the user inputs layout information, the movement path between the initial position and the final position and the robot movement that realizes the movement of the workpiece along the movement path are automatically generated. It can be realized. Also, by setting the processing unit as a local area set in a plane, the processing load of the robot motion information generation process can be reduced, and the processing can be executed at high speed.

  Another aspect of the present invention provides an information acquisition unit that acquires layout information including a first position of a work in a work space of a robot and a second position of the work different from the first position, and the layout Based on the information, the computer is caused to function as a processing unit that generates robot motion information, the processing unit sets a plurality of local regions on a predetermined plane in the work space, and the first of the workpiece By determining the correspondence between the position and the local region and the correspondence between the second position of the workpiece and the local region, the workpiece is moved from the first position to the second position. The present invention relates to a program that generates work movement information expressed using the local area and generates the robot motion information expressed using the local area based on the work movement information.

  In one aspect of the present invention, layout information including an initial position and a final position of a workpiece is acquired, and robot motion information is generated using a local region. Therefore, if the user inputs layout information, the movement path between the initial position and the final position and the robot movement that realizes the movement of the workpiece along the movement path are automatically generated. It can be realized. Also, by setting the processing unit as a local area set in a plane, the processing load of the robot motion information generation process can be reduced, and the processing can be executed at high speed.

  As described above, according to some aspects of the present invention, a process using a local area is performed on the basis of layout information including an initial position and a final position of a workpiece, so that a robot can be operated at high speed with easy input for a user. A control device, a robot system, a robot, a robot motion information generation method, a program, and the like that generate motion information can be provided.

The system configuration example of the control apparatus which concerns on this embodiment. The detailed system configuration example of the control apparatus which concerns on this embodiment. The structural example of the robot system which concerns on this embodiment. The other structural example of the robot system which concerns on this embodiment. An example of the initial state corresponding to the initial position of the workpiece. An example of a local region set in the work space. An example of the correspondence between the initial position of the workpiece and the local area. The final state corresponding to the final position of a workpiece | work, and the correspondence example of a final position and a local area | region. The intermediate state corresponding to the intermediate position of the workpiece, and the correspondence relationship between the intermediate position and the local area. The figure explaining the assembly | attachment operation | work of two workpieces. Explanatory drawing of the movable range and movable range information of each arm. The figure explaining the correspondence of the initial position etc. of a work, and a local field. An example of robot operation information obtained by determining an arm to be used. 14A to 14D are explanatory diagrams of specific examples of robot operations. FIG. 15A to FIG. 15D are other explanatory diagrams of specific examples of the robot operation. FIG. 16A to FIG. 16D are other explanatory diagrams of specific examples of the robot operation. FIG. 17A to FIG. 17D are other explanatory diagrams of specific examples of the robot operation. An example in which a plurality of movement paths and robot operation information are obtained. The figure explaining the work cost calculation process with respect to each robot operation information. The other figure explaining the work cost calculation process with respect to each robot operation information. 21A and 21B show setting examples of the moving speed of the robot. The flowchart for demonstrating the process of this embodiment. The flowchart for demonstrating work cost calculation processing.

  Hereinafter, this embodiment will be described. In addition, this embodiment demonstrated below does not unduly limit the content of this invention described in the claim. In addition, all the configurations described in the present embodiment are not necessarily essential configuration requirements of the present invention.

1. First, the method of this embodiment will be described. Conventionally, in order to cause a robot to perform a desired operation, a technique using teaching by a user is widely known. Specifically, as described above, direct teaching, teaching playback, offline teaching, and the like are known, and a method close to a combination as disclosed in Patent Document 1 is also disclosed.

  However, in recent years, robots have been used in a very wide area. For this reason, it is assumed that there are many cases where a user who teaches a robot does not have sufficient expertise about the robot. In other words, it is desirable that teaching to the robot be performed in an easy-to-understand form so that even a user with insufficient expertise can execute it. In that respect, as described above, the method of Patent Document 1 is not preferable.

  On the other hand, with the method of Patent Document 2, it is not necessary for the user to designate the robot operation one by one using a teaching pendant or the like. Furthermore, the operation by the robot largely depends on a layout (work layout) representing the arrangement of work objects (work) in the work space. Therefore, when a user designates a layout or the like to be a target state, the system that automatically creates a teaching program that realizes the layout is highly useful. However, as described above, in Patent Document 2, when the layout is changed, the work by the user is required, so that the burden on the user is large and it cannot be said that it is a preferable method.

  In view of this, the present applicant proposes a control device that automatically creates a robot operation for moving a workpiece from the initial position to the final position when layout information including the initial position and final position of the workpiece is given. At that time, a plurality of planar local areas are set in the work space of the robot, and processing is performed in units of the set local areas. The local area may be set, for example, by dividing the work space into a planar grid.

  Specifically, as shown in FIG. 1, the control device 100 of the present embodiment includes a first position of a workpiece in a work space of a robot (corresponding to a robot main body 300 of FIG. 3 and the like described later), and the first The information acquisition part 110 which acquires the layout information containing the 2nd position of the workpiece | work different from a position, and the process part 120 which produces | generates robot operation information based on the acquired layout information are included. Then, the processing unit 120 sets a plurality of local regions on a predetermined plane in the work space, the correspondence between the first position of the workpiece and the local region, and the second position of the workpiece and the local region. By obtaining a correspondence relationship, work movement information is generated using the local area to represent the movement of the work from the first position to the second position, and is represented using the local area based on the work movement information. Generate robot motion information.

  Here, the first position of the work represents the position of the given work in the work space, and the second position is the position of the given work in the work space and is different from the first position. Represents the position. In the present embodiment, since the work is assumed to move from the first position to the second position, the work is at the first position at the first timing and is later than the first timing. It is in the second position at the second timing. Specifically, the first position of the work may be a position corresponding to an initial position representing the position of the work in the initial state of the work. Further, the second position of the workpiece may be a position corresponding to the final position representing the position of the workpiece in the final state (target state) of the work. In the following description, the first position is described as the initial position and the second position is the final position. However, the present invention is not limited to this, and the first and second positions are arbitrary workpieces in the work space. These two positions are represented.

  As will be described later, when the work A and the work B in the state of FIG. 5 are assembled and placed in the material removal area C so as to be in the state of FIG. 8, the initial position of the work is as shown in FIG. Represents the position of the workpiece A and the position of the workpiece B. Further, the final position of the work represents, for example, the position of the work A + B (the work A and the work B are assembled) in FIG. Alternatively, the initial position and the final position do not have to uniquely determine the position of the workpiece. For example, as shown in the material removal area C in FIG. 8, a region having a size larger than that of the workpiece may be used as information indicating the initial position or the final position. In the example of FIG. 8, the work A + B may be in any position in the material removal area C in the final state, and does not matter up to a fine position in the material removal area C.

  The layout information is information representing the position of the work in the work space, and is some information representing the state of the work as shown in FIGS. 5 and 8, for example. Specifically, it may be a captured image obtained by capturing the state of FIG. 5 or the like, information obtained from the captured image, or a coordinate value in a given coordinate system set in the work space. .

  Further, the workpiece movement information is information that expresses a path for moving the workpiece from the initial position to the final position in units of local areas, for example, as indicated by R1 and R2 in FIG. The robot motion information is information that expresses a robot motion for realizing the movement of the workpiece represented by the workpiece movement information. Even if the movement path of the workpiece is defined by the workpiece movement information, it can be realized by moving the workpiece with the right hand of the robot, moving it with the left hand, or rotating the waist while holding it with either arm It is unknown whether to do it. The robot operation information in this embodiment is information that defines the operation of the robot that realizes the movement of the workpiece, such as the determination of the arm to be used.

  Thus, when layout information including an initial position and a final position of a workpiece is input, it is possible to generate robot operation information representing a robot operation for moving the workpiece from the initial position to the final position. When the user inputs an initial state and a desired state (final state), the control device 100 automatically generates a movement path between them and a robot operation necessary for the movement. Therefore, it is possible for the user to cause the robot to perform a desired work on the work without complicated input. More broadly, when layout information including two positions of a workpiece is input, it is possible to generate robot operation information representing a robot operation for moving the workpiece between the two positions. .

  Furthermore, in this embodiment, work movement information and robot motion information are generated in units of local areas. By appropriately setting the size of the local region, it is possible to perform processing that places importance on the balance between the accuracy of information to be generated and the amount of calculation required for information generation. Specifically, the robot motion information can be generated at high speed by allowing a certain amount of error.

  Further, when the layout information is processed in units of local areas, it is only necessary to determine which local area corresponds to the position of each workpiece as shown in FIG. For example, as shown in FIG. 5, the layout of the work A, the work B, and the material removal area C on the work table is acquired as layout information, and as shown in FIG. When a divided local region is set, the layout information corresponding to FIG. 5 can be expressed as a column shown as “initial position” in FIG. 12 using the local region. That is, the process of dropping the layout information into the local area is very easy, and even if the input layout information changes, there is an advantage that adjustment work by the user is not required for the change.

  Although it is assumed that the robot motion is inherently three-dimensional, processing is performed using local regions separated by a two-dimensional grid, so that the robot motion creation processing can be performed at high speed. Become. Also, as described above, the robot motion creation process can be speeded up by appropriately setting the size of the local region.

  However, conversely, robot motions created by projecting robot motions that are originally three-dimensional to two-dimensional, or by processing robot motions that should be processed in smaller units in relatively large local areas. The information will be less accurate. The method of the present embodiment allows this, and presupposes that the robot motion creation process at a more detailed level is performed later. In other words, the robot motion creation process is divided into several steps, and relatively coarse robot motion information is created upstream, and relatively detailed robot motion information is created downstream using the upstream creation results. And the method of this embodiment assumes the upstream process.

  Note that the layout information in the present embodiment is information representing the placement of a workpiece and the like, and the movement information from the initial position of the workpiece to the final position is obtained from the layout information including the initial position and final position of the workpiece. become. At that time, when a robot having a plurality of arms such as a double-arm robot is targeted, there are several candidates for the arm used when moving the work. For example, it may be moved only by the right arm, may be moved only by the left arm, or may be changed in the middle. Further, if there are not only movements but also a work for assembling a plurality of workpieces, many combinations of arms to be used are conceivable.

  Therefore, in this embodiment, as shown in FIG. 4, the robot may include a plurality of arms. The robot motion information in this case is information indicating which arm of the plurality of arms is used to move the workpiece from the first position to the second position, as illustrated in FIG. is there.

  As a result, information indicating which arm of the robot is used to move the workpiece along the movement path represented by the workpiece movement information can be generated as the robot operation information. In a robot having a plurality of arms, such as a double-arm robot described later with reference to FIG. 4, the movement of a predetermined workpiece may be realized by either the first arm or the second arm. Actual robot control cannot be performed without deciding whether to use. In addition, the work can be moved with the first arm, but the work may not be able to move with the second arm because the position on the moving path cannot be reached. In that case, it is necessary to use the first arm. If robot operation information using the second arm is generated, robot control according to the information cannot be realized, which is not preferable. That is, in a multi-arm robot, determination of an arm is very important, and the method of this embodiment is useful in that it can be performed up to determination of an arm to be used for inputting layout information.

  Further, as described above, considering that the robot motion creation process is divided into a plurality of processes, it is necessary that the rough robot motion created in the upstream process can be realized by the target robot. . Usually, the robot operation information required when actually performing the robot control is very detailed. Therefore, it is difficult to obtain a processing result in a realistic time if a robot operation with such a granularity is searched in the entire search space. Therefore, when dividing the work process, it is important to limit the search space in the relatively rough process of the upstream process. That is, in the downstream process, a process for making the result obtained in the upstream process more detailed is performed, and generation of a robot operation that is significantly different from the result in the upstream process is not generally performed. In view of the above points, if a robot motion that cannot be realized by the target robot in the upstream process is created, no matter how detailed information is obtained in the subsequent downstream process, the resulting robot motion is Eventually, it becomes impossible to implement, and processing becomes useless. That is, in the upstream process, it is necessary to carefully determine whether or not the created robot operation can be realized.

  Therefore, in the present embodiment, processing for ensuring with high accuracy that the created robot operation can be realized may be performed. Specifically, as will be described later with reference to FIG. 11, the processing unit 120 of the control device 100 obtains the correspondence between the movable range of the arm and the local region for each arm of the plurality of arms, thereby The movable range information representing the movable range is obtained, and the robot motion information is generated based on the movable range information and the workpiece movement information.

  Here, the movable range of the arm represents a range in which a given position of the arm (in a narrow sense, the position of an end point or an end effector provided at the end point) can be reached. The movable range of the arm is determined as a design item of the robot based on the range of joint angles that can be taken by the joint included in the arm, and is stored in a format unrelated to the local region. For example, as shown in FIG. 11, it may be stored in the form of a given range for the robot body, or may be stored as a set of angle ranges where the joint angles of each joint included in the arm can be taken. Good. The processing unit 120 of the present embodiment performs processing for expressing the movable range in units of local regions. In the example shown in FIG. 11, the movable range information is B0E3 for the first arm (left arm) and B1E4 for the second arm (right arm). In this specification, the local area corresponding to the movable range information and the position of the workpiece is represented by arranging the upper left point of the local area and the lower right point of the local area. Can be implemented. For example, a local region ID that uniquely represents each local region may be prepared, and the initial position of the workpiece may be represented by arranging one or a plurality of the IDs.

  Thereby, the robot operation to be created can be realized by considering the movable range. Note that the robot operation that cannot be realized is, for example, an operation of moving a movable part (in a narrow sense, an arm and a hand provided at an arm tip or the like) out of a movable range. At this time, the movable range of the arm can be processed in units of local regions. That is, similar to the workpiece movement information, the robot movement information can be generated at high speed by allowing a certain amount of error in the movable range information. Note that even if an error is allowed, if it is possible to reach a position where the arm cannot originally reach, it is preferable because the generated robot motion information may become unrealizable in practice. Absent. Therefore, ideally, the region represented by the generated movable range information should be narrower than the movable range of the robot.

  Hereinafter, after describing a detailed configuration example of the control device and the robot system according to the present embodiment, details of the processing will be described.

2. System Configuration Example FIG. 2 shows a detailed system configuration example of the control device 100 according to the present embodiment. However, the control device 100 is not limited to the configuration in FIG. 2, and various modifications such as omitting some of these components or adding other components are possible.

  As illustrated in FIG. 2, the processing unit 120 of the control device 100 may include a local region setting unit 121, a work movement information generation unit 123, a movable range information generation unit 125, and a robot motion information generation unit 127. Further, the robot motion information generation unit 127 may include an operation cost calculation unit 129.

  The local area setting unit 121 sets a local area that is a unit of processing in the present embodiment. Specifically, the local area setting unit 121 sets a plurality of planar residence areas in the work space. The workpiece movement information generation unit 123 expresses the movement from the initial position to the final position of the workpiece using the local region by obtaining a correspondence relationship between the set local region and the workpiece position included in the layout information. Generated workpiece movement information.

  The movable range information generation unit 125 generates movable range information representing the movable range of the movable part of the robot. Since the movable range of the movable part is determined as a design matter of the robot, the movable range information generation unit 125 performs processing for expressing the movable range using a local region.

  The robot motion information generation unit 127 generates and outputs robot motion information based on the workpiece movement information and the movable range information. The robot operation information generation unit 127 may include a work cost calculation unit 129. When a plurality of robot operation information is obtained, the work cost calculation unit 129 calculates a work cost for each piece of robot operation information. In this case, the robot motion information generation unit 127 performs a process of determining a small number (one in a narrow sense) of robot motion information as an output for the subsequent unit based on the calculated work cost.

  Details of processing performed in each unit of the processing unit 120 illustrated in FIG. 2 will be described later. As described above, the method of this embodiment corresponds to the upstream process when a series of processes for determining the robot operation is realized in a plurality of processes. Therefore, the robot operation information output as the processing result of this embodiment is used for actual robot control after some processing is performed by the subsequent unit. This subsequent unit may be included in the control device 100 according to the present embodiment, or may be included in another electronic device.

  Further, as shown in FIG. 3, the method of the present embodiment can be applied to a robot system including the control device 100 and a robot (robot main body 300). In the configuration of FIG. 3, the control device 100 of this embodiment is a unit that generates more detailed information for robot control based on the robot motion information in addition to the information acquisition unit 110 and the processing unit 120 of FIG. (The latter-stage unit described above) and a robot control unit that controls the robot main body 300 based on the detailed information for robot control.

  The robot body 300 includes an arm 310 and an end effector 319 provided at the tip of the arm 310 or the like.

  Thus, the layout information is acquired to generate the robot motion information, and the control device 100 that generates the control information for the robot body 300 based on the robot motion information to perform the robot control, and operates according to the control of the control device 100. It is possible to realize a robot that performs the above as one system. In this robot system, the robot body 300 and the control device 100 may be provided separately as shown in FIG.

  Note that the configuration example of the robot system according to the present embodiment is not limited to FIG. For example, as shown in FIG. 4, the robot system may include a robot main body 300 and a base unit unit 400.

  The robot body 300 may be a double-arm robot, and includes a first arm 310 and a second arm 320 in addition to portions corresponding to a head and a torso. In FIG. 4, the first arm 310 includes joints 311 and 313 and frames 315 and 317 provided between the joints. The same applies to the second arm 320, but is not limited thereto. Further, end effectors 319 are provided at the tips of the first arm 310 and the second arm 320. Here, the end effector 319 may be a hand as shown in FIG. 4, or may be another tool or the like. Although FIG. 4 shows an example of a double-arm robot having two arms, the robot of this embodiment may have three or more arms.

  The base unit part 400 is provided at the lower part of the robot body 300 and supports the robot body 300. In the example of FIG. 4, the base unit 400 is provided with wheels and the like so that the entire robot can move. However, the base unit 400 may be configured to be fixed to a floor surface or the like without having wheels or the like. In the robot system of FIG. 4, the robot main body 300 and the control device 100 are integrally configured by storing the control device 100 in the base unit 400.

  Alternatively, the method of the present embodiment can be applied to a robot including the information acquisition unit 110 and the processing unit 120 described above. For example, the information acquisition unit 110 and the processing described above can be performed by a base (more specifically, an IC provided on the base) built in the robot without providing a specific control device like the control device 100. A robot that realizes the unit 120 and executes robot control is conceivable.

3. Details of Processing Next, details of processing according to the present embodiment will be described. Specifically, the details of the layout information acquired by the information acquisition unit 110 will be described, and the local region setting processing by the local region setting unit 121 will be described. Thereafter, the robot motion information generation process will be described. Specifically, the robot motion information generation process includes a movable range information generation process in the movable range information generation unit 125, a work movement information generation process in the work movement information generation unit 123, and a cost in the work cost calculation unit 129. This is realized by calculation processing. Finally, the processing flow of the present embodiment will be described with reference to the flowcharts of FIGS.

3.1 Acquisition of Layout Information First, details of layout information acquired by the information acquisition unit 110 will be described. Hereinafter, as described above, the work A and the work B at the positions shown in FIG. 5 are assembled to the work A on the way, and then the material is removed as shown in FIG. An example of performing work placed in area C will be described. At this time, the workpiece A is assembled with the workpiece A facing down, but the position suitable for the work in front of the robot is the initial position of the workpiece B. Therefore, as shown in FIG. 9, it is assumed that the work B is once moved to the back side with respect to the robot and the work A is moved to a free position.

  That is, the layout information may further include a third position of the workpiece that is different from the first position and the second position. In that case, the processing unit 120 of the control device 100 obtains the correspondence relationship between the third position of the workpiece and the local region, thereby moving the workpiece from the first position to the third position, and the third of the workpiece. The workpiece movement information representing the movement from the position to the second position using the local region is generated, and the robot motion information represented using the local region is generated based on the workpiece movement information.

  Here, the third position of the work is a position of the work in the work space, and represents a position different from the first position and the second position. In this embodiment, since the movement of the workpiece from the first position to the third position and the movement of the workpiece from the third position to the second position are assumed, the workpiece is in the third position. The third timing is a given timing that is after the first timing and before the second timing. Further, the third position of the workpiece may be a position corresponding to a midway position of the workpiece. The intermediate position of the workpiece is the position of the workpiece shown in FIG. 9 as described above, and information representing the state of FIG. 9 is included in the layout information. In the following description, the third position of the workpiece is assumed to be an intermediate position.

  As a result, information including the intermediate position of the workpiece is stored as layout information, and workpiece movement information passing through the intermediate position can be generated, and further, robot operation information can be generated according to the workpiece movement information. If the initial position and the final position of the workpiece are designated, there is a great advantage that the control device 100 automatically creates a route between them, but depending on the situation, not all the routes between them are allowed. For example, in the example of FIG. 9, as described above, the work B needs to be evacuated to secure the work place. Alternatively, an area where the workpiece does not want to enter may be set because there is some obstacle between the initial position and the final position, and in this case, a route that avoids the area where the workpiece does not want to enter is set. Must be generated. In the present embodiment, it is possible to include a midway position in the layout information, and by setting the midway position, it is possible to make the movement path of the workpiece more appropriate.

  The following description assumes that the layout information includes an intermediate position of the workpiece, and the layout information to be given to the control device 100 includes the initial position of the workpiece A and the initial position of the workpiece B shown in FIG. Information indicating the position of the material removal area C corresponding to the final position of the workpiece A + B illustrated in FIG. 8, and information indicating the intermediate position of the workpiece A and the intermediate position of the workpiece B illustrated in FIG. 9.

  Here, various modifications can be made as to how the information representing each position is formatted. For example, a work image may be arranged on an actual work table so as to be in the state of FIGS. 5, 8, and 9, and a captured image obtained by capturing the state from above the work table may be used. In this case, the control device 100 acquires a captured image, and specifies which image region of the captured image corresponds to the work space of the robot. For example, when the work table corresponds to the work space of the robot, an area where the work table is imaged is extracted from the captured image, and a position in which the work A or B is imaged is obtained in the extracted area. Thus, processing for specifying the initial position, final position, and midway position of the workpiece A or the like is performed.

  Alternatively, the actual work space is used, but the actual work does not have to be used. For example, the initial position of the workpiece may be acquired by indicating a point on the work space using some pointing device and specifying the specified point. In this case, the captured image may be captured as described above, and the layout information may be acquired by specifying the initial position and the like based on the position where the pointing device is captured in the captured image. Alternatively, by using a sensor or the like that detects the position of the pointing device itself, the position pointed to by the pointing device without using the captured image may be specified and acquired as layout information. For example, since position measurement systems using ultrasonic waves and magnetism are widely known, the position pointed to by the pointing device may be specified using these methods.

  Alternatively, the layout information may be generated on the simulator without using the actual work space or workpiece. For example, CAD data that reproduces the work space and the size and shape of the work may be created, and the work shown in FIG. 5 or the like may be reproduced by placing the work on a two-dimensional or three-dimensional simulation space. In that case, the information acquisition unit 110 acquires the layout information by specifying the position of each workpiece based on the information associating the simulation space with the actual work space.

  In addition, various modifications can be made to the information acquired by the information acquisition unit 110 and the method of specifying the initial position of the workpiece from the acquired information.

3.2 Local area setting process (work space partitioning process)
The local area setting unit 121 sets a plurality of planar local areas in the robot work space. An example of the local area to be set is shown in FIG. In FIG. 6, a 4 × 4 local area is set by dividing a work space corresponding to the work table by a plane lattice. Note that the length x in the vertical direction and the length y in the horizontal direction of each local region do not have to coincide with each other, and a rectangular local region may be set as shown in FIG. In addition, the vertical length X of the work space may not be an integral multiple of x, and the horizontal length Y of the work space may not be an integral multiple of y. Therefore, as shown in FIG. 6, some local regions may not have a size of vertical x horizontal y.

  In addition, it is assumed that the interval between the planar lattices is constant, and that each local region is basically the same size. However, the present invention is not limited to this, and local regions of different sizes may be set according to the position on the work space. Each local region is not limited to a lattice shape, and may be, for example, a triangle or a hexagon.

  In addition, a given fixed value may be used as the size of the local area to be set, but local area size setting processing may be performed based on the work size information included in the layout information. Specifically, the processing unit 120 may specify size information of a work that is a work target based on the layout information, and set the size of the local region based on the size information.

  This makes it possible to set the size of the local area based on the work size. In this embodiment, the robot motion information generation process is performed in units of local areas. Therefore, the size of the local area is directly related to processing accuracy and calculation amount (calculation time).

  For example, if the size of the local region is relatively small, the accuracy of processing using the local region can be improved. Specifically, if a local area of 1 cm × 1 cm is set for a work of about 10 cm in size, the position of the work represented by the local area and the movement path of the work are considerably high accuracy. Become. On the other hand, even if the position and movement path of the workpiece are slightly shifted (for example, about 1 cm), the local area corresponding to the position of the workpiece is different. It takes time to create information. As described above, the method of this embodiment corresponds to the upstream process in the robot motion determination process, and more detailed processing is performed in the subsequent unit. Therefore, in the processing steps corresponding to the present embodiment, the demand for accuracy is low, and rather high-speed processing is important. That is, the local region should not be overly small in size.

  On the other hand, if the size of the local area is made relatively large, the amount of calculation of processing using the local area is reduced, and the speed can be increased. Specifically, if a local region having a size of 20 cm × 20 cm is set, the number of local regions per unit area is 1/400 compared to a case where a local region of 1 cm × 1 cm is set. Therefore, as shown in R1 and R2 in FIG. 18 and the like, when determining a movement route in units of local regions, it is possible to reduce the number of local regions that are routed, or when a fine local region is used, it is recognized as a different route. A movement route can be processed as one movement route, and the amount of calculation can be reduced. On the other hand, the processing accuracy decreases as the processing unit increases. As a result, since the required movement route and robot operation are excessively rough information, the search space in the downstream process cannot be sufficiently limited, and the processing load on the subsequent unit responsible for the downstream process may increase. There is sex. In addition, if the processing accuracy is too low, there is a possibility that it is erroneously determined whether or not the created robot operation can be executed by the target robot.

  In view of the above points, the local region may be set to a size that balances accuracy and processing load. At this time, the accuracy of processing is determined not only by the size of the local area but also by the relative relationship between the local area size and the work size. For example, when a local area of 3 cm × 3 cm is used, the error is within a range that falls within 3 cm × 3 cm. At this time, if the target workpiece is as small as a screw and has a size of about 1 cm × 1 cm, the allowable error is very large with respect to the workpiece size. In this case, an error of 300% with respect to the workpiece size is allowed, so it cannot be said that the processing accuracy is sufficient. On the other hand, even when the same 3 cm × 3 cm local region is used, if the workpiece is large and has a size of about 30 cm × 30 cm, an allowable error is suppressed to about 10% with respect to the workpiece size.

  Therefore, in the present embodiment, the local area size setting process may be performed based on the workpiece size information. For example, a local area having a size of t% of the size of the smallest workpiece among the workpieces given as the layout information may be set. In this way, for each workpiece, an error based on the size of the workpiece can be suppressed to at most t%. Although various values of t can be considered, for example, a value such as 50 may be used.

3.3 Robot Motion Information Creation Processing Next, processing for creating robot motion information based on layout information and local areas will be described.

  First, the work movement information generation unit 123 performs a process of dropping the given layout information into the local area. Specifically, the initial position, final position, and midway position of the work are expressed using local regions.

  A specific example will be described with reference to the drawings. It is known from the layout information that the initial positions of the work A and the work B are the positions shown in FIG. Further, since the local region is a region divided by the plane lattice shown in FIG. 6, the combination of these local regions results in FIG. Thereby, it can be seen that the initial position of the workpiece A corresponds to one local region between B and C in the vertical direction and between 0 and 1 in the horizontal direction. Therefore, as shown in FIG. 12, the initial position of the workpiece A is expressed as B0C1 using the local region.

  Similarly, the initial position of the workpiece B can be expressed as C1D2. Similarly, for the work A + B in which the work A and the work B are assembled, the material removal area C is the final position, and the material removal area C is an area that spans four local areas as shown in FIG. Therefore, as shown in FIG. 12, the final position of the workpiece A + B can be expressed as B2D4. The same is true for the midway position. Since the midway position of the workpiece A spans two local areas, it is expressed as C1D3, and the midway position of the workpiece B is expressed as B1C2.

  Here, in order to shift from the intermediate state shown in FIG. 9 to the final state shown in FIG. That is, as shown in FIG. 10, a step of moving the workpiece B onto the workpiece A at the position C1D3 is also required, and FIG. 12 shows this operation in the second column from the right. That is, not only the intermediate position B1C2 shown in FIG. 9 but also C1D3 shown in FIG. 10 is used as the intermediate position for the workpiece B. 10 may be given as layout information, or may be created by the processing unit 120 of the control device 100 based on the information of FIGS. 8 and 9.

  With the above processing, information is obtained that the workpiece A may be moved B0C1-> C1D3, the workpiece B may be moved C1D2-> B1C2-> C1D3, and the workpiece A + B may be moved C1D3-> B2D4.

  Further, the workpiece movement information generation unit 123 generates one or a plurality of workpiece movement information by interpolating between the initial position, intermediate position, and final position of each workpiece as necessary. For example, a path of B0C1 → C1D2 → C1D3 passing through the local area represented by C1D2 may be created from the movement of workpiece A from B0C1 → C1D3, or B0C1 → B1C2 → passing through the local area represented by B1C2. A route called C1D3 may be created.

  The same applies to work B and work A + B. For example, C1D3 → B2D4 of the workpiece A + B may be interpolated as C1D3 → C2D3 → B2D4. In the example of FIG. 5 and the like, since the initial position and the intermediate position of each work or the intermediate position and the final position are close to each other, there may be a case where no interpolation process is particularly required as in the case of C1D2 → B1C2 → C1D3 of the work B.

  As shown in R1, R2, etc. in FIG. 18, even if the initial position, midway position, and final position are determined, the workpiece movement information obtained from these positions is not necessarily one. Therefore, the workpiece movement information generation unit 123 of the present embodiment may output a plurality of workpiece movement information.

  Next, movable range information generation processing will be described. The movable range information generation unit 125 generates the movable range information by expressing the movable range of the movable unit of the robot using a local region. For example, when the movable range of the first arm (here, the left arm) and the movable range of the second arm (right arm) of the target robot are the ranges shown in FIG. The movable range information is generated as B0E3 for one arm and B1E4 for the second arm.

  By obtaining the movable range information, it is possible to easily obtain a region that cannot be reached by any robot using the target robot. In the example of FIG. 11, the area represented by B0E4 can be reached by either the first or second arm, but the area represented by A0B4 in the work space is an area that cannot be operated by the target robot. It becomes. Therefore, even if the workpiece movement information generation unit 123 generates workpiece movement information passing through such an area, the robot movement information along the workpiece movement information is not executable. There is no advantage of processing.

  Therefore, in this embodiment, in order to suppress the generation of such unnecessary workpiece movement information, the movable range information obtained by the movable range information generation unit 125 is output to the workpiece movement information generation unit 123, and the workpiece movement is performed. The information generating unit 123 may generate work movement information using the input movable range information. Specifically, when the workpiece movement information is obtained by interpolating the initial position, halfway position, and final position of the workpiece (the movement path of the workpiece is obtained in units of local areas), a process that excludes a path that is out of the movable range is performed. Just do it. In this way, the movement of the work along the generated work movement information can be realized by using any arm of the robot.

  Moreover, you may use the input movable range information for determination, such as the initial position of a workpiece | work. If the obtained initial position or the like of the work is not included in the movable range of any arm (in FIG. 11, the position within A0B4), the work is performed in accordance with the layout information. I can't do that. In this case, the target robot may perform error processing such as determining that the input layout information is inappropriate and requesting re-input of the layout information.

  Based on the workpiece movement information and the movable range information generated by the above processing, the robot operation information generation unit 127 generates robot operation information. Here, the robot operation information is information corresponding to the robot operation that realizes the movement of the workpiece represented by the workpiece movement information. However, since more detailed processing is performed in the subsequent unit, the robot operation information of the present embodiment does not have to finely define the joint angle of the arm or the like. For example, the robot motion information of the present embodiment may be information obtained by specifying an arm that performs the movement of the workpiece represented by the workpiece movement information. That is, for the movement of the workpiece A from B0C1 to C1D3 (including movement between the interpolated B0C1 and C1D3), etc., it may be specified whether the movement is performed by the first arm or the second arm. . Therefore, when the workpiece A is moved from B0C1 to C1D3 using the first arm, it is not necessary to define how to control the angles of the joints included in the first arm.

  In selecting an arm to be used, it is necessary to use the first arm for movement including a position that can be reached by the first arm but cannot be reached by the second arm. Similarly, it is necessary to use the second arm for movement including a position that can be reached by the second arm but cannot be reached by the first arm. Therefore, of the movements included in the workpiece movement information, a position that can be reached by only one of the arms is determined to be moved by the reachable arm.

For example, the initial position B0C1 of the workpiece A can be reached with the first arm (left arm) but cannot be reached with the second arm. Therefore, in the example of FIG. 5, the movement of the workpiece A from B0C1 to C1D3 is performed by the first arm, and in this embodiment, the movement using the first arm is expressed as L (B0C1 → C1D3) _A. To do.

Similarly, the final position B2D4 of the arm A + B after assembly includes a region that can be reached by the second arm (right arm) but cannot be reached by the first arm. Therefore, in the example of FIG. 8, the movement of the workpiece A + B from C1D3 to B2D4 is performed by the second arm, and in this embodiment, the movement using the second arm is expressed as R (C1D3 → B2D4) _{A + B.} To do.

On the other hand, the movement at a position where the plurality of arms can be reached may be performed by the first arm or the second arm. For example, any arm may be used for (C1D2 → B1C2) _B and (B1C2 → C1D3) _B, etc., which are movements of the work B.

Further, as the order of work, in order to move the work A to the assembly work position C1D3, it is necessary to retract the work B from the initial position C1D2 to the intermediate position B1C2. Based on these, as shown in FIG. 13, the robot motion information of this embodiment is R (C1D2 → B1C2) _B , L (B0C1 → C1D3) _A , R (B1C2 → C1D3) _B , R (C1D3 → B2D4) ) _{A + B.}

FIGS. 14A to 14D illustrate the above flow. First, as an operation corresponding to R (C1D2 → B1C2) _B , as shown in FIG. 14A, the work B is moved from C1D2 to B1C2 using the right arm. Next, as an operation corresponding to L (B0C1 → C1D3) _A , as shown in FIG. 14B, the work A is moved from B0C1 to C1D3 using the left arm. Next, as an operation corresponding to R (B1C2 → C1D3) _B , as shown in FIG. 14C, the work B is moved from B1C2 to C1D3 using the right arm, and the assembling work is performed. Finally, as an operation corresponding to R (C1D3 → B2D4) _{A + B} , as shown in FIG. 14D, the work A + B after assembly work is moved from C1D3 to B2D4 (material removal area C) using the right arm. .

However, R (C1D2-> B1C2) _B in the work order 1 can be changed to L (C1D2-> B1C2) _B , and R (B1C2-> C1D3) _B in the work order 3 is changed to L (B1C2-> C1D3) _B It can be changed. In other words, four types of robot motion information candidates are obtained as the selection results of the left and right arms. Specifically, the operation in which R (C1D2 → B1C2) _B in the work order 1 is changed to L (C1D2 → B1C2) _B is as shown in FIGS. 15 (A) to 15 (D). When compared with FIG. 14A, the left arm is used for the operation of moving the workpiece B from C1D2 to B1C2 as shown in FIG. 15A. Similarly, the operations in which R (B1C2 → C1D3) _B in the work order 3 is changed to _B (B1C2 → C1D3) _B are as shown in FIGS. 16A to 16D, and both the work order 1 and the work order 3 are The changed operation is shown in FIGS. 17 (A) to 17 (D).

As described above, for each movement, a plurality of candidates for the interpolation path between the start point and the end point can be considered. For example, L (B0C1 → C1D3) _A may be L (B0C1 → C1D2 → C1D3) _A or L (B0C1 → B1C2 → C1D3) _A.

  Further, in the above description, an arm that is used with reference to the initial position, final position, and intermediate position before interpolation is set. For example, the movement of the workpiece A from B0C1 to C1D3 is always performed by one arm. However, the robot motion information creation process is not limited to this, and the arm to be used may be determined in smaller units. For example, when the movement of B0C1 → C1D3 is realized by the movement of B0C1 → C1D2 → C1D3, different arms may be used for the movement of B0C1 → C1D2 and the movement of C1D2 → C1D3.

  In view of the above points, a plurality of robot motion information is obtained as candidates for one layout information input, and in some cases, a very large number of robot motion information candidates may be acquired. . In the present embodiment, all the acquired robot operation information may be output.

  However, in view of the fact that the processing of this embodiment corresponds to an upstream process among a plurality of processing steps and requires a rough robot motion, the number of robot motions created is limited to a small number (one in a narrow sense). It is desirable that As described above, in order to reduce the calculation amount in the downstream process, it is important to limit the search space in the relatively rough process in the upstream process.

  Therefore, when a plurality of pieces of robot motion information are generated, the processing unit 120 of the control device 100 obtains a work evaluation value (work cost) for each of the plurality of robot motion information, as shown in FIG. 19 and FIG. One piece of robot motion information determined based on the obtained work evaluation value is output.

  This makes it possible to select appropriate information from among a plurality of robot motion information and to limit the number of robot motion information to be output. If the robot operation information is limited to one (or a small number that is two or more but less than a given threshold value), it is possible to suppress the processing load of the subsequent unit. Further, since the output is determined using the work cost, it is possible to leave more desirable robot operation information. Note that what kind of robot motion information is desired depends on what evaluation value is used as the work evaluation value. In the example described later with reference to FIG. 19, the one having a short moving distance (and the moving time if constant velocity motion is assumed) is selected with priority, and in the example in FIG. 20, the one having a short moving time is prioritized. To select. Alternatively, the work evaluation value may be calculated in consideration of the structure of the robot. For example, as shown in the flowchart shown in FIG. 23 to be described later, the work evaluation value is calculated for each of the first arm and the second arm. Also good. Specific examples are shown below.

  There are various methods for calculating the work cost. For example, the travel time may be used as the cost. For example, as shown in FIG. 18, when a given work is moved from an initial position to a final position through a halfway position, two routes indicated by R1 and R2 are candidates. At this time, if it is assumed that the acceleration / deceleration of the arm need not be taken into account at all, the time required for the movement of the work is proportional to the number of local regions that pass through. That is, as shown in FIG. 19, when the number of local areas in R1 is 21 and the number of local areas in R2 is 14, the cost of each route is the number of local areas itself, so in R1, 21 , R2 is 14. In this case, R2 may be output as R2 having a lower cost (shorter movement time) than R1.

  Further, the normal arm gradually accelerates from the speed 0 state to reach the maximum speed, and gradually decelerates even when stopped to reach the state where the speed is 0. That is, the speed change when acceleration / deceleration is not taken into consideration is as shown in FIG. 21A, but it is more accurate to process the speed change using trapezoidal approximation or the like as shown in FIG. It will be something. In other words, when considering acceleration / deceleration, the cost is not calculated based on the number of local areas, but changes in speed are taken into consideration. For example, a value corresponding to the speed may be assigned to each local region on the movement route as shown in FIG. In FIG. 20, the greater the value, the greater the speed, and the maximum speed is 4. In this case, since the time required to move the local area is represented by the reciprocal of the speed assigned to each local area, the work cost may be the sum of the reciprocals in the local area on the movement path. In the example in which acceleration / deceleration shown in FIG. 20 is taken into account, if the maximum speed can be maintained for a long time as in the case where the vehicle can move without stopping for a long distance, the cost can be reduced accordingly. Therefore, there is a possibility that a route having a longer moving distance may be less expensive than a route having a shorter moving distance.

  When performing the trapezoidal approximation in FIG. 21B, the acceleration and deceleration (that is, the slope of the trapezoid) may be constant, or may be changed according to user input or the like. For example, if the user places importance on moving the workpiece at high speed, the acceleration and deceleration may be set to the maximum value. The maximum value here is a value corresponding to the maximum acceleration and the maximum deceleration determined by, for example, the mechanical performance of a motor (an actuator in a broad sense) that drives a joint. On the other hand, there are cases where it is important to move slowly while avoiding sudden acceleration / deceleration, as in the case of moving a cup or the like containing liquid. In that case, a setting may be made to moderately accelerate and decelerate based on a user input “I want to move slowly”. If the acceleration and deceleration are changed, the speed value assigned to each local region also changes accordingly, and the calculated cost also changes. That is, the calculated work cost may be changed according to what viewpoint the user places importance on.

  Further, the work cost may be calculated using information such as the arm structure. For example, when an operation is performed with a given arm, the operation performed with the arm folded is easier than the operation performed with the arm extended. This is due to the fact that when the arm is extended, the hand position is far from the operating axis (rotating axis) of the arm and a larger moment is generated, which makes it difficult to stop at the desired position. That is, even when the movement is performed at a position that can be reached by either of the left and right arms, the robot operation information using the closer arm is easier to realize. Considering this point, even when a given workpiece is moved along a given route, the work cost calculated along the route depends on whether the first arm or the second arm is used. You may perform the work cost calculation process from which a value differs. Specifically, a calculation process is performed such that the work cost when the closer arm is used is smaller than the work cost when the far arm is used.

  In addition, when an arm includes a plurality of joints, an operation that can be realized by driving only one joint is more preferable than an operation that requires two or more joints to be driven in conjunction with each other. Easy to control. In the processing of the present embodiment, it is assumed that fine steps are not driven, but the movement of the hand by driving a single joint is circular (or a shape obtained by projecting a three-dimensional circle into two dimensions). This can be understood without detailed analysis of the arm structure. Therefore, the work cost calculation processing may be performed from the viewpoint of whether or not the movement path of the workpiece is close to a circle.

  In addition, the work cost calculation method and the robot operation information selection method using the calculated cost can be variously modified.

3.4 Process Flow of the Present Embodiment The process flow of the present embodiment will be described with reference to FIGS. FIG. 22 is a basic flow of processing of the present embodiment. When this processing is started, layout information corresponding to the start state is first acquired (S101). The start state is the state shown in FIG. 5, for example, and as the layout information, a captured image obtained by capturing the work A and the work B arranged at the position shown in FIG.

  Then, the work size information is acquired from the layout information acquired in S101 (S102), and the work space size is also acquired (S103). In the example of FIG. 5 and the like, the size of the work space corresponds to the size of the work table. Based on the workpiece size information acquired in S102, the size of the local area is set (S104). For example, as described above, the size of the local area may be ½ of the workpiece size (the minimum workpiece size among a plurality of workpieces in a narrow sense). Then, the work space is partitioned into local regions using the local region size (planar lattice spacing) set in S104 (S105). The processing result of S105 is, for example, as shown in FIG.

  Further, layout information corresponding to the final state is acquired (S106). The final state is, for example, the state shown in FIG. In the flowchart of FIG. 22, the layout information corresponding to the start state and the layout information corresponding to the end state are acquired at different timings. May be. Further, in the flowchart of FIG. 22, the intermediate state (intermediate state) corresponding to the intermediate position of the workpiece is not considered, but as described above, the intermediate position of the workpiece may be included in the layout information.

  Next, using the layout information acquired in S101 and S106 and the local area set in S105, the initial position and final position of the work are associated with the local area (S107). This corresponds to, for example, the process of acquiring information illustrated in FIG.

  At the same time, the movable range information is generated from the correspondence between the movable range of the movable part of the robot and the local region (S108). The movable range information is information such as B0E3 for the first arm and B1E4 for the second arm in the example of FIG.

  Work movement information is generated based on the information created in S107 and S108 (S109). In addition to the information shown in FIG. 12, the workpiece movement information in the present embodiment assumes information interpolated using a local region for the route between the initial position and the final position. For example, R1 and R2 in FIG. 18 may be used as work movement information.

  Then, the robot operation information is generated based on the movable range information and the workpiece movement information. This is realized by calculation processing (S110) of work cost (work evaluation value), creation (selection) of robot motion information based on the calculated work cost, and output processing (S111). In S111, a process of determining, as an output, robot motion information that minimizes the work cost calculated in S110 among a plurality of robot motion information candidates is performed.

  An example of a specific flow of the processing of S110 is shown in FIG. In the work cost calculation process, first, the number of local regions (moving masses) along the moving path is accumulated (S201). Here, the movement at a constant speed is assumed as shown in FIGS. 19 and 21A, but a speed change or the like may be considered as shown in FIGS. 20 and 21B. In that case, the evaluation values obtained for each local region may be integrated.

  Although the processing result of S201 may be the work cost, the work cost is obtained in consideration of the difference between the first and second arms in the flowchart of FIG. Specifically, each local area along the movement path is divided into a local area handled by the first arm and a local area handled by the second arm, and an integration process is performed for each arm (S202). Then, by obtaining the work cost for each arm, the work cost for the entire robot operation information to be processed is obtained (S203). At this time, the evaluation value assigned to each local region may be different depending on the arm used as described above.

  Note that the control device 100 or the like of the present embodiment may realize part or most of the processing by a program. In this case, the control device 100 according to the present embodiment is realized by a processor such as a CPU executing a program. Specifically, a program stored in a non-temporary information storage medium is read, and a processor such as a CPU executes the read program. Here, the information storage medium (computer-readable medium) stores programs, data, and the like, and functions as an optical disk (DVD, CD, etc.), HDD (hard disk drive), or memory (card type). It can be realized by memory, ROM, etc. A processor such as a CPU performs various processes according to the present embodiment based on a program (data) stored in the information storage medium. That is, in the information storage medium, a program for causing a computer (an apparatus including an operation unit, a processing unit, a storage unit, and an output unit) to function as each unit of the present embodiment (a program for causing the computer to execute processing of each unit) Is memorized.

  Although the present embodiment has been described in detail as described above, it will be easily understood by those skilled in the art that many modifications can be made without departing from the novel matters and effects of the present invention. Accordingly, all such modifications are intended to be included in the scope of the present invention. For example, a term described at least once together with a different term having a broader meaning or the same meaning in the specification or the drawings can be replaced with the different term in any part of the specification or the drawings. Further, the configuration and operation of the control device, the robot system, etc. are not limited to those described in the present embodiment, and various modifications can be made.

100 control device, 110 information acquisition unit, 120 processing unit, 121 local region setting unit,
123 workpiece movement information generation unit, 125 movable range information generation unit,
127 robot motion information generation unit, 129 work cost calculation unit, 300 robot body, 310 first arm, 311, 313 joint,
315, 317 frame, 319 end effector, 320 second arm,
400 Base unit

Claims (10)

An information acquisition unit that acquires layout information including a first position of the work in the work space of the robot and a second position of the work different from the first position;
A processing unit that generates robot motion information based on the acquired layout information;
Including
The processor is
Setting a plurality of local areas on a predetermined plane in the work space;
By obtaining a correspondence relationship between the first position of the workpiece and the local region and a correspondence relationship between the second position of the workpiece and the local region, the first position of the workpiece is determined from the first position. Generating movement information representing movement to the position of 2 using the local region,
A control device that generates the robot motion information represented by using the local region based on the workpiece movement information. In claim 1,
The robot includes a plurality of arms,
The robot operation information is
The control apparatus according to claim 1, wherein the control apparatus is information indicating which of the plurality of arms is used to move the workpiece from the first position to the second position. In claim 2,
The processor is
For each arm of the plurality of arms, by obtaining a correspondence relationship between the movable range of the arm and the local region, the movable range information representing the movable range using the local region is obtained,
A control device that generates the robot motion information based on the movable range information and the workpiece movement information. In any one of Claims 1 thru | or 3,
The layout information further includes a third position of the workpiece different from the first position and the second position,
The processor is
By obtaining a correspondence relationship between the third position of the workpiece and the local region, the movement of the workpiece from the first position to the third position, and the third position of the workpiece from the third position Generating the workpiece movement information representing movement to a second position using the local region;
A control device that generates the robot motion information represented by using the local region based on the workpiece movement information. In any one of Claims 1 thru | or 4,
The processor is
A control device that specifies size information of the work that is a work target based on the layout information, and sets the size of the local region based on the size information. In any one of Claims 1 thru | or 5,
The processor is
When a plurality of the robot motion information is generated, a work evaluation value for each of the plurality of robot motion information is obtained,
A control apparatus that outputs one piece of the robot motion information determined based on the work evaluation value. A control device according to any one of claims 1 to 6;
The robot;
A robot system characterized by including: An information acquisition unit for acquiring layout information including a first position of the work in a predetermined work space and a second position of the work different from the first position;
A processing unit that generates robot motion information based on the acquired layout information;
Including
The processor is
Setting a plurality of local areas on a predetermined plane in the work space;
By obtaining a correspondence relationship between the first position of the workpiece and the local region and a correspondence relationship between the second position of the workpiece and the local region, the first position of the workpiece is determined from the first position. Generating movement information representing movement to the position of 2 using the local region,
The robot motion information represented using the local area is generated based on the work movement information. Performing a process of acquiring layout information including a first position of the work in the work space of the robot and a second position of the work different from the first position;
Based on the acquired layout information, perform robot motion information generation processing for generating robot motion information,
As the robot motion information generation process,
Setting a plurality of local areas on a predetermined plane in the work space;
By obtaining a correspondence relationship between the first position of the workpiece and the local region and a correspondence relationship between the second position of the workpiece and the local region, the first position of the workpiece is determined from the first position. Generating movement information representing movement to the position of 2 using the local region,
A robot motion information generation method characterized by performing processing for generating the robot motion information represented by using the local area based on the workpiece movement information. An information acquisition unit that acquires layout information including a first position of the work in the work space of the robot and a second position of the work different from the first position;
As a processing unit that generates robot motion information based on the layout information,
Make the computer work,
The processor is
Setting a plurality of local areas on a predetermined plane in the work space;
By obtaining a correspondence relationship between the first position of the workpiece and the local region and a correspondence relationship between the second position of the workpiece and the local region, the first position of the workpiece is determined from the first position. Generating movement information representing movement to the position of 2 using the local region,
A program for generating the robot motion information represented by using the local area based on the work movement information.

Images