End effector selection device, end effector selection method, and program

Abstract

The present invention provides an end effector selection device that contributes to selecting an appropriate end effector with respect to a target environment in which work is to be performed by an end effector, a target article on which the work is to be performed by the end effector, and the content of the work. The end effector selection device comprises an end effector selection unit that receives information pertaining to an obstacle and an article arrangement in a target environment in which work is to be performed by an end effector and information pertaining to the operation of the end effector, and outputs, on the basis of the information pertaining to the obstacle and the article arrangement and the information pertaining to the operation of the end effector, information pertaining to the type of the end effector to be selected.

Description

End effector selection device, end effector selection method, and program

[0001] The present invention relates to an end effector selection device, an end effector selection method, and a program.

[0002] Regarding the selection of an end effector, the following literature can be cited:

[0003] Patent Document 1 relates to a robot simulation device that selects a robot hand shape based on the possibility of grasping a workpiece, which is calculated for multiple robot hand shapes based on a robot hand shape model and depth data.

[0004] JP 2015-100866 A

[0005] The following analysis is given by the inventor.

[0006] When controlling a robot arm to handle a target object (for example, picking, placing, pushing, etc.) using an end effector, an example is shown in Patent Document 1, in which an end effector is selected based on the shape of the target object and the object placement, such as the depth of the location where the target object is placed.

[0007] However, in addition to the shape and placement of the target object, the end effector required to handle the target object may change depending on, for example, the environment in which the target object is placed and the work to be performed on the target object.

[0008] One of the objects of the present invention is to provide an end effector selection device, an end effector selection method, a program, etc. that contribute to selecting an appropriate end effector for the environment in which the end effector will be used to perform work, the object on which the end effector will be used to perform work, and the work content.

[0009] According to a first aspect of the present invention, there is provided an end effector selection device including an end effector selection unit that receives information regarding obstacles and item locations in an environment where work is to be performed by an end effector, and information regarding the operation of the end effector, and outputs information regarding the type of end effector to select based on the information regarding the obstacles and item locations and the information regarding the operation of the end effector.

[0010] According to a second aspect of the present invention, there is provided an end effector selection method in which a computer included in an end effector selection device receives information on obstacles and item locations in an environment where work is to be performed by the end effector, and information on the operation of the end effector, and outputs information on the type of end effector to be selected based on the information on the obstacles and item locations and the information on the operation of the end effector. This method is linked to a specific machine, namely a computer, that executes the above method.

[0011] According to a third aspect of the present invention, a program can be provided that causes a computer included in an end effector selection device to execute the following processes: receiving information about obstacles and item locations in the environment where work is to be performed by the end effector, and information about the operation of the end effector; and outputting information about the type of end effector to be selected based on the information about the obstacles and item locations and the information about the operation of the end effector.

[0012] These programs can be recorded on a computer-readable storage medium. The storage medium can be a non-transitory medium such as a semiconductor memory, a hard disk, a magnetic recording medium, or an optical recording medium. The present invention can also be embodied as a computer program product.

[0013] According to the present invention, it is possible to provide an end effector selection device, an end effector selection method, a program, etc. that contribute to selecting an appropriate end effector for the environment on which the end effector is to be used to perform work, the object on which the end effector is to be used to perform work, and the work content.

[0014] 1 is a block diagram showing an example of a configuration of an end effector selection device according to the present disclosure. FIG. 1 is a block diagram showing an example of a configuration including additional components of an end effector selection device according to the present disclosure. FIG. 2 is a block diagram showing an example of an environment in which work is to be performed by an end effector according to the present disclosure, and an example of a sensor that acquires information about the environment. FIG. 3 is a block diagram showing an example of another configuration of an end effector selection device according to the present disclosure. FIG. 4 is a block diagram showing an example of a flowchart illustrating an example of an operation of an end effector selection device having another configuration according to the present disclosure. FIG. 5 is a block diagram showing an example of a different configuration of an end effector selection device according to the present disclosure. FIG. 6 is a block diagram showing an example of a modified configuration of an end effector selection device according to the present disclosure. FIG. 7 is a flowchart showing an example of an operation of an end effector selection device having a modified configuration according to the present disclosure. FIG. 8 is a diagram showing the configuration of a computer that constitutes an end effector selection device according to the present disclosure.

[0015] In this disclosure, the drawings may relate to one or more embodiments. In addition, each embodiment described below can be combined with other embodiments as appropriate, and the present invention is not limited to each embodiment.

[0016] First, an overview of one embodiment will be described with reference to the drawings. Note that the reference numerals in this overview are added to each element for convenience as an example to facilitate understanding, and are not intended to limit the present invention to the illustrated form. Furthermore, connecting lines between blocks in the drawings and the like referred to in the following description include both bidirectional and unidirectional lines. Unidirectional arrows are used to schematically indicate the flow of the main signal (data) and do not exclude bidirectionality.

[0017] 1 is a block diagram showing an example of the configuration of an end effector selection device 100 according to the present disclosure. Referring to FIG. 1, the end effector selection device 100 includes an end effector selection unit 110.

[0018] The end effector selection unit 110 receives information 111 about obstacles and item locations in the environment where work is to be performed by the end effector, and information 112 about the operation of the end effector, and outputs information 113 about the type of end effector to be selected based on the information 111 about obstacles and item locations and the information 112 about the operation of the end effector. Note that the information 112 about the operation of the end effector includes information about the object on which work is to be performed by the end effector and information about the work content.

[0019] As will be described later with reference to FIG. 3, information 111 regarding obstacles and item placement in the environment where work is to be performed by the end effector may be extracted by providing a sensor information processing unit 65 downstream of the sensor 60 shown in FIG. 3, and inputting the information 111 regarding obstacles and item placement in the environment where work is to be performed by the end effector to the end effector selection device 100 shown in FIG. 1 as information 111 regarding obstacles and item placement in the environment where work is to be performed by the end effector shown in FIGS. 1 and 2.

[0020] As described above, according to one embodiment, it is possible to provide an end effector selection device, an end effector selection method, a program, etc. that contribute to selecting an appropriate end effector for the environment on which work is to be performed by the end effector, the object on which work is to be performed by the end effector, and the work content.

[0021] [First Embodiment] Next, a first embodiment will be described in detail with reference to the drawings. Fig. 2 is a block diagram showing an example of a configuration including additional components of an end effector selection device according to the present disclosure. Fig. 2 shows a configuration including a relationship information storage unit 130 that is referenced by the end effector selection unit 110 of the end effector selection device 100 shown in Fig. 1 to select an end effector.

[0022] 2, the end effector selection device 100B includes an end effector selection unit 110B and a relationship information storage unit 130. Note that information 112B regarding an object on which work is to be performed by the end effector and information regarding the work content, which are included in information 112 regarding the operation of the end effector, may be input to the end effector selection device 100B shown in FIG. 2. The relationship information storage unit 130 may also be configured to be located outside the end effector selection device 100B and to be connected to the end effector selection unit 110B via communication.

[0023] The relationship information storage unit 130 stores relationship information between the type of end effector, the target item and task content on which the end effector is to perform the work, and the obstacles and item placement. Note that the relationship information storage unit 130 may store a trained relationship determination model generated by training using training data that includes, as correct answer labels, information on the target item and task content on which the end effector is to perform the work, information on the obstacles and item placement, and information on the type of end effector. The relationship information may also be a trained relationship determination model.

[0024] The end effector selection unit 110B uses the relationship information stored in the relationship information storage unit 130 regarding the type of end effector, the target item and work content on which the end effector is to be used, and obstacles and item placement, to output information 113 regarding the type of end effector to be selected based on the input information regarding the target item, information regarding the work content 112B, and information regarding obstacles and item placement 111.

[0025] As will be described later with reference to FIG. 3, information 111 regarding obstacles and item placement in the environment where work is to be performed by the end effector may be extracted by providing a sensor information processing unit 65 downstream of the sensor 60 shown in FIG. 3, and inputting the information 111 regarding obstacles and item placement in the environment where work is to be performed by the end effector to the end effector selection device 100B shown in FIG. 2 as information 111 regarding obstacles and item placement in the environment where work is to be performed by the end effector shown in FIGS. 1 and 2.

[0026] As described above, according to the first embodiment, it is possible to provide an end effector selection device, an end effector selection method, a program, etc. that contribute to selecting an appropriate end effector for the environment on which work is to be performed by the end effector, the object on which work is to be performed by the end effector, and the work content.

[0027] [Second Embodiment] Next, a second embodiment will be described in detail with reference to the drawings. Fig. 3 is a block diagram showing an example of an environment in which work is to be performed by an end effector according to the present disclosure and a sensor that acquires information about the environment. Referring to Fig. 3, an environment 10 in which work is to be performed and a sensor 60 that acquires information about the environment are shown as an example.

[0028] The environment 10 where work is to be performed includes a shelf 20, target items 30, 31, and 32, a robot arm 40, and an end effector 50 arranged at the tip of the robot arm 40. Also arranged on the shelf 20 are target items 30, 31, and 32 on which work is to be performed by the end effector. As an example, it is assumed that a shelf 21 and pillars 22 and 23 of the shelf 20 are obstacles to work by the robot arm 40, and the arrangement of the shelf 21 and pillars 22 and 23 is information about obstacle arrangement. It is also assumed that the arrangement of the target items 30, 31, and 32 is information about item arrangement.

[0029] For example, the sensor 60 may be a camera that acquires images. For example, the sensor 60 acquires still images or videos of the environment 10 where work is to be performed, and outputs the images as information (sensor information) 70 about the environment where work is to be performed by the end effector 50. The information (sensor information) 70 about the environment where work is to be performed by the end effector may be point cloud data (also referred to as point cloud). The information (sensor information) 70 about the environment where work is to be performed by the end effector includes information about obstacles and item locations. A sensor information processing unit 65 may be provided downstream of the sensor 60 to extract information 111 about obstacles and item locations and input the extracted information 111 to the end effector selection device 100 in FIG. 1 and the end effector selection device 100B in FIG. 2 as information 111 about obstacles and item locations in the environment where work is to be performed by the end effector, as shown in FIGS. 1 and 2 .

[0030] Fig. 4 is a block diagram showing another example of the configuration of an end effector selection device according to the present disclosure. Referring to Fig. 4, the end effector selection device 100C includes an end effector selection unit 110C, a relationship information storage unit 130, a simulator construction unit 140, and an operation possibility evaluation unit 150. The relationship information storage unit 130 may be configured to be located outside the end effector selection device 100C and connected to the end effector selection unit 110C via communication.

[0031] Next, the operation of the end effector selection device 100C shown in Fig. 4 will be described with reference to Fig. 5. Fig. 5 is a flowchart showing an example of the operation of the end effector selection device 100C having another configuration according to the present disclosure. The operation starts in step S501.

[0032] Next, in step S502, information (sensor information) 70 about the environment in which work is to be performed by the end effector, a 3D CAD (3-dimensional computer-aided design) model 141 of the robot and end effector, and a 3D CAD model 142 about the environment and layout are input to the simulator construction unit 140. The simulator construction unit 140 constructs a simulator that simulates the three-dimensional layout of the environment in which work is to be performed, based on the input information (sensor information) 70 about the environment in which work is to be performed by the end effector, the 3D CAD model 141 of the robot and end effector, and the 3D CAD model 142 about the environment and layout. The constructed simulator simulates the three-dimensional layout, and the simulator construction unit 140 outputs information 111B about obstacles and item placement.

[0033] That is, the information 111B regarding obstacles and item placement is generated by a simulator that simulates the three-dimensional layout of the environment 10 where the work is to be performed, which is generated based on information (sensor information) 70 regarding the environment where the end effector is to perform the work. Note that, as an example, the simulator construction unit 140 may generate a scene graph or the like, which is environmental information, for the range that the robot can reach, and output it as the information 111B regarding obstacles and item placement. Note that, for the simulator that simulates the three-dimensional layout of the environment where the work is to be performed that is constructed in step S502, a simulator model 143 is output from the simulator construction unit 140 and sent to the operability evaluation unit 150, and the simulator model 143 is placed in the operability evaluation unit 150, and the simulator is executed.

[0034] Next, in step S503, the end effector selection unit 110C uses the relationship information stored in the relationship information storage unit 130 between the type of end effector, the target item and work content to be worked on by the end effector, and the obstacles and item placement to output information 113 regarding the type of end effector to be selected based on the input information 112B regarding the target item and work content and information 111B regarding the obstacles and item placement.

[0035] Next, in step S504, the operability evaluation unit 150 uses a model of the end effector corresponding to the information 113 regarding the type of end effector to be selected to evaluate the work content using a simulator located in the operability evaluation unit 150.

[0036] Next, in step S505, the operability evaluation unit 150 determines whether the evaluation of the task content satisfies a predetermined criterion. If the predetermined criterion is met (YES in S505), the process proceeds to step S506, where the operability evaluation unit 150 determines that the information 113 regarding the type of end effector to be selected is an operable type of end effector.

[0037] Next, the operability assessment unit 150 outputs information 113 about the type of end effector to be selected that has been determined to be an operable type of end effector in step S507. Next, the processing ends in step S509.

[0038] On the other hand, if the evaluation of the work content does not satisfy the predetermined criteria in step S505 (S505 NO), the process proceeds to step S508, and the operability evaluation unit 150 sends an instruction 152 to the end effector selection unit 110C to change the type of end effector to be selected.

[0039] Next, the process proceeds to step S503, where the end effector selection unit 110C, having received an instruction 152 to change the type of end effector to be selected, executes the same process as step S503 again, changes the type of end effector to be selected, and outputs information 113 regarding the changed type of end effector to be selected.

[0040] As described above, according to the second embodiment, it is possible to provide an end effector selection device, an end effector selection method, a program, etc. that contribute to selecting an appropriate end effector for the environment on which work is to be performed by the end effector, the object on which work is to be performed by the end effector, and the work content.

[0041] Third Embodiment Next, a third embodiment will be described in detail with reference to the drawings. It should be noted that the target environment 10 where work is to be performed and the sensor 60 that acquires information about the target environment are, as an example, the same as the target environment 10 where work is to be performed and the sensor 60 that acquires information about the target environment described with reference to FIG. 3 . FIG. 6 is a block diagram showing an example of a different configuration of an end effector selection device according to the present disclosure. Referring to FIG. 6 , the end effector selection device 100D includes an end effector selector 110D and an action condition model storage unit 160. The end effector selector 110D also includes an action feasibility verification unit 1110 and a selection unit 1140, and the action feasibility verification unit 1110 includes an action plan calculation unit 1120 and a feasibility determination unit 1130. In FIG. 6 , components with the same reference numerals as those in FIG. 4 indicate the same components. In addition, the operation condition model storage unit 160 may be configured to be located outside the end effector selection device 100D and to be connected via communication to the operation plan calculation unit 1120 of the operation possibility verification unit 1110 of the end effector selection unit 110D.

[0042] The action condition model storage unit 160 stores, as information on the action of the end effector, information on the target object on which the end effector performs work and information on the work content, and the action condition model 112C for each type of end effector among the action condition models 112C for each type of end effector. Note that the action condition model storage unit 160 may store, as the action condition model 112C, a learned action condition model generated for each end effector by training using training data including information on each target object and information on the placement of each obstacle and object, information on the initial state and terminal state of each end effector, and action condition evaluation values ​​as correct labels.

[0043] Next, the operation of the end effector selection device 100D shown in Fig. 6 will be described with reference to Fig. 7. Fig. 7 is a flowchart showing an example of the operation of the end effector selection device 100D having a different configuration according to the present disclosure. The operation starts in step S701.

[0044] Next, in step S702, the information 111 relating to the obstacle and item placement, the information 112B relating to the target item and the information 112B relating to the work content are input to the end effector selection device 100D.

[0045] Next, in step S703, the motion plan calculation unit 1120 determines a motion plan for each type of end effector from information 111 about obstacles and item locations in the environment where the end effector is to perform work, and information about the motion of the end effector. Note that, as an example, a motion planning problem is a problem where, by solving this problem, the motion of the end effector is output, and the motion plan can be determined by solving the motion planning problem. Furthermore, the information about the motion of the end effector may include information 112B about the item on which the end effector is to perform work, information about the work content, and a motion condition model 112C for each type of end effector.

[0046] Next, in step S704, the motion plan calculation unit 1120 calculates an evaluation value of an optimal motion plan for each of one or more types of end effectors.

[0047] Next, in step S705, the feasibility determination unit 1130 checks whether the actions in the action plan are feasible and updates the evaluation value according to the check result. For example, the evaluation value is updated so that if there is a collision, the evaluation value will be significantly lowered.

[0048] Next, in step S706, the selection unit 1140 selects the type of end effector having the largest updated evaluation value, and outputs information about the type of end effector to be selected.

[0049] Next, in step S707, the end effector selection device 100D outputs the information on the type of end effector to be selected output by the end effector selection unit 110D.

[0050] As described above, according to the third embodiment, it is possible to provide an end effector selection device, an end effector selection method, a program, etc. that contribute to selecting an appropriate end effector for the environment on which work is to be performed by the end effector, the object on which work is to be performed by the end effector, and the work content.

[0051] Fourth Embodiment Next, a fourth embodiment will be described in detail with reference to the drawings. It should be noted that the target environment 10 where work is to be performed and the sensor 60 that acquires information about the target environment are, as an example, the same as the target environment 10 where work is to be performed and the sensor 60 described with reference to FIG. 3 . FIG. 8 is a block diagram showing an example of the configuration of a modified example of an end effector selection device according to the present disclosure. The end effector selection device 100E shown in FIG. 8 differs from the end effector selection device 100D shown in FIG. 6 in the following respects. Referring to FIG. 8, the end effector selection device 100E includes an end effector selection unit 110E, a simulator construction unit 140, and an action condition model storage unit 160. In FIG. 8, components with the same reference numerals as those in FIGS. 4 and 6 indicate the same components.

[0052] In the end effector selection device 100D shown in FIG. 6 , information 111 related to obstacles and item placement in the environment on which work is to be performed by the end effector, information 112B related to the item on which work is to be performed by the end effector and information related to the work content, and information related to the operation of the end effector including an operation condition model 112C for each type of end effector are input to an operation plan calculation unit 1120 of an operation possibility verification unit 1110 of an end effector selection unit 110D, and the operation plan calculation unit 1120 is configured to determine an operation plan for each type of end effector.

[0053] In contrast to this, in the end effector selection device 100E shown in FIG. 8, a simulator model 143 is output from the simulator construction unit 140 to an operation plan calculation unit 1120E of an operation possibility verification unit 1110E of an end effector selection unit 110E, and the operation plan calculation unit 1120E is configured to determine an operation plan for each type of end effector from the simulator model 143, information on the target object on which work is to be performed by the end effector, information on the work content 112B, and information on the operation of the end effector including an operation condition model 112C for each type of end effector.

[0054] The simulator construction unit 140 may construct a model of the reachable range of the end effector or a model of the reachable range of a robot arm equipped with an end effector that performs work, and output these as the simulator model 143. That is, the simulator model 143 may include information equivalent to the information 111 regarding obstacles and item placement that is input to the end effector selection device 100D shown in Fig. 6. Furthermore, the simulator model 143 may be generated by the simulator construction unit 140 as, for example, a scene graph or the like that is environmental information for the reachable range of the robot.

[0055] In this case, the motion plan calculation unit 1120E may determine a motion plan for each type of end effector for one or more types of end effectors from the simulator model 143, which is a model of the range that the end effector can reach during work or a model of the range that a robot arm equipped with an end effector can reach during work, information on the object on which the end effector will work and information on the work content 112B, and information on the operation of the end effector including an operation condition model 112C for each type of end effector.

[0056] In addition, the feasibility determination unit 1130E may check whether the operation of the operation plan is feasible using a simulator model 143, which is a model of the range that a robot arm equipped with an end effector can reach during work, and update the evaluation value according to the confirmation result.

[0057] FIG. 9 is a flowchart showing an example of the operation of the end effector selection device 100E having the modified configuration according to the present disclosure.

[0058] Steps S904, S906, and S907 in the flowchart shown in Fig. 9 are the same as steps S704, S706, and S707 in the flowchart shown in Fig. 7, and steps S902, S903, and S905 in Fig. 9 are different from the flowchart shown in Fig. 7. The operation starts at step S901.

[0059] Next, in step S902, information (sensor information) 70 about the environment of the target where work is to be performed by the end effector, a 3D CAD (3 Dimensional Computer Aided Design) model 141 of the robot and end effector, and a 3D CAD model 142 about the environment and layout are input to the simulator construction unit 140. The simulator construction unit 140 constructs a simulator that simulates the three-dimensional layout of the environment of the target where work is to be performed, from the input information (sensor information) 70 about the environment of the target where work is to be performed by the end effector, the 3D CAD model 141 of the robot and end effector, and the 3D CAD model 142 about the environment and layout, and outputs a simulator model 143.

[0060] Next, in step S903, the motion plan calculation unit 1120E determines a motion plan for each type of end effector from the simulator model 143, information about the object on which the end effector is to work, information about the work content 112B, and information about the operation of the end effector including an operation condition model 112C for each type of end effector.

[0061] Next, in step S904, similar to step S704 in FIG. 7, the motion plan calculation unit 1120E calculates an evaluation value of an optimal motion plan for each of one or more types of end effectors.

[0062] Next, in step S905, the feasibility determination unit 1130E uses the simulator model 143 to check whether the operation of the operation plan is feasible, and updates the evaluation value according to the check result. For example, the evaluation value is updated so that if a collision occurs, the evaluation value will be significantly lowered.

[0063] Next, in step S906, similar to step S706 in FIG. 7, the selection unit 1140 selects the type of end effector having the largest updated evaluation value, and outputs information about the selected type of end effector.

[0064] Next, in step S907, similar to step S707 in Fig. 7, the end effector selection device 100E outputs the information on the type of end effector to be selected output by the end effector selection unit 110E. The process ends in step S908.

[0065] As described above, according to the fourth embodiment, it is possible to provide an end effector selection device, an end effector selection method, a program, etc. that contribute to selecting an appropriate end effector for the environment on which work is to be performed by the end effector, the object on which work is to be performed by the end effector, and the work content.

[0066] Although each embodiment of the present invention has been described above, the present invention is not limited to the above-described embodiments, and further modifications, substitutions, and adjustments can be made without departing from the basic technical concept of the present invention. For example, the configurations of the elements shown in the drawings are examples to aid in understanding the present invention, and the present invention is not limited to the configurations shown in these drawings. Furthermore, "A and / or B" is used to mean at least either A or B.

[0067] Furthermore, the procedures described in the above-described embodiment and the first to fourth embodiments can be realized by a program that causes a computer (9000 in FIG. 10) functioning as the end effector selection device according to the present invention to realize the functions of the end effector selection device. Such a computer is exemplified by a configuration including a CPU (Central Processing Unit) 9010, a communication interface 9020, a memory 9030, and an auxiliary storage device 9040 in FIG. 10. That is, the CPU 9010 in FIG. 10 executes the control program for the end effector selection device, and updates the calculation parameters stored in the auxiliary storage device 9040, etc.

[0068] The memory 9030 is a RAM (Random Access Memory), a ROM (Read Only Memory), or the like.

[0069] That is, each part (processing means, function) of the end effector selection device shown in the above-mentioned embodiment and the first to fourth embodiments can be realized by a computer program that causes the processor of the above-mentioned computer to execute each of the above-mentioned processes using its hardware.

[0070] Finally, preferred embodiments of the present invention are summarized below. [First Embodiment] In an end effector selection device, an end effector selection unit may receive information regarding obstacles and item locations in an environment where work is to be performed by the end effector, and information regarding the operation of the end effector. The end effector selection unit may output information regarding the type of end effector to be selected based on the information regarding the obstacles and item locations and the information regarding the operation of the end effector. [Second Embodiment] In the end effector selection device according to the first embodiment, the information regarding the operation of the end effector preferably includes information regarding the object on which work is to be performed by the end effector, and information regarding the work content. [Third Mode] The end effector selection device according to the second mode preferably further includes a relationship information storage unit, wherein the end effector selection unit uses relationship information between the type of end effector, the target object and task content on which the end effector is to be operated, and the obstacles and item placement stored in the relationship information storage unit to output information about the type of end effector to be selected based on the input information about the target object, the task content, and the obstacles and item placement. [Fourth Mode] In the end effector selection device according to the third mode, the relationship information storage unit preferably stores a learned relationship determination model generated by training using training data including, as correct answer labels, information about the target object and task content on which the end effector is to be operated, information about the obstacles and item placement, and information about the type of end effector, wherein the relationship information is the learned relationship determination model.[Fifth Mode] The end effector selection device according to any one of the second to fourth modes is further configured such that the information regarding the obstacles and item placement is generated by a simulator that simulates a three-dimensional layout of the environment, which is generated based on information regarding the environment in which the end effector is to perform work, and further includes an operation feasibility evaluation unit that evaluates the work content using an end effector model corresponding to the type of end effector to be selected by the simulator, and the operation feasibility evaluation unit preferably outputs the type of end effector to be selected when the evaluation of the work content satisfies a predetermined criterion. [Sixth Mode] In the end effector selection device according to the first mode, it is preferable that the end effector selection unit includes a motion plan calculation unit, a feasibility determination unit, and a selection unit, the motion plan calculation unit determines a motion plan for each type of end effector from information on the obstacle and item placement and information on the motion of the end effector, the motion plan calculation unit calculates an evaluation value of an optimal motion plan for each of one or more types of end effector, the feasibility determination unit confirms whether the motion of the motion plan is feasible and updates the evaluation value in accordance with the confirmation result, and the selection unit selects the type of end effector with the largest updated evaluation value, and outputs information on the type of end effector to be selected. [Seventh Mode] In the end effector selection device described in the sixth mode, the information on obstacles and item placement is a simulator model of a simulator that simulates a three-dimensional layout of the environment, generated based on information on the environment of a target on which the end effector is to perform work; the motion plan calculation unit determines a motion plan for each type of end effector from the simulator model and information on the motion of the end effector; the information on the motion of the end effector includes information on a target item on which work is to be performed by the end effector, information on the work content, and an motion condition model for each type of end effector; and it is preferable that the feasibility determination unit uses the simulator model to confirm whether the motion of the motion plan is feasible, and updates the evaluation value in accordance with the confirmation result.[Eighth Mode] In the end effector selection device according to the seventh mode, the operation condition model for each type of end effector preferably includes a learned operation condition model generated by training using training data including, for each end effector, information on each target object, information on each obstacle and object location, information on the initial state and final state of each end effector, and an operation condition evaluation value as a correct label. [Ninth Mode] In the end effector selection method, a computer included in the end effector selection device may receive information on obstacles and object locations in an environment where an operation is to be performed by the end effector, and information on the operation of the end effector. The computer may output information on the type of end effector to be selected based on the information on the obstacles and object locations and the information on the operation of the end effector. [Tenth Mode] A program may cause a computer included in the end effector selection device to execute a process of receiving information on obstacles and object locations in an environment where an operation is to be performed by the end effector, and information on the operation of the end effector. The program may execute a process of outputting information regarding the type of end effector to be selected based on the information regarding the obstacle and object placement and the information regarding the operation of the end effector. Note that the ninth and tenth embodiments can be expanded into the second to eighth embodiments, similar to the first embodiment.

[0071] The disclosures of the above-cited patent documents are incorporated herein by reference. Modifications and adjustments of the embodiments and examples are possible within the scope of the entire disclosure of the present invention (including the claims), and further based on the basic technical concepts thereof. Furthermore, various combinations and selections of the various disclosed elements (including each element of each claim, each element of each embodiment or example, each element of each drawing, etc.) are possible within the scope of the disclosure of the present invention. In other words, the present invention naturally includes various modifications and alterations that would be possible by a person skilled in the art in accordance with the entire disclosure and technical concepts, including the claims. In particular, with regard to the numerical ranges described herein, any numerical value or subrange within the range should be construed as specifically described, even if not otherwise specified. Furthermore, the disclosures of the above-cited documents, when used in part or in whole in combination with the disclosures herein as part of the disclosure of the present invention, in accordance with the spirit of the present invention, are also deemed to be included in the disclosures of this application.

[0072] 10 Environment 20 Shelf 21 Shelf 22, 23 Pillar 30, 31, 32 Target item 40 Robot arm 50 End effector 60 Sensor 65 Processing unit 70 Information about the environment of the target on which work is performed by the end effector (sensor information) 100, 100B, 100C, 100D, 100E End effector selection device 110, 110B, 110C, 110D, 110E End effector selection unit 111, 111B Information about obstacles and item placement 112 Information about the operation of the end effector 112B Information about the target item and information about the work content 112C Operation condition model 130 Relationship information storage unit 140 Simulator construction unit 150 Operation possibility evaluation unit 160 Operation condition model storage unit 1110, 1110E Operation possibility verification unit 1120, 1120E Motion plan calculation unit 1130, 1130E Feasibility determination unit 1140 Selection unit 9000 Computer 9010 CPU 9020 Communication interface 9030 Memory 9040 Auxiliary storage device

Claims

1. An end effector selection device including an end effector selection unit that receives information regarding obstacles and item placement in the environment where work is to be performed by the end effector, and information regarding the operation of the end effector, and outputs information regarding the type of end effector to select based on the information regarding the obstacles and item placement and the information regarding the operation of the end effector.

2. The end effector selection device according to claim 1, wherein the information relating to the operation of the end effector includes information relating to an object on which work is to be performed by the end effector and information relating to the work content.

3. An end effector selection device as described in claim 2, further comprising a relationship information storage unit, wherein the end effector selection unit uses relationship information stored in the relationship information storage unit between the type of end effector, the target item and work content on which the end effector is to perform work, and the obstacles and item placement, to output information regarding the type of end effector to be selected based on the input information regarding the target item, information regarding the work content, and information regarding the obstacles and item placement.

4. The end effector selection device of claim 3, wherein the relationship information storage unit stores a learned relationship determination model generated by training using training data including, as correct answer labels, information on the target item on which the end effector is to work and information on the work content, information on obstacles and item placement, and information on the type of end effector, and the relationship information is the learned relationship determination model.

5. An end effector selection device as described in any one of claims 2 to 4, wherein the information regarding the obstacles and item placement is generated by a simulator that simulates a three-dimensional layout of the environment, which is generated based on information regarding the environment in which the end effector is to perform work, and further includes an operation feasibility evaluation unit that evaluates the work content using the simulator using an end effector model corresponding to the type of end effector to be selected, and the operation feasibility evaluation unit outputs the type of end effector to be selected if the evaluation of the work content meets a predetermined criterion.

6. The end effector selection device according to claim 1, wherein the end effector selection unit includes a motion plan calculation unit, a feasibility determination unit, and a selection unit, wherein the motion plan calculation unit determines a motion plan for each type of end effector from information on the obstacle and item placement and information on the motion of the end effector, the motion plan calculation unit calculates an evaluation value of the optimal motion plan for each type of end effector for one or more types of end effector, the feasibility determination unit checks whether the motion of the motion plan is feasible and updates the evaluation value in accordance with the check result, and the selection unit selects the type of end effector with the largest updated evaluation value and outputs information on the type of end effector to be selected.

7. The end effector selection device described in claim 6, wherein the information regarding the obstacles and item placement is a simulator model of a simulator that simulates a three-dimensional layout of the environment, generated based on information regarding the environment of the target on which the end effector is to perform work; the motion plan calculation unit determines a motion plan for each type of end effector from the simulator model and information regarding the motion of the end effector; the information regarding the motion of the end effector includes information regarding the target item on which the end effector is to perform work, information regarding the work content, and an motion condition model for each type of end effector; and the feasibility determination unit uses the simulator model to confirm whether the actions of the motion plan are feasible and updates the evaluation value in accordance with the confirmation result.

8. The end effector selection device of claim 7, wherein the operating condition model for each type of end effector includes a learned operating condition model generated by training using training data that includes, for each end effector, information on each target item and information on each obstacle and item placement, information on the initial state and terminal state of each end effector, and operating condition evaluation values ​​as correct labels.

9. An end effector selection method in which a computer included in an end effector selection device receives information about obstacles and item placement in the environment where work will be performed by the end effector, and information about the operation of the end effector, and outputs information about the type of end effector to select based on the information about obstacles and item placement and the information about the operation of the end effector.

10. A program that causes a computer included in an end effector selection device to perform the following processes: receiving information about obstacles and item placement in the environment where work is to be performed by the end effector, and information about the operation of the end effector; and outputting information about the type of end effector to be selected based on the information about obstacles and item placement and the information about the operation of the end effector.

Images