JP7228292B2 - Control device - Google Patents

Description

The present invention relates to control devices.

2. Description of the Related Art In recent years, in order to cope with the declining working population, etc., industrial robots have been actively developed to perform work that has been performed by natural humans.
For example, there has been proposed a technique of gripping and stacking cardboard boxes or the like using an article gripping mechanism including a manipulator (see, for example, Patent Document 1).

JP 2018-176313 A

However, the article to be gripped in the technique described in Patent Document 1 is a cardboard box. In other words, although conventional techniques including the technique described in Patent Literature 1 above were able to grip a standard-sized article, for example, they were sufficiently capable of stably gripping an irregular-shaped article. I couldn't say
That is, for example, it is difficult to grip an object having factors such as irregular shape, soft body, and adhesiveness. Here, an object having an atypical element is an object that does not have the same size or shape for each object. An object having an element such as a soft object is an object whose shape changes with a small amount of force. Also, an object having an element such as adhesive force is an object having adhesiveness.

SUMMARY OF THE INVENTION The present invention has been made in view of such circumstances, and an object of the present invention is to provide a technology capable of gripping an object having factors such as an atypical shape, a soft object, and an adhesive force.

In order to achieve the above object, a control device according to one aspect of the present invention includes:
A control device that executes control to grip an object having at least one element of an atypical shape, a soft object, and an adhesive force with an arm,
Execution of control for gripping a predetermined gripping weight of the object as a gripping target in consideration of the volume of a region surrounded by the arm when gripping the object;
gripping control means;
Prepare.

ADVANTAGE OF THE INVENTION According to this invention, the technique which can hold|grip the object which has factors, such as an atypical shape, a soft body, and adhesive force, can be provided.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram illustrating an overview of an example of an external configuration of a grasping system including a control device and an arm according to an embodiment of the present invention; FIG. 2 is a diagram showing an example of a grasping portion included in the arm of FIG. 1; 2 is a diagram showing an example of a control flow of the control device of FIG. 1; FIG. 1. It is a figure which shows an example, such as a vat which stores the foodstuffs grasped by the arm of FIG. 5 is a diagram showing an example of spaghetti height distribution in the bat of FIG. 4. FIG. It is a figure which shows an example of the operation|movement which grasp|grips food based on the distribution of FIG. It is a figure which shows an example of the operation|movement which discharge|ejects the foodstuffs hold|gripped by the operation|movement of FIG. 2 is a diagram showing an example of a configuration of an information processing system portion including the control device of FIG. 1 in the grasping system of FIG. 1; FIG. 9 is a block diagram showing an example of a hardware configuration of a control device in the information processing system of FIG. 8; FIG. FIG. 10 is a functional block diagram showing an example of the functional configuration of the control device of FIG. 9; FIG. 11 is a flowchart illustrating an example of the flow of gripping processing executed by the control device having the functional configuration of FIG. 10;

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

The control device of the present invention can perform control to cause the arm to grip an object having at least one element of the control device, atypical, soft object, and adhesive force.
Specifically, for example, in the embodiments described below, spaghetti is targeted for control as an object having at least one of atypical, soft, and sticky.
That is, the control device according to the embodiment of the present invention is capable of executing control for discharging spaghetti onto a plate or a lunch box after having the arm grasp spaghetti of a predetermined target weight from a vat containing a large amount of spaghetti. . The target weight is the weight of the object (here, spaghetti) that is gripped by the arm and discharged to a predetermined location (here, plate or lunch box), which is the weight targeted for control.

FIG. 1 is a diagram illustrating an outline of an example of the external configuration of a grasping system including a control device and an arm according to one embodiment of the present invention.

The gripping system shown in FIG. 1 is installed in a line in a food factory that discharges spaghetti placed in vats into plates or lunch boxes. This grasping system comprises a control device 1 according to one embodiment of the invention, an arm A, and cameras C1 and C2.

The control device 1 controls the arm A to grip a target weight of spaghetti S out of the spaghetti S stored in the vat B and to discharge the spaghetti to the plate D1 or D2. In addition, hereinafter, the plates D1 and D2 will be referred to as "dish D" when it is not necessary to distinguish them individually.
The arm A is a so-called manipulator, and has a grip portion A1 at the tip of a multi-joint structure. That is, the arm A has an arm side, which is an interconnected body of links and joints for the purpose of positioning the gripping portion A1, and a gripping portion A1.

The control device 1 uses, as parameters, the volume of the area surrounded by the gripping portion A1 of the arm A in which the spaghetti S is gripped and the density of the spaghetti S surrounded by the gripping portion A1 of the arm A to grip the spaghetti S in the gripping portion A1. Execute the control for grasping S.

That is, as shown in the following formula (1), the actual gripping weight M of the spaghetti S is determined by the volume V of the area surrounded by the arm A, the density ρ of the spaghetti S contained in the area, and the external factor C determined.

・・・（１）

... (1)

Here, the external factor C refers to variations that occur even though the arm A is similarly controlled. For example, an example of the external factor C is a change in the properties of the spaghetti S gripped by the arm A in FIG. be.

To summarize the above, basically, the gripping weight M is determined by the volume V and the density ρ of the area surrounded by the gripping portion A1 of the arm A. Conversely, in order to secure the target weight as the gripping weight M of the object to be gripped, in principle, the volume V and the density ρ of the enclosed area should be controlled. However, in order to accurately secure the target weight as the gripping weight M of the object to be gripped, it is necessary to perform control considering not only the volume V and the density ρ of the enclosed area but also the external factor C.
Therefore, the control device 1 is based on variably controlling the volume V and the density ρ of the enclosed area, and also performs control to reduce the influence of the external factor C. FIG. This makes it possible to appropriately discharge the spaghetti S of the target weight onto the plate D.
The details of the variable control of the volume V and the density ρ by the control device 1 and the control for reducing the influence of the external factor C will be described later.

A spaghetti S is an object to be gripped by the arm A. Specifically, the spaghetti S is an example of an object that is boiled in a process (not shown) in a food factory and has elements such as atypical shape, soft body, and stickiness.
Vat B is a container for storing spaghetti S. Although details will be described later, the vat B has a structure that allows the arrangement of the spaghetti S to be arranged.
The camera C1 takes an image of the vat B and the spaghetti S stored in the vat B as a subject. Data of an image obtained as a result of imaging by the camera C<b>1 is provided to the control device 1 .
Here, although only one camera C1 is shown in FIG. 1, there are actually two or more cameras. The distance (depth) to the subject (from the camera C1) can be calculated in pixel units based on the data of the captured images captured by each of the two or more cameras C1.

The control device 1 controls the arm A and the bat B based on the captured image data and the target weight provided from the camera C1 in this way. As a result, of the spaghetti S stored in the vat B, the target weight of the spaghetti S can be discharged.
The outline of control by such a control device 1 is as follows.

First, the spaghetti S is gripped by the gripper A1 of the arm A and taken out from the bat B, so the control device 1 grasps the arrangement of the spaghetti S in the bat B and selects the region to grip.
That is, the control device 1 acquires information on each depth of the bat B and the spaghetti S based on the data of the captured image from the camera C1. Further, the control device 1 selects an area for gripping the spaghetti S from the bat B based on the information on each depth of the bat B and the spaghetti S.
Here, when the control device 1 determines that the spaghetti S of the predetermined target weight cannot be gripped or when the spaghetti S of the target weight is actually not gripped, it is one of the controls for reducing the external factor C. By controlling the operation of the bat B, the arrangement of the spaghetti S can be arranged.

Next, the control device 1 executes control to change the position of the gripping portion A1 in order to allow the arm A to grip the spaghetti S in the selected region of the bat B. FIG. As a result, the gripping portion A1 of the arm A operates to grip the spaghetti S in the selected area of the bat B. As shown in FIG.

To summarize the above, the arrangement of vat B and spaghetti S stored in vat B varies. Therefore, the control device 1 executes control for changing the area gripped by the gripping portion A1 of the arm A and arranging the arrangement of the spaghetti S stored in the vat B, thereby achieving a predetermined target weight. of spaghetti S can be properly gripped by the arm A.

That is, until now, there has been a technique for separating solid foods such as potato chips that do not have an element of stickiness in a fixed quantity.
However, the prior art could not be applied to foods that have at least one factor of irregular shapes such as spaghetti S, soft bodies, and stickiness. In other words, the lines of many food factories were labor-intensive, not even mechanized. In other words, the task of sorting food items with factors such as atypical shapes, soft bodies, and stickiness in fixed quantities has not been fully automated, and human operators have been indispensable.
Therefore, the control device 1 of this embodiment is adopted. As a result, the controller 1 can automatically operate the industrial robot (arm A) to pick up a fixed amount (target weight) from the spaghetti S piled on the vat B and transfer it to a container (dish D). In other words, it is possible to automate the work of sorting food items having factors such as atypical shapes, soft bodies, and stickiness. Furthermore, it will help to compensate for labor shortages due to future population decline.

Here, the belt conveyor L conveys plates D1 and D2 as conveyed objects.
The dish D1 is the destination of the spaghetti S of the predetermined target weight that has already been discharged.
The plate D2 is the destination of the target weight of spaghetti S to be gripped from now on.
In this way, the belt conveyor L moves the dish D1 on which the predetermined target weight of spaghetti S that has already been discharged is placed, and moves the dish D2 to which the target weight of spaghetti S is next discharged. In this way, the belt conveyor L provides arm A with the next destination for the spaghetti.

The camera C2 takes an image of the dish D2, which is the next discharge destination, as a subject.
Data of an image obtained as a result of imaging by the camera C2 is provided to the control device 1. FIG.
Here, camera C2 is basically the same camera as camera C1 described above. That is, it is possible to calculate the distance (depth) to the subject (from the camera C2) in units of pixels based on the data of the captured images captured by each of the two or more cameras C2.

The control device 1 controls the arm A and the grasping of the arm A based on the data of the captured image provided from the camera C2 in this way and the discharge coordinates, which are the position coordinates of the discharge destination obtained by analyzing the data. It controls part A1. This enables the spaghetti S to be discharged to a discharge coordinate, such as the top of the plate D or a predetermined position on the plate D.
The outline of control by such a control device 1 is as follows.
First, the dish D to which the spaghetti S is to be discharged is not necessarily placed in the same position every time. Therefore, based on the depth information of the plate D, the control device 1 grasps the coordinates at which the gripped spaghetti S should be discharged.

Next, the control device 1 executes control to discharge the gripped spaghetti S to the grasped coordinates.
Summarizing the above, the position of the dish D into which the spaghetti S is served can vary. Therefore, the control device 1 analyzes the discharge coordinates at which the spaghetti S should be discharged based on the information on the depth of the dish D, and controls the arm A so that the spaghetti S is discharged at the discharge coordinates. of the target weight of spaghetti S can be discharged appropriately.

The example of the food factory implemented by the arm A controlled by the control device 1 according to one embodiment of the control device of the present invention has been described above with reference to FIG.
2 to 7, each configuration and operation of the grasping system shown in FIG. 1 will be described.

First, using states A and B shown in FIG. 2, the structure and operation of the grip portion A1 of the arm A in FIG. 1 will be described.
Although the details will be described later, the state A shown in FIG. 2 shows the grasping portion A1 of the arm A of FIG. 1 before the grasping operation. 2 shows the grasping portion A1 of the arm A of FIG. 1 after the grasping operation. When it is not necessary to distinguish between the A state shown in FIG. 2 and the B state shown in FIG. 2, they are collectively referred to as "FIG. 2".
FIG. 2 is a diagram showing an example of a grasping portion included in the arm of FIG. 1. FIG. Here, it is assumed that the gripping portion A1 of the arm A in FIG. 2 is arranged on the bat B in which the spaghetti S is stored as shown in FIG.

In the description of the grip portion A1 of the arm A in FIG. 2, directions defined as follows are used unless otherwise specified.
That is, hereinafter, a plane perpendicular to the direction in which gravity acts is called an Xa-Ya plane consisting of the axis Xa and the axis Ya. Here, it is assumed that the grasping portion A1 of the arm A is set so as to grasp the spaghetti S placed on the Xa-Ya plane (hereinafter referred to as "horizontal plane" as appropriate).
Here, on the Xa-Ya plane, when spaghetti S is gripped, the axis Xa is taken in the direction of connecting the arm side of the arm A when viewed from the gripping portion A1 of the arm A. In this case, when viewed from the grip portion A1 of the arm A, the direction in which the arm side of the arm A is connected is called the "positive direction of the axis Xa", and the opposite direction is called the "negative direction of the axis Xa". do.
Also, the axis Za and its direction are oriented in the direction opposite to the direction in which gravity acts. That is, the direction in which the spaghetti S is higher than the Xa-Ya plane (horizontal plane) on which the spaghetti S is arranged is called the "positive direction of the axis Za", and the opposite direction is called the "negative direction of the axis Za". shall be called.
In addition, the axis Ya is defined so that the three-dimensional orthogonal coordinate system consisting of the axis Xa, the axis Ya, and the axis Za becomes a right-handed system, and the respective directions are defined as "positive direction of the axis Ya" and "negative direction of the axis Ya". call.

In state A shown in FIG. 2, of the arm A in FIG. 1, the grasping portion A1 before the grasping operation is illustrated.
As in the state A shown in FIG. 2, the grip portion A1 of the arm A is composed of a surface A11, a surface A12, a surface A13, and a surface A14.
That is, the grasping portion A1 of the arm A has surfaces A11 and A13 parallel to the Ya-Za axis plane. Further, the surface A11 of the grip portion A1 is provided with the axis Xa in the positive direction when viewed from the surface A13 of the grip portion A1.
In addition, the gripping portion A1 of the arm A has surfaces A12 and A14 parallel to the Xa-Ya plane. Further, the surface A12 of the grip portion A1 is provided with the axis Xa in the positive direction when viewed from the surface A14 of the grip portion A1.

The plane A11 and the plane A12 are connected such that the side of the plane A11 on which the axis Za is positive and the side of the plane A12 on which the axis Xa is positive are in contact with each other. .
The surface A12 and the surface A13 are connected such that the side of the surface A12 on which the axis Xa is negative and the side of the surface A13 on which the axis Za is positive are in contact with each other. .
The plane A13 and the plane A14 are connected such that the side of the plane A13 with the negative axis Za and the side of the plane A14 with the negative axis Xa are in contact with each other. . In addition, surface A13 and surface A14 have a hinge in the edge|side which contacts. That is, the plane A14 is variable in angle parallel to the Ya-Za plane. Also, when the axis Za is inserted into the spaghetti S arranged in the negative direction when viewed from the grip portion A1 of the arm A (the axis Za is moved in the negative direction), the angle is controlled by the control device 1. That is, when the grip portion A1 of the arm A is inserted, the surface A14 is controlled to be parallel to the Ya-Za plane. Further, after the grip portion A1 of the arm A is inserted, the plane A14 is controlled to be parallel to the Xa-Ya plane.
Also, as in the state A shown in FIG. 2, the length from the side of the surface A12 in which the axis Xa is in the negative direction to the side where the surface A12 contacts the surface A11 is the length L1. Also, the length from the side of the surface A13 in which the axis Za is in the negative direction to the side where the surface A13 contacts the surface A12 is the length L2.

In the state B shown in FIG. 2, the grasping portion A1 of the arm A of FIG. 1 after the grasping operation is illustrated. For easy understanding, the spaghetti S to be gripped is not shown in the gripping portion A1 after the gripping operation shown in the state B shown in FIG.
Unlike the surface A11 shown in the state A shown in FIG. 2, the surface A11 shown in the state B shown in FIG.
Unlike the surface A12 shown in the state A shown in FIG. 2, the surface A12 shown in the state B shown in FIG. . That is, as in the state B shown in FIG. 2, the length from the side of the surface A12 on which the axis Xa is in the negative direction to the side where the surface A12 contacts the surface A11 is the length L3. At this time, the length L3 is shorter than the length L1.
Unlike the surface A13 shown in the state A shown in FIG. 2, the surface A13 shown in the state B shown in FIG. . That is, as in the state B shown in FIG. 2, the length from the side of the surface A13 in which the axis Za is in the negative direction to the side where the surface A13 contacts the surface A12 is the length L4. At this time, the length L4 is shorter than the length L2.

As explained in FIG. 1, the control device 1 can control the gripping weight by controlling the volume and density of the area surrounded by the gripping portion A1 of the arm A. FIG.
When a constant target weight of spaghetti S is to be dispensed onto the dish D, the controller 1 needs to confine the spaghetti S to the volume of the enclosed area with high reproducibility. That is, the control device 1 needs to confine the pasta so that the density of the enclosed area is constant. Density here is mass per volume. In addition, the density can be controlled within a certain range by applying a certain amount of pressure even to an object having elements such as atypical, soft, and cohesive.

Here, hereinafter, it will be explained that the density can be controlled within a certain range by applying a certain pressure even to an object having elements such as atypical, soft, and adhesive force.
For example, if objects with factors such as irregular shapes, soft objects, and adhesiveness placed in bat B are scooped out in a container with a predetermined volume and separated, the mass of the object will vary over multiple times. do.
Specifically, for example, when an object having an element of atypical shape is scooped out in a container of a predetermined volume and separated, due to reasons such as variations in the shape of the object, the space between the objects (the object does not exist) space) exists. In addition, for example, when an object having elements such as a soft object is scooped out in a container of a predetermined volume and separated, the shape of the object is not constant (deformation), etc., and the space between the objects (the object space) exists. Also, for example, when an object having an element such as adhesive force is scooped out in a container of a predetermined volume and separated, a space ( There exists a space where the object does not exist.
Furthermore, the above-mentioned space in which the object does not exist varies between multiple times. In other words, the mass of an object contained in a given volume fluctuates due to variations in the size of the space.

However, when a gripping system including a control device and arm A is employed, even objects with atypical, soft, and adhesive factors can be reduced to within a certain range of density by applying a constant pressure. can be controlled to
That is, for example, when the arm A grips an object, a predetermined pressure is generated on the object. Objects with stressed atypical, soft, and cohesive elements are arranged such that the space between them is reduced.
Specifically, for example, when pressure is applied to an object having an element of atypical shape, concave portions and convex portions, which are variations in the shape of the object, become stuck in each other due to the application of pressure. It is arranged so that the space between is reduced. Further, for example, when pressure is applied to an object having an element such as a soft object, the object is deformed or the like, so that the space between the objects is reduced. In addition, for example, when pressure is applied to an object having an element such as adhesive force, the object peels off or slides the adhesion between the objects or between the object and the container, thereby Arranged in such a way that it takes up less space.
Thus, when a gripping system including arm A is employed, it is possible to exert a constant pressure on objects having atypical, soft and sticky factors when gripping. In addition, objects that have elements such as irregular shapes, soft objects, and adhesive force applied with a certain pressure are arranged so that the space between the objects is reduced. This results in a range of densities for objects with atypical, soft and cohesive components.

Furthermore, when a grasping system including an arm A is employed, when using the arm to grasp an object having factors such as irregular shapes, soft objects, and adhesive force, the area surrounded by the grasping portion A1 of the arm A volume and density can be controlled. Thereby, the gripping system including the control device 1 and the arm A can control the gripping weight of the object.
Therefore, the gripping portion A1 of the arm A in FIG. 2 can control the volume and density of the enclosed area by the following structure and control.

The gripping portion A1 of the arm A in FIG. 2 can grip the spaghetti S by the volume V enclosed by the surfaces A11 to A14. That is, the control device 1 can grip the spaghetti S by controlling the arm A having the gripping portion A1.
Specifically, for example, the control device 1 executes the following control. First, such a grip A1 is inserted into the spaghetti S with the axis Za moved in the negative direction. Next, the grip part A1 changes the length L1 to the length L3 and the length L2 to the length L4. Further, the grip portion A1 is brought into substantial contact with the surface A14 by varying the mutual positions of the surfaces A11 to A14. In this way, the volume V holds the spaghetti S.
At this time, the control device 1 can change the position of the side where the surface A12 contacts the surface A11 and the position of the side where the surface A13 contacts the surface A12. In addition, the control device 1 can change the force for changing the mutual positions of the planes A11 to A14. Thereby, the volume V and the pressure for holding the spaghetti S can be changed.

Moreover, when changing the pressure, it is necessary to hold the control device 1 with constant pressure in the vertical direction (the direction of gravity) and the horizontal direction.
Further, the control device 1 performs the following control in order to grasp the spaghetti S with a constant density.

That is, the control device 1 maintains the state of the spaghetti S. Specifically, for example, the degree of boiling of the spaghetti S, the length of time after boiling, and the like change the elasticity and adhesive strength of the spaghetti S. In this case, the relationship between the pressure and the density when gripping the spaghetti S changes. Therefore, the control device 1 performs control so that the state of the spaghetti S is maintained.
Further, the control device 1 selects a place where the spaghetti S is to be held. That is, for example, as described above with reference to FIG. 1, based on the data of the imaged image obtained as a result of imaging by the camera C1, for example, a region in which spaghetti S is not reduced and arranged in a mountain shape is grasped. Select the target area.

The control device 1 also controls the force in the direction of gravity. That is, for example, by controlling the position of the surface A12 of the grip portion A1 of the arm A in FIG. 2, the force in the direction of gravity is controlled.
The controller 1 also controls lateral forces. That is, for example, by controlling the angle of the surface A11 of the grip portion A1 of the arm A in FIG. 2 and the position of the surface A13, the force in the lateral direction is controlled.

The structure of the gripping portion A1 that grips the spaghetti S is not limited to that shown in FIG. 2, and any structure can be adopted. That is, it is sufficient for the structure of the grasping part A1 to be able to change the volume V and the pressure for grasping the object to be grasped.
The control device 1 changes the volume V and the pressure for gripping the object to be gripped in order to grip the object to be gripped by a predetermined target weight. As a result, the control device 1 can grip an object having factors such as an atypical shape, a soft object, and an adhesive force by a predetermined target weight of the object to be gripped.

The structure and operation of the grip portion A1 of the arm A in FIG. 1 have been described above with reference to FIG.
3, a description will be given of how an article to be gripped is gripped by a predetermined target weight in the food factory shown in FIG.

3 is a diagram illustrating an example of a control flow of the control device of FIG. 1; FIG.
A series of white arrows shown in the horizontal direction in the upper part of FIG. That is, each of the plurality of white arrows indicates each step until the spaghetti S is served. The character string inside one white arrow describes the name of the process indicated by the white arrow.
Specifically, the process of gripping the spaghetti S includes the steps of "preparation", "recognition", "gripping", and "arranging".

The advance preparation step ST1 is a step of preparing spaghetti S to be gripped.
The recognition step ST2 is a step of recognizing the arrangement of the spaghetti S to be gripped.
The grasping step ST3 is a step of actually grasping spaghetti S by a predetermined target weight.
The serving step ST4 is a step of discharging (arranging) a predetermined target weight of spaghetti S on a plate D. As shown in FIG.

A series of white arrows shown in the horizontal direction at the bottom of FIG. 3 indicate specific steps in each process until the spaghetti S is discharged (arranged). That is, each of the plurality of white arrows at the bottom of FIG. 3 indicates specific steps until the spaghetti S is gripped and discharged (arranged). A character string inside one white arrow describes the name of the step indicated by the white arrow.

Specifically, as specific steps of the process until the spaghetti S is served, "state maintenance", "gripping area selection", "machine parameter determination", "food gripping", "disturbance elimination", "weight There are respective steps of "Confirm" and "Eject".

A state maintenance step ST11 exists as a step corresponding to the preparation step ST1.
That is, in the state maintenance step ST11, the control device 1 maintains the state of the spaghetti S to be gripped.
For example, the arrangement of the spaghetti S stored in the vat B may vary depending on the previous grasping or the like. Thus, for example, spaghetti S may be arranged in different amounts in different regions of vat B. FIG. Further, for example, the spaghetti S may be taken out by the previous grasping, and the placement of the spaghetti S may be different from the previous one. Such a difference in arrangement of the spaghetti S stored in the bat B can be said to be a change in the state of the spaghetti S to be gripped. That is, although details will be described later, in step ST11 for maintaining the state, the control device 1 can arrange the spaghetti S stored in the vat B in order to maintain the state of the spaghetti S to be gripped. That is, in step ST11 of maintaining the state, the control device 1 can physically eliminate external factors related to gripping of the spaghetti S of the target weight.

As a step corresponding to the recognition step ST2, there is a grasping area selection step ST21. That is, in step ST21 for gripping area selection, the control device 1 selects which area of the spaghetti S to grip from the bat B in which the spaghetti S to be gripped is stored. Although details will be described later, for example, there may be less spaghetti S in the area gripped in the previous gripping process. Therefore, based on the data of the captured image obtained as a result of the imaging by the camera C1, the control device 1 selects the region where the spaghetti S is not reduced, or more specifically, the region arranged in a mountain shape, as the region to be gripped. Select. That is, in step ST21 for gripping region selection, the control device 1 executes control for maintaining the state of spaghetti S before step ST3 for gripping, which will be described later. physical factors can be eliminated.

As steps corresponding to the gripping process ST3, there are a machine parameter determination step ST31, a food gripping step ST32, a disturbance elimination step ST33, and a weight confirmation step ST34.

First, in step ST31 for determining machine parameters, the control device 1 determines machine parameters for gripping spaghetti S for a predetermined target weight. That is, for example, the control device 1 determines the depth to which the spaghetti is to be inserted into the volume V of the grasping portion A1 by the specified volume. Further, for example, the control device 1 determines the volume V and the pressure of the grasping portion A1 for grasping spaghetti S for a predetermined target weight.

Next, in step ST32 of food gripping, the control device 1 executes control for gripping a predetermined target weight of spaghetti S based on the machine parameters determined in step ST31 of machine parameter determination. That is, for example, the control device 1 executes control for gripping the spaghetti S by changing the position of the gripping portion A1 or changing the volume V or the pressure of the gripping portion A1. That is, in step ST32 of food gripping, the control device 1 can physically eliminate external factors related to gripping of the target weight of spaghetti S while the control of gripping the spaghetti S is being executed. can.

Next, in step ST33 of food gripping, the control device 1 releases the spaghetti S gripped in excess of the target weight due to tangling or sticking out of the spaghetti S gripped in step ST32 of food gripping. Execute control. That is, for example, the control device 1 loosens the grip of the spaghetti S by shaking the gripping portion A1 of the arm A or changing the mutual positions of the surfaces A11 to A14 of the gripping portion A1 of the arm A. It is possible to release the spaghetti S gripped more than the target weight. That is, in step ST33 of food gripping, the control device 1 releases the part of the spaghetti S gripped by the arm A that exceeds the target weight from the state gripped by the arm A during the gripping operation by the arm A. It is possible to physically eliminate external factors related to gripping of the target weight of spaghetti S by executing the control to allow the weight to be increased.

Next, in a weight confirmation step ST34, the control device 1 confirms the weight of the spaghetti S held after the disturbance elimination step ST33. That is, for example, the control device 1 estimates how much spaghetti S is being gripped based on the pressure or the like required when changing the volume V of the gripping portion A1 of the arm A. FIG. That is, for example, the control device 1 actually grasps the spaghetti S based on the relationship between how the pressure changes by changing the volume V of the grasping portion A1 of the arm A and whether the spaghetti S is actually grasped. The weight of the cooked spaghetti S can be estimated.
Further, for example, the control device 1 may measure the weight of the actually gripped spaghetti S by placing the actually gripped spaghetti S on a weight scale (not shown). Further, for example, although not shown, the arm A may be provided with a weight scale such as a spring balance, and the control device 1 may acquire the weight of the spaghetti S measured by the weight scale and confirm the weight of the spaghetti S. .
It should be noted that it is preferable to estimate the actual weight based on the relationship between the volume V and the pressure, etc., instead of measuring the actual weight using a weighing scale. That is, when a weighing scale is used, it takes time until the numerical value of the weighing scale stabilizes. However, when estimating based on the relationship between the volume V and the pressure, etc., the actual weight can be estimated almost simultaneously with gripping control. That is, by estimating based on the relationship between the volume V and the pressure, etc., the processing time for the entire processing can be shortened.
In addition, in the weight confirmation step ST34, if the confirmed weight of the actually gripped spaghetti S does not fall within a predetermined range determined based on a predetermined target weight, it is determined to be NG. The process of step ST31 is executed again. That is, after executing control such as returning the actually gripped spaghetti S to the bat B, the control device 1 determines the machine parameters again and executes control for gripping the spaghetti S. FIG.
Further, in the weight confirmation step ST34, if the confirmed weight of the actually gripped spaghetti S is within a predetermined range determined based on a predetermined target weight, it is determined as OK, which will be described later. The process of step ST41 is executed.
That is, steps ST31 to ST34 are repeatedly executed until the confirmed weight of the spaghetti S actually gripped falls within a predetermined range determined based on a predetermined target weight.

A discharging step ST41 exists as a step corresponding to the arranging step ST4. That is, in the discharge step ST41, the control device 1 grasps the coordinates to be discharged of the spaghetti S based on the information of the depth of the belt conveyor L including the dishes D1 and D2, and executes the discharge control. Further, for example, when the spaghetti S is discharged, the control device 1 performs control for releasing the spaghetti S adhered to the arm A from the arm A by shaking the gripping portion A1 of the arm A. All of the spaghetti S can be physically eliminated. That is, the control device 1 releases the spaghetti S sticking to the arm A as much as possible among the gripped spaghetti S, so that the target weight of the spaghetti S can be discharged onto the tray D of the discharge destination. can.

In the above, with reference to FIG. 3, it has been described how to grip an article to be gripped by a predetermined target weight in the food factory shown in FIG.
4 and 5, how to select the region for gripping the spaghetti S in step ST21 for selecting the gripping region shown in FIG. 3 will be described.

As described above with reference to FIG. 3, in step ST21 for gripping area selection, the control device 1 selects an area for gripping the spaghetti S from the bat B in which the spaghetti S to be gripped is stored.
FIG. 4 is a diagram showing an example of a vat or the like that stores food gripped by the arms of FIG. 1 .
FIG. 4 shows a spaghetti S, a vat B in which the spaghetti S is stored, and a camera C1.
A bat B in FIG. 4 has a tray portion B1 and a push-in portion B2.
The bat B in the example of FIG. 4 is composed of five surfaces other than the top surface of the rectangular parallelepiped structure. The spaghetti S is stored from the upper surface of the vat B. The tray portion B1 of the bat B is composed of three of the five surfaces forming the bat B and the bottom surface. In addition, the pushing portion B2 of the bat B has one side surface of the five surfaces forming the bat B, and a bar that pushes the side surface closer to the opposite side surface. The stick of the pushing portion B2 of the bat B can push or pull the surface of the pushing portion B2 of the bat B by a predetermined mechanism.

4 and 5, unless otherwise specified, directions defined as follows are used.
That is, hereinafter, a plane perpendicular to the direction in which gravity acts is referred to as an Xb-Yb plane consisting of the axis Xb and the axis Yb. Here, it is assumed that the bat B is set on the Xb-Yb plane (hereinafter referred to as a "horizontal plane" as appropriate) and the spaghetti S is placed thereon.
Here, in the Xb-Yb plane, the direction of movement of the surface of the pushing portion B2 of the bat B described above is taken as the axis Xb. In this case, the direction in which the surface of the pushing portion B2 of the bat B after being pushed is called the "positive direction of the axis Xb" when viewed from the surface of the pushing portion B2 of the bat B before being pushed. The opposite direction shall be referred to as "the direction in which the axis Xb is negative".
Also, the axis Zb and its direction are oriented in the direction opposite to the direction in which gravity acts. That is, the direction in which the bat B is higher than the Xb-Yb plane (horizontal plane) is called the "positive direction of the axis Zb", and the opposite direction is called the "negative direction of the axis Zb". shall be called.
Further, the axis Yb is defined so that the three-dimensional orthogonal coordinate system consisting of the axis Xb, the axis Yb, and the axis Zb is a right-handed system, and the respective directions are defined as "positive direction of the axis Yb" and "negative direction of the axis Yb". call.

FIG. 4 illustrates bat B and camera C1 of FIG.
As shown in FIG. 4, the bat B is composed of a tray portion B1 and a pushing portion B2.
The bat B has a rectangular parallelepiped structure having sides parallel to the axis Xb, the axis Yb, and the axis Zb, and has no surface facing the positive direction of the axis Zb. In addition, by pushing in the surface of the pushing portion B2 of the bat B on which the axis Xb is in the negative direction, the distance between the surface on which the axis Xb is in the positive direction and the surface on which the axis Xb is in the negative direction can be shortened. . That is, the push-in portion B2 of the bat B can collect the spaghetti S by pushing in the surface where the axis Xb is in the negative direction, thereby narrowing the area where the spaghetti S is present.

As described above, the camera C1 takes an image of the vat B and the spaghetti S stored in the vat B as a subject. That is, an area where the amount of spaghetti S is not reduced, or in other words, an area arranged in a mountain shape is selected as an area to be gripped.
In the example of FIG. 4, among the spaghetti S stored in the vat B, there are many spaghetti S only in the area R indicated by the two-dot chain line where the axis Xb is in the positive direction and the axis Yb is in the negative direction.
That is, the arrangement of the spaghetti S based on the data of the imaged image obtained as a result of imaging by the camera C1 is obtained as information shown in FIG. 5, for example.

5 is a diagram showing an example of spaghetti height distribution in the bat of FIG. 4. FIG.
In state A shown in FIG. 5, heights (coordinates in the direction of axis Zb) based on data of captured images obtained as a result of imaging by camera C1 in FIG. FIG. 10 is a diagram showing an example of the distribution shown;

5, the first column from the positive direction of the axis Xb (the region where the axis Yb is constant), The first row from the positive direction (the area where the axis Xb is constant) and the first row from the negative direction of the axis Yb are all 15 in height. This height of 15 means that the bat B has a height of 15 in the direction of the axis Zb. That is, the numerical value of the area inside the area with a height of 15 is the numerical value of the height of the spaghetti S.
Thus, the height based on the data of the captured image obtained as a result of capturing by the camera C1 can be expressed as the distribution of the height of each region on the Xb-Yb plane.

State B shown in FIG. 5 is a diagram showing an example of distribution obtained by roughly calculating each region of the distribution of heights in state A shown in FIG.
That is, for example, the number of regions can be reduced by averaging the height distribution of each region on the Xb-Yb plane. Among the regions in state B shown in FIG. Among them, the spaghetti S with a height of 8.75 has the highest height. Therefore, based on this, the control device 1 can execute gripping control with the area having a height of 8.75 as the gripping area.
Further, for example, when there is no region suitable for selection as the gripping region, as described above, the push-in portion B2 of the bat B pushes the surface of the bat B in the negative direction of the axis Xb so that the region where the spaghetti S exists. Spaghetti S can be collected by narrowing it. Thereby, for example, spaghetti S can be collected like the height distribution shown in state C shown in FIG.

State C shown in FIG. 5 is an example of the height distribution of spaghetti S when the spaghetti S having the height distribution in state B shown in FIG. FIG. 4 is a diagram showing;
That is, for example, the control device 1 can collect the spaghetti S by pushing the area where the spaghetti S exists in half on the Xb-Yb plane.
Thus, the control device 1 acquires the arrangement of the spaghetti S based on the data of the captured image obtained as a result of being captured by the camera C1. In addition, the control device 1 can change the arrangement of the spaghetti S by controlling the push-in portion B2 of the bat B, for example, gathering the spaghetti S together. That is, the control device 1 can execute control for gripping a certain amount of spaghetti S by such control.

4 and 5, how to select the region for gripping the spaghetti S in step ST21 for selecting the gripping region shown in FIG. 3 has been described.
Hereinafter, how to eliminate the disturbance when the spaghetti S is grasped in the disturbance elimination step ST33 shown in FIG. 3 will be described with reference to FIG.

FIG. 6 is a diagram showing an example of the action of gripping food based on the distribution of FIG.
In the state A shown in FIG. 6, a portion of the spaghetti S stored in the vat B is gripped by the gripping portion A1 of the arm A, so that the spaghetti HS is placed in the area surrounded by the gripping portion A1 of the arm A. Fig. 3 shows an existing state;
State A shown in FIG. 6 is a state in which the spaghetti S in the gripping region R described in FIG. Here, in the state A shown in FIG. 6, the area surrounded by the arm A is further entangled with spaghetti HS1. In this state, the control device 1 swings the arm A, thereby executing control to release the spaghetti HS1 gripped in excess of the target weight due to tangling or sticking.

State B shown in FIG. 6 is a diagram showing an example of a state in which another spaghetti HS1 has fallen onto the bat B as a result of the control of swinging the arm A under the control of the control device 1 . In this way, the control device 1 executes the control to swing the arm A, thereby causing the spaghetti S to become entangled or sticky due to the spaghetti S having factors such as atypical shape, soft body, and stickiness. It is possible to eliminate the disturbance such as
In the above, using FIG. 6, how to eliminate the disturbance when the spaghetti S is grasped in the disturbance elimination step ST33 shown in FIG. 3 has been described.
Hereinafter, how to eliminate the disturbance when the spaghetti S is discharged in the discharging step ST41 shown in FIG. 3 will be described with reference to FIG.

FIG. 7 is a diagram showing an example of the operation of discharging the food gripped by the operation of FIG. 6. FIG.
A state shown in FIG. 7 is a diagram showing a state before the spaghetti HS gripped by the gripping portion A1 of the arm A is discharged onto the plate D. FIG.
In state A shown in FIG. 7, a dish D, an arm A for gripping spaghetti HS, and a camera C2 are shown. As described above, the camera C2 takes an image of the dish D2, which is the next destination, as a subject. The control device 1 grasps the coordinates at which the gripped spaghetti S should be discharged based on the data of the captured image obtained as a result of the imaging by the camera C2. The dashed-dotted line drawn on the plate D in state A shown in FIG. When the spaghetti HS is discharged, the control device 1 moves the arm A to a position where the spaghetti HS is discharged at the discharge coordinate T (for example, directly above the discharge coordinate T when the spaghetti HS is discharged directly below). to perform control. After that, the control device 1 executes control for discharging the spaghetti HS.

State B shown in FIG. 7 is a diagram showing a state after the spaghetti HS gripped by the arm A is discharged onto the plate D. FIG.
State B shown in FIG. 7 shows spaghetti HS that has been discharged onto a plate D, and spaghetti HS2 that has not been discharged together with the spaghetti HS and has adhered to the arm A. FIG. That is, spaghetti HS2, which has elements such as atypical, soft, and sticky, is attached to arm A. The control device 1 can physically eliminate external factors in the gripping control of the spaghetti S by executing the control to release the object attached to the arm A from the arm A. That is, for example, the control device 1 can release the adhered spaghetti HS2 from the arm A by vibrating the arm A. FIG.
In the above, using FIG. 7, how to eliminate the disturbance in discharging the spaghetti S in the discharging step ST41 shown in FIG. 3 has been described.
8 to 11, a gripping system including the control device 1 that executes the above-described control, the configuration of the control device 1, and the like will be described.

FIG. 8 is a diagram showing an example of a configuration of an information processing system portion including the control device of FIG. 1 in the grasping system of FIG.
The information processing system shown in FIG. 8 is configured to include a control device 1 , a model DB 2 and a learning device 3 .

The control device 1 acquires a model obtained (or updated) as a result of learning by the learning device 3 from the model DB 2 and determines the volume and density when the spaghetti S is gripped. Using the model, the control device 1 also predicts the weight actually gripped based on the volume and density of the spaghetti S gripped. In addition, the control device 1 acquires the model obtained (or updated) as a result of learning by the learning device 3 from the model DB 2, and calculates the difference between the predicted weight of the gripped spaghetti S and the target weight. Determine the parameters to be corrected.
The learning device 3 performs machine learning on the relationship between the volume, density, and pressure when gripping the spaghetti S from the actual gripped weight and the estimated gripped weight of the spaghetti S gripped as a result. It is generated or updated and stored in the model DB2.

9 is a block diagram showing an example of a hardware configuration of a control device in the information processing system of FIG. 8. FIG.

FIG. 9 is a block diagram showing an example of the hardware configuration of the control device 1 in the information processing system of FIG.

The control device 1 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a bus 14, an input/output interface 15, an output section 16, and an input section 17. , a storage unit 18 , a communication unit 19 , and a drive 20 .
The CPU 11 executes various processes according to programs recorded in the ROM 12 or programs loaded from the storage unit 18 to the RAM 13 .
The RAM 13 also stores data necessary for the CPU 11 to execute various processes.

The CPU 11 , ROM 12 and RAM 13 are interconnected via a bus 14 . An input/output interface 15 is also connected to this bus 14 . An output unit 16 , an input unit 17 , a storage unit 18 , a communication unit 19 and a drive 20 are connected to the input/output interface 15 .
The output unit 16 includes a display, a speaker, and the like, and outputs images and sounds.
The input unit 17 is composed of a keyboard, a mouse, etc., and inputs various kinds of information according to user's instruction operation.
The storage unit 18 is configured by a hard disk or the like, and stores data of various kinds of information.

The communication unit 19 controls communication with other terminals (for example, arm A or model DB 2 in FIG. 8) via a network.
A removable medium 31 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory is mounted in the drive 20 as appropriate. A program read from the removable medium 31 by the drive 20 is installed in the storage unit 18 as necessary. The removable medium 31 can also store various data stored in the storage unit 18 in the same manner as the storage unit 18 .

Although not shown, the learning device 3 of the information processing system shown in FIG. 8 has basically the same hardware configuration as shown in FIG.
Further, for convenience of explanation, the learning device 3 is provided separately from the control device 1, but it is not limited to this, and each function of the learning device 3 and the control device 1 can be integrated into one information processing device. may be aggregated.

10 is a functional block diagram showing an example of the functional configuration of the control device of FIG. 9. FIG.

As shown in FIG. 10S, in the CPU 11 of the control device 1, a region selection section 111, a grip control section 112, and a disturbance elimination section 113 function. A model DB 2 is provided in one area of the storage unit 18 .

Here, the model DB 2 stores models obtained (or updated) as a result of learning by the learning device 3, as described above. To simplify the explanation, the model DB 2 is assumed to be provided in the storage unit 18 of the control device 1 hereinafter.
However, the model DB 2 may not be provided in the control device 1, and the model may be stored in the learning device 3 and a copy of the model may be stored in the control device 1, for example. That is, the information processing device in which the model DB 2 is provided is arbitrary.

As described above, the camera C1 takes an image of the vat B and the spaghetti S stored in the vat B as a subject. Data of an image obtained as a result of imaging by the camera C<b>1 is provided to the control device 1 .

The area selection unit 111 recognizes the spaghetti S to be controlled before the gripping control unit 112 performs control to grip the spaghetti S in step ST21 for gripping area selection in FIG. A region including the spaghetti S to be gripped is selected based on. That is, the area selection unit 111 recognizes the arrangement of the bat B and the spaghetti S stored in the bat B based on the data of the captured image obtained as a result of the imaging by the camera C1, and based on the recognition result. to select a region including the spaghetti S to be gripped.
Specifically, for example, the area selection unit 111 selects an area with a large amount of spaghetti S or an area where it is easy to grasp the spaghetti S for the target weight, as described above with reference to FIGS. 4 and 5 .

Of the spaghetti S stored in the vat B, the gripping control unit 112 treats the gripping weight of the spaghetti S stored in the bat B as the gripping target, and considers the volume of the area surrounded by the arm A when gripping the spaghetti S, and grips the gripping target. to perform control. Further, the gripping control unit 112 considers the density of the spaghetti S inside the arm A when gripping the spaghetti S with a predetermined gripping weight out of the spaghetti S stored in the vat B, Execute control for gripping a grip target.

The gripping control unit 112 is provided with a machine parameter determination unit 121 , a gripping operation control unit 122 , a discharge control unit 124 and a weight confirmation unit 123 .

The machine parameter determination unit 121 determines the machine parameters used for the control of gripping the spaghetti S.
In addition, the machine parameter determining unit 121 can acquire the actual weight of the spaghetti S confirmed by the weight confirming unit 123, and determine the machine parameters to be used for subsequent control based on the actual weight.
Here, the parameters determined by the machine parameter determination unit 121 are the following parameters. 2, the angle at which the gripping portion A1 of the arm A is inserted into the spaghetti S, the mutual positions of the surfaces A11 to A14 of the gripping portion A1 of the arm A. , the pressure with which the surfaces A11 to A14 press the spaghetti S, the depth of insertion of the grip portion A1 of the arm A into the spaghetti S, and the like.
Note that the machine parameter determination unit 121 acquires a model related to determination of the machine parameters from the model DB 2 and determines the machine parameters.

The gripping motion control section 122 controls the gripping motion of the arm A based on the machine parameters determined by the machine parameter determination section 121 .
The gripping motion control section 122 has a gravity direction control section 131 and a lateral direction control section 132 .

The gravity direction control unit 131 controls the direction in which the spaghetti S is sandwiched between the surfaces A12 and A14 of the gripping portion A1 of the arm A based on the machine parameters determined by the machine parameter determination unit 121 .

The lateral direction control unit 132 controls the direction in which the spaghetti S is sandwiched between the surfaces A11 and A13 of the gripping portion A1 of the arm A based on the machine parameters determined by the machine parameter determination unit 121 .

The weight confirmation unit 133 confirms the actual weight of the grasped object grasped by the arm A under the control of the grasping operation control unit 122 .
That is, for example, the weight confirmation unit 133 acquires the relationship between the volume and the pressure of the gripped spaghetti S by controlling the gripping portion A1 of the arm A. FIG. Furthermore, the weight confirmation unit 133 reduces the volume of the gripped spaghetti S and acquires the pressure at that time. The weight confirmation unit 133 confirms by estimating the actual weight of the gripped spaghetti S based on the relationship between the volume of the gripped spaghetti S and the pressure thus acquired. In this case, the weight confirmation unit 133 acquires a model related to estimation of the actual weight of the spaghetti S from the model DB2, and estimates the actual weight of the spaghetti S.
Further, for example, the weight confirmation unit 133 discharges the gripped spaghetti S to a weight scale (not shown), and acquires the weight of the discharged spaghetti S obtained by the weight scale. That is, the weight confirming unit 133 confirms the actual weight of the gripped spaghetti S based on the measurement result of the weighing scale (not shown).

If the actual weight confirmed by the weight confirmation unit 133 is not within a predetermined range determined based on the target weight, the disturbance elimination unit 113, which will be described later, executes the elimination of the disturbance.

The ejection control unit 124 acquires the ejection coordinates T of the dish D, and after the gripping control unit 112 performs control, controls to eject the gripped object gripped by the arm A to the ejection coordinates T of the dish D. Execute.
Here, the camera C2 takes an image of the dish D2, which is the next destination, as described above. The data of the imaged image obtained as a result of imaging by the camera C2 is provided to the ejection control section 124, which will be described later. That is, the discharge control unit 124 acquires the discharge coordinates T on the plate D based on the data of the captured image obtained as a result of the imaging by the camera C2, and replaces the grasped object grasped by the arm A with the plate D control to discharge to the discharge coordinate T of . At this time, the discharge control unit 124 can acquire a model for estimating the discharge coordinates T on the plate D from the model DB2, and estimate the discharge coordinates T on the plate D from the model DB2.

The disturbance eliminator 113 eliminates external factors that change the gripped weight of the spaghetti S by at least one of physical and algorithmic factors.

The disturbance elimination unit 113 is provided with a pre-grip control disturbance elimination unit 141 and a grip control disturbance elimination unit 142 .

The pre-gripping control disturbance elimination unit 141 physically eliminates external factors by executing control for maintaining the state of the spaghetti S before the gripping control unit 112 executes control for gripping the gripping target. . That is, for example, the pre-gripping control disturbance elimination unit 141 physically eliminates external factors by executing control for maintaining a state such as the arrangement of spaghetti S stored in the bat B in FIG. Note that in this case, the pre-gripping control disturbance elimination unit 141 acquires from the model DB 2 a model related to estimation of the state of the spaghetti S based on the relationship between the state of the spaghetti S and the gripping weight. state can be estimated. Based on the state of the spaghetti S estimated in this way, the pre-gripping control disturbance elimination unit 141 executes control to maintain the state of the spaghetti S. FIG.

The grip control disturbance elimination unit 142 eliminates the external factors while the grip control unit 112 is performing control.
That is, for example, the disturbance elimination unit 142 during gripping control performs control to release the portion of the spaghetti S gripped by the arm A that exceeds the target weight from the state gripped by the arm A during the gripping operation by the arm A. to physically eliminate the external factors. That is, for example, when the spaghetti S is gripped more than the target weight under the control of the gripping operation control unit 122, which will be described later, the gripping control disturbance elimination unit 142 shakes the arm A while gripping the spaghetti S, thereby increasing the adhesive force. Part or all of the spaghetti S adhering to each other is removed. As a result, the grasping control disturbance elimination unit 142 releases the portion of the spaghetti S grasped by the arm A that exceeds the target weight from the grasping state by the arm A. At this time, the grasping control disturbance elimination unit 142 determines how much of the spaghetti S grasped by the arm A exceeds the target weight based on the actual weight confirmed by the weight confirmation unit 133 described above. can be obtained.
Further, for example, the gripping control disturbance eliminating unit 142 executes control for releasing the spaghetti S attached to the arm A from the arm A after the gripping operation by the arm A. Physically eliminate the cause. That is, for example, after the spaghetti S gripped on the plate D is discharged under the control of the discharge control unit 124, control for releasing the spaghetti S attached from the arm A due to the adhesive force of the spaghetti S is executed. Specifically, for example, the gripping control disturbance elimination unit 142 releases the spaghetti S attached to the arm A from the arm A by executing control to swing the arm A on the plate D. FIG.

FIG. 11 is a flowchart illustrating an example of the flow of gripping processing executed by the control device having the functional configuration of FIG. 10;
When the next dish D on which the spaghetti S is to be discharged is provided, the gripping process is started and the following steps S1 to S12 are executed.

That is, in step S1, the pre-grip control disturbance elimination unit 141 maintains the state. That is, in step S<b>1 , the pre-gripping control disturbance elimination unit 141 performs control to maintain the state of the spaghetti S before the gripping control unit 112 performs control to grip the gripping target, thereby eliminating external factors. are physically eliminated.

In step S2, the area selection unit 111 selects a gripping area. That is, in step S1, before the gripping control unit 112 performs gripping control for the spaghetti S, the spaghetti S to be gripped is recognized based on the recognition result, and the spaghetti S to be gripped is included. Select an area.

In step S3, the machine parameter determination unit 121 determines gripping parameters. That is, in step S3, the machine parameter determination unit 121 determines the machine parameters used for the control of gripping the spaghetti S.

In step S4, the pre-grip control disturbance elimination unit 141 prepares the grip region. That is, the pre-gripping control disturbance elimination unit 141 physically eliminates external factors by executing control to maintain the state of the spaghetti S stored in the vat B in FIG. 4, such as the arrangement and the degree of boiling.

In step S5, the gripping control unit 112 determines whether gripping control is performed for the first time in the trial in gripping processing.
Here, "the first trial" refers to the first trial of the grasping motion for discharging the spaghetti S onto the plate D. As shown in FIG. That is, although the details will be described later, when the weight prediction result of the gripped spaghetti S is not within the range, the gripping operation is performed again. In this case, the number of trials increases to the second or third time within the trial.
If it is the first time within the trial, it is determined as YES in the process of step S5, and the process proceeds to step S7.
If it is not the first time within the trial, it is determined to be NO in the process of step S5, and the process proceeds to step S6. When the process proceeds to step S6, the process proceeds to step S7 after the process of step S6. Therefore, in the following description, it is assumed that the determination in step S5 is NO.

In step S6, the machine parameter determination unit 121 corrects the gripping parameters. That is, the machine parameter determination unit 121 acquires the actual weight of the spaghetti S confirmed by the weight confirmation unit 123, and based on the actual weight, determines the machine parameters to be used for subsequent control. Here, the next and subsequent times refer to the number of times the gripping operation is performed again when the weight prediction result of the gripped spaghetti S, which will be described later, is not within the range, or when the spaghetti S is discharged to the dish D and the next dish is selected. Includes times to execute control for discharging to D.

In step S7, the gravity direction control section 131 controls the force in the direction of gravity. That is, the gravity direction control unit 131 controls the direction in which the spaghetti S is sandwiched between the surfaces A12 and A14 of the gripping portion A1 of the arm A based on the machine parameters determined by the machine parameter determination unit 121 .

In step S8, the lateral direction control section 132 controls the force in the lateral direction. That is, the lateral direction control unit 132 controls the direction in which the spaghetti S is sandwiched between the surfaces A11 and A13 of the gripping portion A1 of the arm A based on the machine parameters determined by the machine parameter determination unit 121 .

In step S9, the weight confirmation unit 123 determines whether or not the weight prediction result is within the range.
If the confirmed weight prediction result is within the range, YES is determined in the process of step S9, and the process ends.
If the confirmed weight prediction result is not within the range, NO is determined in the process of step S9, and the process proceeds to step S10. In the following description, it is assumed that the determination in step S9 is NO.

In step S10, the disturbance elimination unit 142 during gripping control determines whether or not adjustment is possible.
If it is determined that the adjustment is possible, YES is determined in the process of step S10, and the process proceeds to step S11.
If it is determined that the adjustment is not possible, NO is determined in the process of step S10, and the process proceeds to step S5. That is, if the adjustment is not possible, the gripping control disturbance elimination unit 142 discharges the gripped spaghetti S to the vat B, and proceeds to the process of step S5.
In the following description, it is assumed that YES is determined in the process of step S10.

In step S11, the discharge control unit releases the adhering amount. That is, for example, in step S11, after the spaghetti S gripped on the plate D is discharged, the discharge control unit 124 performs control to release the spaghetti S attached from the arm A due to the adhesive force of the spaghetti S.

In step S12, the weight confirmation unit 123 determines whether or not the discharged weight is within a predetermined range. That is, for example, the weight confirmation unit 123 determines whether or not the weight of the discharged spaghetti S is within a predetermined range based on the weight difference before and after the discharge acquired by the weight sensor provided on the arm A.
If it is determined that the discharged weight is within the predetermined range, the determination in step S12 is YES, and the process ends.
If it is determined that the discharged weight is not within the predetermined range, NO is determined in the process of step S12, and the process proceeds to step S5. That is, for example, if the discharged weight is not within the predetermined range, the spaghetti S discharged onto the plate D is not within the predetermined range based on the target weight, and thus the product is determined to be defective.
The weight of the discharged spaghetti S may be determined based on various information such as measurement of the weight of the dish D containing the discharged spaghetti S and the state of the arm A after the spaghetti S is discharged. .

The flowchart describing an example of the flow of the gripping process executed by the control device having the functional configuration of FIG. 10 has been described above with reference to FIG. 11 .

The embodiments of the grasping system to which the present invention is applied have been described above. However, embodiments to which the present invention is applied may be, for example, as follows.

In the present embodiment, the following control is performed in order to eliminate external factors that may cause fluctuations in the gripping weight of the spaghetti S, but the present invention is not particularly limited to this.

That is, for example, in the present embodiment, the control device 1 changes the volume V and the density of gripping the object to be gripped in order to grip the object to be gripped by a predetermined target weight. Not limited. That is, it is sufficient for the control device 1 to change at least one of the volume V and the density of gripping the object to be gripped in order to grip the object to be gripped by a predetermined target weight. That is, an object having at least one element of an atypical shape, a soft object, and an adhesive force can be gripped by a constant weight by controlling the volume V or pressure as a parameter. That is, it is sufficient for the control terminal 1 to control at least one of the volume V and the density in order to grip a predetermined target weight.
It is preferable that the control terminal 1 determines or changes the pressure for gripping the object to be gripped (here, the spaghetti S) so as to satisfy various conditions. Specifically, for example, the pressure for gripping the spaghetti S is preferably a pressure that does not cut the spaghetti S and a pressure that allows the spaghetti S to be lifted.

In addition, in the present embodiment, the following control is performed in order to eliminate external factors that may be a variable factor in the gripping weight of the spaghetti S, but the control is not particularly limited to this.

Further, for example, in the present embodiment, by moving the spaghetti S stored in the vat B, an external factor that causes a change in the gripping weight of the spaghetti S is eliminated, but the present invention is not particularly limited to this. That is, for example, a bat B that does not have the pushing portion B2 may be controlled to swing the entire bat. As a result, the spaghetti S stored in the vat B can be leveled evenly. That is, it is sufficient if the arrangement of the spaghetti S on the bat B can be changed.

Further, for example, in the present embodiment, the control terminal 1 determines the gripping area based on the height distribution of the spaghetti S on the bat B, but the present invention is not particularly limited to this. That is, for example, the control terminal 1 may determine the density of the spaghetti S based on the data of the captured image obtained as a result of imaging by the camera C1, and determine the gripping area based on the density information. Even in this case, it is possible to eliminate external factors in the control for discharging the target weight of gripped objects. That is, it is sufficient for the control terminal 1 to select the grip region based on the placement of the grip target.
Furthermore, in this embodiment, as shown in FIG. 4, the area R parallel to the Xb-Yb plane is selected as the gripping area, but it is not particularly limited to this. That is, for example, a region R with a predetermined height may be selected as the gripping region. Specifically, for example, the area selection unit 111 may select a three-dimensional gripping area at a position on the Xb-Yb plane of the bat B and a position on the axis Zb as the gripping area.
Also, for example, the grip region may be selected by the following algorithm. That is, for example, instead of determining the density based on the arrangement of the spaghetti S, the shape of the mountain of the spaghetti S is determined based on the height at each position, and by integrating the shape of the mountain, the volume of the spaghetti S may be calculated to select a three-dimensional grip region.
That is, various algorithms may be adopted as the gripping area selection algorithm depending on various conditions such as the amount of information processing related to selection of the gripping area and the accuracy of the camera C1.

In addition, the spaghetti S may have the following external factors. Specifically, for example, due to the entanglement of the noodles, excess noodles can be removed at a portion that is not in contact with the arm. In addition, for example, the density cannot be lifted while maintaining the density, and the density becomes sparse. Also, for example, the weight and adhesive strength of noodles differ between freshly boiled noodles and aged noodles. Also, for example, if the height of the noodles is low, it is not possible to press them down with sufficient force. Therefore, in order to eliminate these external factors, the control terminal 1, in any one of the steps before execution of the control for gripping the spaghetti S, during the gripping operation, and after the gripping operation, performs external By eliminating the factors, the gripped weight can be brought closer to the target weight.

Also, for example, in the present embodiment, spaghetti was used as an object having at least one of atypical, soft, and adhesive, but the following objects can also be used. That is, for example, foods having elements such as atypical, soft, and sticky include not only noodles such as the above-mentioned spaghetti S, but also a wide range of prepared dishes such as hijiki and meat. For example, in a food factory to which this embodiment is applied, the following foods can be gripped. That is, noodles (ramen, pasta, udon, etc.), perishable fruits and vegetables (tomatoes, peaches, strawberries, etc.), small side dishes (hijiki, dried daikon radish, bonito flakes, etc.), foods of different sizes (hamburgers, grilled fish, bread, etc.) etc.) can be grasped. Furthermore, for example, it may be an industrial product or the like. That is, the present invention can be applied to all fields other than food factories.

Further, for example, in the present embodiment, the structure and operation of the grasping portion A1 of the arm A have been described with reference to FIG. 2, but the present invention is not particularly limited to this. The gripping portion A1 of the arm A only needs to have a structure capable of controlling the volume and density of the area surrounded by the gripping portion A1 of the arm A.

Further, when the control device 1 determines that the spaghetti S of the predetermined target weight cannot be picked up or when the spaghetti S of the target weight cannot actually be picked up, the control device 1 controls the operation of the bat B to allow the spaghetti S to be picked up. Although it has been described that the arrangement of S can be adjusted, it is not particularly limited to this. That is, the manner in which the control device 1 controls the operation of the bat B to arrange the spaghetti S is not limited to the above case. Specifically, for example, the control device 1 may control the operation of the bat B to arrange the spaghetti S every time before selecting an area for gripping the spaghetti S from the bat B. FIG.

In addition, in the description of the state maintenance step ST11 and the pre-gripping control disturbance elimination unit 141, the controller 1 maintains the state of the spaghetti S to be gripped. It is not limited to the arrangement of the spaghetti S that has been made. Specifically, for example, the control device 1 can also maintain the state of the spaghetti S as follows.
For example, even if the spaghetti S is gripped with the same volume V and pressure, the over-cooked spaghetti S is gripped by a weight different from that of the properly boiled spaghetti S. The control device 1 confirms the weight of the spaghetti S actually gripped, and the control device detects whether or not the state of the spaghetti S is maintained based on the confirmed weight. When the state of spaghetti S is not maintained, the control device 1 feeds back the information "the state of spaghetti S is not maintained" to the previous step of the grasping system shown in FIG. Specifically, for example, when the spaghetti S has lost its elasticity, it can be determined that the reason why the elastic state is not maintained is that the spaghetti S has been overcooked. In this case, the control device 1 also feeds back information on the reason. As a result, adjustments such as shortening the time for boiling the spaghetti S are made in the process prior to the grasping system. As a result, in step ST11, the controller 1 detects that the state of the spaghetti S is maintained. The state of the spaghetti S is maintained by repeating the state maintenance step ST11.
That is, in step ST11 for maintaining the state, the control device 1 can perform control to maintain various states of the spaghetti S to be gripped, such as the quality of the spaghetti S such as the placement in the vat B, the boiling time, and the amount of oil. can.

That is, for example, although overcooking was given as an example of the reason why the state of spaghetti S is not maintained, the control terminal 1 discovers another reason and feeds back the reason to the previous process of the grasping system. It's okay. That is, it is preferable that the control terminal 1 can discover the reason why various states of the grasped object are not maintained. Even if the information that the state of the spaghetti S is not maintained is merely notified to the manager or the like of the gripping system, it is useful for accurately discharging the target weight of the spaghetti S.

Further, although the machine parameter determining unit 121 acquires a model related to determination of machine parameters from the model DB 2 and determines the machine parameters, the present invention is not particularly limited to this. That is, for example, in repeating gripping a plurality of times, the control terminal 1 may update the model using the results of gripping (parameters related to gripping control and results of checking the gripping weight). Also, at this time, the control terminal executes the next grasping control based on the updated model. By updating the model by repeating gripping a plurality of times in this manner, the following effects can be obtained.
That is, the state of the spaghetti S is maintained by eliminating external factors by maintaining the state. However, changes in the state of the spaghetti S, such as the boiling time, the elapsed time after boiling, and the amount of oil coated, occur not a little. Therefore, the control terminal 1 can execute gripping control adapted to the latest state of the spaghetti S by applying the results of gripping repeated multiple times to the model and updating the model.
Note that the model update process may be performed by executing a deep learning method in the learning device 3 by providing the learning device 3 with gripping result information from the control terminal 1 . In this case, it is preferable that the load of the CPU and the like for updating the model is provided by the learning device 3 instead of the control terminal 1 .

Further, for example, in the description of the present embodiment, in step S6 of FIG. 11, the machine parameter determining unit 121 corrects the gripping parameter, but the correction can be performed as follows.
That is, for example, for gripping parameter correction, a model related to gripping parameter correction can be acquired from the model DB 2 to determine gripping parameters (machine parameters). Specifically, for example, first, the weight confirmation unit 123 predicts the actual weight of the gripped spaghetti S based on the machine parameters before correction, the model, and the relationship between the volume and pressure of the gripped spaghetti S. The machine parameter determination unit 121 can determine arm parameters for correcting the difference between the predicted actual weight and the target weight as the correction of the gripping parameter.
Further, the gripping parameters corrected in step S6 of FIG. 11 may be various parameters such as the volume V, the pressure, the angle at which the gripping portion A1 of the arm A is inserted into the spaghetti S, and the like. Specifically, for example, a gripping parameter to be corrected may be a volume V or a density (pressure in control). However, of the volume V and the density, the volume V has less error in relation to the grip weight. Therefore, the volume V is preferable as the gripping parameter to be corrected.

Also, for example, in the above description, the camera C2 is assumed to capture the image of the dish D2, which is the next discharge destination, as a subject. In addition, the control device 1 controls the arm A and the position of the arm A based on the data of the captured image provided from the camera C2 and the discharge coordinates, which are the position coordinates of the discharge destination obtained by analyzing the data. Although the grip part A1 is controlled, it is not particularly limited to this. That is, for example, the discharge coordinates may be the position coordinates of the discharge destination specified in advance. Specifically, for example, the dish D2 may be conveyed to a fixed position by a belt conveyor L or the like. In this case, fixed positional coordinates where the dish D2 is transported may be input in advance as discharge coordinates.

Also, for example, the series of processes described above can be executed by hardware or by software.
In other words, the functional configuration described above is merely an example and is not particularly limited.
In other words, it is sufficient for the information processing system to have a function capable of executing the series of processes described above as a whole, and there is no particular limitation on what kind of functional block is used to realize this function. Also, the location of the functional block is not particularly limited, and may be arbitrary.
Also, one functional block may be composed of hardware alone, software alone, or a combination thereof.

When a series of processes is to be executed by software, a program constituting the software is installed in a computer or the like from a network or a recording medium.
The computer may be a computer built into dedicated hardware.
Also, the computer may be a computer capable of executing various functions by installing various programs, such as a server, a general-purpose smart phone, or a personal computer.

A recording medium containing such a program not only consists of a removable medium (not shown) that is distributed separately from the device main body in order to provide the program to the user, etc., but is also preinstalled in the device main body and stored in the user's memory. It is composed of a recording medium etc. provided for

In this specification, the steps of writing a program recorded on a recording medium are not necessarily processed chronologically according to the order, but may be executed in parallel or individually. It also includes the processing to be performed.
Further, in this specification, the term "system" means an overall device composed of a plurality of devices, a plurality of means, or the like.

In other words, the control device to which the present invention is applied can take various embodiments having the following configurations.

That is, the control device to which the present invention is applied (for example, the control device 1 in FIGS. 1 and 10) is
A control device that executes control to grip an object (e.g., spaghetti S in FIG. 1) with at least one element of an atypical shape, a soft object, and adhesive force with an arm (e.g., arm A in FIG. 1) ,
Among the objects, a predetermined target weight of the objects to be grasped (for example, the spaghetti HS grasped in FIG. 6) is grasped, and the volume of the area surrounded by the arm when grasping the object (for example, the volume V ), and perform control to grip the target to be gripped,
gripping control means (eg, gripping control unit 112 in FIG. 10);
is sufficient as long as the control device is provided with

This makes it possible to change the actually gripped volume of the gripped object having at least one of the atypical, soft, and adhesive forces. That is, the volume, which is one of the factors that change the weight of the gripped object, can be changed. As a result, the present invention can execute control for gripping the target weight as the gripping target.

Further, the control device to which the present invention is applied (for example, the control device 1 in FIGS. 1 and 10) is
A control device that executes control to grip an object (e.g., spaghetti S in FIG. 1) with at least one element of an atypical shape, a soft object, and adhesive force with an arm (e.g., arm A in FIG. 1) ,
Among the objects, a predetermined target weight is set as a grasped object (for example, the grasped spaghetti HS in FIG. 6), and the density of the object inside the arm when grasping the object (for example, the volume in FIG. Execute control to grip the gripping target in consideration of the pressure corresponding to the force when changing
gripping control means (eg, gripping control unit 112 in FIG. 10);
is sufficient as long as the control device is provided with

As a result, it is possible to change the actual gripping pressure of the gripping target object, which has at least one of atypical, soft, and adhesive force. That is, the pressure, which is one of the factors that change the weight of the gripped object, can be changed. As a result, the present invention can execute control for gripping the target weight as the gripping target.

Furthermore, the gripping control means
Furthermore, it is possible to perform control for gripping the gripping target in consideration of the volume of the area surrounded by the arm when gripping the object (for example, the volume in FIG. 2).

This makes it possible to change the actually gripped volume and density of the gripped object having at least one of atypical, soft, and cohesive. That is, it is possible to change the product of volume and density, which is a factor that changes the weight of the grasped object. As a result, the present invention can more accurately execute the control of gripping the target weight as the gripping target.

Furthermore, disturbance elimination means (for example, disturbance elimination section 113 in FIG. 10) that eliminates external factors that are fluctuation factors of the gripping weight of the object by at least one of physical and algorithmic policies,
can further comprise

As a result, it is possible to eliminate external factors at various timings such as before execution of gripping control, during the gripping operation, and after the gripping operation.

Furthermore, the disturbance elimination means is
By executing control for maintaining the state of the object (for example, the arrangement and boiling degree of spaghetti S stored in the vat B in FIG. 4) before the control is executed by the gripping control means, first disturbance elimination means (for example, pre-grip control disturbance elimination section 141 in FIG. 10) that physically eliminates the physical factor;
can have

As a result, it is possible to reduce or eliminate deviations in gripping control results due to physical reasons such as the state of the gripping target before it is gripped (for example, the presence or absence of overcooking and differences in placement).

Furthermore, the disturbance elimination means is
a second disturbance elimination means (for example, a grip control disturbance elimination section 142 in FIG. 10) that eliminates the external factor while the control is being executed by the grip control means;
can have

As a result, a physical state such as a state during gripping of the grip target (for example, a state in which the grip target object is over-gripped due to adhesion or entanglement, or a state in which the grip target object cannot be discharged due to adhesion). Deviations in grip control results due to reasons can be eliminated.

Furthermore, the second disturbance elimination means is
By executing control for canceling the state of gripping by the arm during the operation of gripping by the arm, of the gripped object of the gripping weight gripped by the arm that exceeds the target weight. , the external factors can be physically excluded.

As a result, it is possible to eliminate deviations in gripping control results due to physical reasons, such as the state during gripping operation of the gripping target (for example, the state of excessive gripping due to adhesion or entanglement of the gripping target object). .

Furthermore, the second disturbance elimination means is
By executing control for releasing the object attached to the arm from the arm after the gripping operation by the arm, the external factor in the control can be physically eliminated.

This makes it possible to eliminate deviations in gripping control results due to physical reasons such as the state after the gripping action of the gripping target (for example, a state in which the object to be gripped cannot be completely removed due to stickiness).

Furthermore, the gripping control means
machine parameter determination means (for example, machine parameter determination unit 121 in FIG. 10) for determining machine parameters used for the control;
gripping motion control means (for example, gripping control unit 112 in FIG. 10) for controlling gripping motion by the arm based on the determined machine parameter;
can have

Thereby, the control means can determine machine parameters for how to actually control the arm, and control the arm based on the machine parameters.

Furthermore, the gripping control means
weight confirmation means for confirming the actual weight of the gripped object gripped by the arm under the control of the gripping means (for example, weight confirmation unit 123 in FIG. 10);
further having
The machine parameter determining means,
It is possible to obtain the actual weight confirmed by the weight confirmation means, and determine the machine parameters to be used for the control from the next time onward based on the actual weight.

As a result, the control device determines the machine parameters again based on the actual weight and executes gripping control again, or determines the machine parameters for the next gripping control based on the latest gripping result. be able to.

Furthermore, before the control is performed by the grip control means, an area selection means (for example, a 10 region selection unit 111),
can further comprise

Thereby, the control device can control gripping based on the arrangement of the object to be controlled before gripping.

Furthermore, ejection means (e.g., ejection control unit 124 in FIG. 10) that executes control for ejecting the gripped object gripped by the arm to a predetermined location after the gripping control means performs the control;
can further comprise

Thereby, the control device can discharge the gripped object to a predetermined location by controlling the arm. That is, a series of processes of gripping and discharging can be performed.

Furthermore, the ejection means
Coordinates at which the predetermined location exists may be obtained, and the control for ejecting the grasped object at the coordinates may be executed.

Thereby, the control device can execute control for accurately ejecting to the coordinates of the ejection destination by controlling the arm.

DESCRIPTION OF SYMBOLS 1... Control apparatus, 11... CPU, 18... Storage part, 111... Area|region selection part, 112... Grasping control part, 113... Disturbance exclusion part, 121... Machine parameter Determination unit 122 Grasping operation control unit 123 Weight confirmation unit 124 Discharge control unit 131 Gravity direction control unit 132 Lateral direction control unit 141 Disturbance elimination unit before gripping control, 142 Disturbance elimination unit during gripping control

Claims (11)

A control device that executes control to grip an object having at least one element of an atypical shape, a soft object, and an adhesive force with an arm,
To set the volume of a region surrounded by the arm when gripping a predetermined gripping weight of the object to be gripped, and to keep the density of the region constant or within a predetermined range. After the pressure is set to a predetermined pressure, control to grip the gripping target is executed,
gripping control means;
A control device comprising: Disturbance elimination means for eliminating, by at least one of physical and algorithmic measures, external factors that are fluctuation factors of the grip weight of the object;
further comprising
A control device according to claim 1 . The disturbance elimination means is
a first disturbance elimination means for physically eliminating the external factor by executing control for maintaining the state of the object before the control is executed by the gripping control means;
has a
3. A control device according to claim 2 . The disturbance elimination means is
a second disturbance elimination means for eliminating the external factor while the control is being executed by the gripping control means;
has a
4. A control device according to claim 2 or 3 . The second disturbance elimination means is
During the operation of gripping by the arm, control is executed to cancel the state of gripping by the arm by the amount exceeding a target weight of the gripped object of the gripping weight gripped by the arm. to physically eliminate the external factors,
5. A control device according to claim 4 . The second disturbance elimination means is
Physically eliminating the external factor in the control by executing control to release the object attached to the arm from the arm after the gripping operation by the arm;
6. A control device according to claim 4 or 5 . The gripping control means is
machine parameter determining means for determining machine parameters used for the control;
gripping motion control means for controlling gripping motion by the arm based on the determined machine parameter;
has a
7. A control device as claimed in any one of claims 1 to 6 . The gripping control means is
weight confirmation means for confirming the actual weight of the gripped object gripped by the arm under the control of the gripping operation control means;
further having
The machine parameter determining means,
Acquiring the actual weight confirmed by the weight confirmation means, and determining machine parameters to be used for the control from the next time on based on the actual weight;
A control device according to claim 7 . region selection means for recognizing the object to be controlled before the control by the gripping control means and selecting a region including the gripping target based on the result of the recognition;
further comprising
9. A control device as claimed in any one of claims 1 to 8 . Ejecting means for executing control to eject the gripped object gripped by the arm to a predetermined location after the gripping control means performs the control;
further comprising
10. A control device as claimed in any one of claims 1 to 9 . The ejection means is
Acquiring coordinates at which the predetermined location exists, and executing the control to eject the grasped object at the coordinates;
11. Control device according to claim 10 .

Images