WO2022085408A1

WO2022085408A1 - Robot control device and robot control method

Info

Publication number: WO2022085408A1
Application number: PCT/JP2021/036652
Authority: WO
Inventors: 浩司白土; 宏規土橋
Original assignee: 三菱電機株式会社
Priority date: 2020-10-19
Filing date: 2021-10-04
Publication date: 2022-04-28
Also published as: DE112021005493T5; JPWO2022085408A1; JP7337285B2; CN116194255A

Abstract

Provided is a robot control device for controlling a robot (10) and a robot hand (20) to grasp an object (70), the robot control device comprising a grasp point generation unit (31) that generates a grasp point of the object (70) to be grasped by the robot hand (20). The grasp point generation unit (31) includes a deformation evaluation unit (33) that calculates shape deformation information when the shape of the object (70) is deformed by a grasping operation of the robot band (20), and a grasp point determination unit (36) that determines a grasp point of the object (70), on the basis of the shape deformation information.

Description

Robot control device and robot control method

The present disclosure relates to a robot control device and a robot control method that give an operation command to a robot and an end effector attached to the fingertip of the robot in order to grip the object so as not to drop it.

In the conventional robot control device, in order to determine the gripping point of the end effector (robot hand) with respect to the object, the shape and weight of the object are measured based on the measurement information of the object, and the position of the center of gravity of the object is estimated. However, there is a method in which a point passing near the position of the center of gravity is set as a gripping point (see, for example, Patent Document 1).

Japanese Unexamined Patent Publication No. 2008-49459 (pages 9 to 10, FIG. 2)

In the conventional robot control device, when picking an object with two or more irregular objects arranged, it is not always possible to grasp a position that passes through the position of the center of gravity in consideration of interference, and due to deformation due to gripping. An object may fall from the robot hand. As a result, there is a problem that the success rate of picking work is lowered and the production efficiency is lowered.

This disclosure is made to solve the above-mentioned problems, and grips an object with a high success rate in a picking operation using a flexible object or an amorphous object as an object. It is possible to obtain a robot control device that can shorten the tact time and maintain high production efficiency.

The robot control device according to the present disclosure controls the robot and the robot hand of the robot in order to grip the object, and includes a grip point generation unit that generates a grip point of the object to be gripped by the robot hand. The grip point generation unit includes a deformation evaluation unit that calculates shape deformation information when the shape of the object is deformed by the gripping motion of the robot, and a grip point determination unit that determines the grip point of the object based on the shape deformation information. It has.

According to the present disclosure, in picking work for a flexible object or an amorphous object, the object can be grasped with a high success rate, the takt time is shortened, and the production efficiency is maintained high. can.

It is an overall view of the robot system which concerns on Embodiment 1. FIG. It is a block diagram which shows the structure of the robot control apparatus which concerns on Embodiment 1. FIG. It is a block diagram which shows the structure of the gripping point generation part which concerns on Embodiment 1. FIG. It is a figure which shows the positional relationship between a finger and an object which concerns on Embodiment 1. FIG. It is a figure which shows the positional relationship between a finger and an object when the object which concerns on Embodiment 1 can move. It is a flowchart which shows the operation of the robot control apparatus which concerns on Embodiment 1. It is a physical characteristic model which combined the spring element and the damping element which concerns on Embodiment 2. It is a block diagram which shows the structure of the gripping point generation part which concerns on Embodiment 3. It is a flowchart which shows the operation of the robot control apparatus which concerns on Embodiment 3. It is a figure which shows the positional relationship between a finger and an object which concerns on Embodiment 4. FIG. It is a figure which shows the positional relationship between a finger and an object which concerns on Embodiment 5. It is a block diagram which shows the structure of the gripping point generation part which concerns on Embodiment 6. It is a block diagram which shows the structure of another gripping point generation part which concerns on Embodiment 6. It is a block diagram which shows the structure of the gripping point generation part which concerns on Embodiment 7. It is a schematic diagram of the object before grasping which concerns on Embodiment 9. FIG. It is a figure which shows the hardware composition of the robot control apparatus which concerns on Embodiment 1-9.

Embodiment 1.
FIG. 1 is an overall view of a robot system according to the first embodiment for carrying out the present disclosure. The robot system is based on the configuration of a robot and a robot control device that controls and operates the robot. The robot 10 may perform a task called material handling, such as gripping an object. In this case, a measuring device 60 that measures information such as the shape of the object 70 in order to acquire the position information and shape information of the object 70, and a robot hand 20 (end effector) for gripping the object 70 are provided. The configuration to be provided is added. The information of the object 70 measured based on the measuring device 60 is processed by the measuring device controller 50, and the information of the object 70 is input to the robot control device 30.

The robot control device 30 controls the robot 10 and the robot hand 20 of the robot 10 in order to grip the object 70, and the grip point generation unit 31 that generates the grip points of the object 70 gripped by the robot hand 20. It is equipped with. In the robot control device 30, in order for the robot hand 20 to grip the object 70 based on the input information of the object 70, the position command value to the robot 10 and the robot hand 20 at the position where the object 70 is gripped are obtained. Calculate the opening position of the finger (fingertip). The robot control device 30 controls the joint of the arm of the robot 10 and the finger of the robot hand 20 so that the finger of the robot hand 20 is controlled at a desired position. Hereinafter, the robot control device 30 controls either the joint of the arm of the robot 10 or at least one of the fingers of the robot hand 20 so that the finger of the robot hand 20 moves to an appropriate position. .. As the information of the object 70, the position shape and the shape information of the object 70 are exemplified.

Further, the robot control device 30 outputs the calculated position command value, the open position command value of the finger of the robot hand 20, and the closed position command value to the robot 10. The robot control device 30 determines the timing at which the position command value for the robot hand 20 is executed with respect to the position command value of the robot 10, and transmits the position command value to the robot 10 as the position command value at each time t. As a result, it is possible to realize an operation in which the fingers of the robot hand 20 are opened and approached, and the fingers of the robot hand 20 are closed at the gripping point of the object 70. Here, unless otherwise specified, the position command value of the robot 10 is related to 6 degrees of freedom of translation 3 degrees of freedom and rotation 3 degrees of freedom. Further, the position command value of the finger of the robot hand 20 depends on the type of the hand, but if it is a link structure, it is defined by the finger tip position or the opening width. In addition, it may refer to the position command value of each drive unit, but here, it refers broadly to the position command value that can be specified without being limited to the structure of the hand. Further, the finger of the robot hand 20 can control the gripping force when the actuator can control the pressure, the force, or the torque. Hereinafter, when the gripping force is specified, it is assumed that the gripping force command value is given to the gripping point candidate.

The "grasping point" means the position and posture of the finger in which the robot hand 20 can grip the object 70. In actual robot control, as described above, a position command value at each time t is required in addition to the position and orientation of the gripping point, but each of the robots 10 so that the robot hand 20 can reach the gripping point of the object 70. The joint position target value shall be calculated separately. The information that can be used to calculate the gripping point of the object 70 does not have to be limited to the position shape and shape information of the object 70. That is, as the information of the object 70, in addition to the direct information such as the position information and the shape information, the position of the object 70 is used by using indirect information such as the temperature information, the distance information, and the color information of the object 70. It is possible to estimate information and shape information.

FIG. 2 is a block diagram showing the configuration of the robot control device 30. As shown in FIG. 2, the robot control device 30 is mainly composed of a gripping point generation unit 31 and a command value generation unit 39. As shown in FIG. 2, the robot control device 30 calculates the position of the gripping point where the robot 10 should move, moves the robot hand 20 to the gripping point, and controls the robot 10 so that the robot 10 grips and operates. The gripping point generation unit 31 outputs the gripping point of the object 70 by using the shape information of the object 70 to be gripped by the robot 10.

Specifically, the target shape information is obtained by acquiring and calculating the image information or the distance information of the object 70 obtained by the visual sensor as the measuring device 60 as a point cloud. Alternatively, the object 70 may be actually gripped once by the finger of the robot hand 20 by actually using the robot hand 20, and the shape information may be acquired based on the position information of the finger at the time of gripping. Further, the measuring device 60 may use a distance measuring sensor to acquire shape information based on the cross-sectional shape of the object. Further, the temperature sensor may be used as the measuring device 60 to acquire shape information based on the approximate position and shape of the object 70. As described above, the measuring device 60 is not limited to the visual sensor. Further, position information, shape information, temperature information, distance information, color information and the like may be obtained from other than the measuring device 60.

FIG. 3 is a block diagram showing the configuration of the grip point generation unit 31. As shown in FIG. 3, the grip point generation unit 31 is composed of a grip point candidate generation unit 32, a deformation evaluation unit 33, and a grip point determination unit 36. The deformation evaluation unit 33 calculates shape deformation information when the shape of the object 70 is deformed by the gripping operation of the robot hand 20. Further, the gripping point determining unit 36 determines the gripping point of the object based on the deformation amount of the object included in the shape deformation information and the geometric constraint condition after the deformation of the object. decide. Hereinafter, each component will be described.

The gripping point candidate generation unit 32 generates gripping point candidates that can be gripped by the robot hand 20 attached to the robot 10 based on the target shape information input to the robot control device 30. At this time, as a method of generating the gripping point candidate, it is possible to search by a method of completely searching between any two points on the entire circumference from the target shape information based on the stroke (opening width) of the finger of the robot hand 20. For example, a case where a 2-finga gripper is taken as an example is shown in FIG. 4 described later.

The search for an object is an elliptical shape, that is, a process of selecting any two points on the outer circumference of the object. The deformation evaluation unit 33 moves a finger inside the object with respect to the two selected points, and performs deformation evaluation described later on the assumption that the grip is performed. At this time, since the finger has a movable direction and a movable range of the finger as a constraint condition, the search itself is not a full search, but the distance between the two candidate points and the open / close distance L0 with the opening / closing distance L0 of the finger as a constraint condition. It is possible to execute a search method under the constraint of comparison, and the search method itself is not limited.

Next, the deformation evaluation unit 33 evaluates and outputs the expected shape deformation information as shown in FIG. 4 for each case of the plurality of grip point candidates generated by the grip point candidate generation unit 32. do. The shape deformation information includes the shape information after deformation. In order to evaluate the shape deformation information, it is possible to calculate the shape deformation information expected in the model in which each finger causes deformation with respect to the object 70, assuming gripping by point contact by each finger.

Further, for the point contact portions provided in the robot hand 20 as many as the number of fingers, an appropriate fixed gripping force Fi is assumed for the generated points i (i = 1, 2, 3 ...). The mechanical relationship in which deformation occurs with respect to force can be calculated at each point contact point. At this time, in order to calculate the amount of deformation with respect to the force, the object 70 is treated as a uniform shape, the spring multiplier is K, the damping coefficient is C, and a rigid body, an elastic body or a rheology is used in the direction in which the finger force is generated. The shape deformation information can be evaluated by approximating the characteristics of the object as an object. For example, in the case of a rheological object, it is known that the relational expression between displacement and force is established as described in Non-Patent Document 1. As described above, the expected shape deformation information can be obtained by designating the gripping point and causing the shape deformation under appropriate conditions.

Finally, the gripping point determination unit 36 performs a process of determining the gripping point based on the shape deformation information generated by the deformation evaluation unit 33. The gripping point determination unit 36 extracts geometrically constrained grip points for each shape deformation information generated for the plurality of gripping point candidates. Specifically, FIG. 4 is a case of being restrained, and a case as shown in FIG. 5 is a case of not being restrained. Here, FIG. 4 is a diagram showing the positional relationship between the finger of the robot hand 20 and the object 70. 4 (a) shows the positional relationship before the finger of the robot hand 20 grips the object 70, and FIGS. 4 (b) and 4 (c) show the positional relationship after the finger of the robot hand 20 grips the object 70. Is shown. FIG. 5 is a diagram showing the positional relationship between the finger and the object 70 when the object 70 can move. In FIGS. 4 and 5, the direction in which the finger of the robot hand 20 opens and closes is defined as the X axis, the direction perpendicular to the direction along the finger, and the direction perpendicular to the X axis is defined as the Y axis.

The black arrows pointing outward from the object 70 shown in FIGS. 4 and 5 indicate the direction in which the object 70 is to be moved by applying a force from the outside. In this case, according to the gripping method of FIG. 4, based on the deformed shape information of the object 70, the object is geometrically applied even if an external force is applied in the X-axis direction and the Y-axis direction. Since it is restrained, it is easy to maintain the gripped state. This is because the geometrical constraint is established from the shape information of the object 70 after the deformation and the positional relationship of the fingers to be gripped after the deformation of the object 70. The present embodiment is characterized by paying attention to this point and extracting gripping point candidates that realize stable gripping. On the other hand, looking at the shape information of the object 70 after deformation and the positional relationship of the fingers in FIG. 5, the gripping stability due to the geometrical restraint also acts in the X direction, and the object cannot move. However, in the minus Y direction, the geometric constraint does not act and the gripping stability is low, and the object can be moved.

Hereinafter, an example of a method in which the gripping point determining unit 36 evaluates gripping stability is shown. As shown in FIG. 4C, the deformation evaluation unit 33 outputs a plurality of discrete points DP1, DP2, ... As shape deformation information of the object 70. The discrete points DP1, DP2, ... Are set based on the contour of the shape expected in the model that causes deformation with respect to the object 70. As an example, the gripping point determining unit 36 determines whether or not a geometric constraint is established from the relationship between the positions of the discrete points DP1, DP2, ... And the finger positions FP1, FP2, thereby gripping. Evaluate stability.

As another example, the gripping point determination unit 36 obtains a first approximate curve for a plurality of discrete points located in the vicinity of the finger position FP1. The gripping point determination unit 36 sets a plurality of discrete points (not shown) with respect to the finger based on the contour of the finger at the position FP1, and obtains a second approximate curve for the plurality of discrete points. The gripping point determination unit 36 determines the shape of the object 70 near the position FP1 of the finger (unevenness information, etc.) and the shape of the finger at the position FP1 (arc, rectangle) based on the first approximate curve and the second approximate curve. Etc.), and the gripping stability is evaluated by determining whether or not the geometrical constraint is established. Examples of the comparison method include the magnitude relation of the curvature between the shape of the object 70 and the shape of the finger, and the height difference between the maximum point and the minimum point of the first approximation curve. The gripping point determination unit 36 similarly obtains approximate curves for a plurality of discrete points located in the vicinity of the finger position FP2 and a plurality of discrete points (not shown) relating to the finger at the position FP2, and the same method as described above. Evaluate grip stability.

As yet another example, the gripping point determination unit 36 confirms the position coordinates of the discrete points DP1, DP2, ... When the virtual force Fvir is applied to the object 70, and the position coordinates before the addition are confirmed. By determining whether or not the amount of change is equal to or less than a predetermined value, it is determined whether or not the geometric constraint is established, and the gripping stability is evaluated. The gripping point determination unit 36 may determine that the gripping stability is low when the amount of change exceeds a predetermined value even in one of the plurality of discrete points, or the amount of change in a part of the plurality of discrete points. May be determined that the gripping stability is low when the value exceeds a predetermined value. It is assumed that the virtual force Fvir is applied to the object 70 from an arbitrary direction.

The above three methods have been mentioned as methods for the gripping point determining unit 36 to evaluate gripping stability, but the method is not limited thereto.

When a plurality of grip points are extracted, for example, the grip point closest to the position of the center of gravity can be selected. When the distance between the center of gravity and the gripping point is short, it is expected that the couple can be kept small even when the gripping force near the gripping point fluctuates due to disturbance or the like.

Here, the operation of the robot control device 30 will be described. FIG. 6 is a flowchart showing the operation of the robot control device. First, in step S101, the target shape information is input. Next, in step S102, the gripping point candidate generation unit 32 generates gripping point candidates that can be gripped by the robot hand 20 based on the input target shape information. Next, in step S103, the deformation evaluation unit 33 evaluates and outputs the shape deformation information for each case of the plurality of gripping point candidates. Then, in step S104, the gripping point determining unit 36 determines the gripping point based on the shape deformation information.

As described above, in the robot control device 30 provided with the grip point generation unit 31, in particular, the grip point generation unit 31 is a deformation evaluation unit that calculates shape deformation information when the shape of the object is deformed by the gripping operation of the hand. By having the gripping point determining unit 36 that determines the gripping point of the object based on the shape deformation information 33, an irregular object such as a flexible object grips the selected gripping point with respect to the object. By doing so, the gripping failure is greatly reduced, the object can be gripped with a high success rate, the tact time can be shortened, and the production efficiency can be maintained high, which is a special effect.

Note that production efficiency refers to the speed of work such as picking work. For example, as an example of production efficiency, it refers to takt time, and if one operation of 1 second is tried 100 times and succeeded 100 times, it is evaluated as an average of 1 second / time takt time, and the same work is tried 100 times. If it succeeds only 50 times, it is evaluated as an average tact time of 2 seconds / time. From the above, the fewer failures there are, the higher the production efficiency.

Embodiment 2.
In the present embodiment, the deformation evaluation unit 33 further adds a configuration for evaluating whether or not the upper limit of the force calculated based on the relational expression between the force applied to the object 70 and the displacement of the object 70 is exceeded. Is different from the first embodiment. In the robot control device 30 according to the first embodiment, a plurality of grip point candidates are extracted by the grip point candidate generation unit 32 after satisfying the condition of being geometrically constrained. From these gripping point candidates, the deformation evaluation unit 33 determines whether or not the object 70 exceeds the allowable deformation amount by the gripping force F (t) (value that changes at time t) expressed in time series. It is a feature of this embodiment that it is evaluated and added to the constraint condition.

The deformation evaluation unit 33 sets an upper limit value of the gripping force applied to the object 70 based on the relational expression between the gripping force applied to the object 70 and the displacement of the object 70 and the amount of deformation of the object 70 that the object 70 can tolerate. calculate. Then, the deformation evaluation unit 33 evaluates whether or not the gripping force applied to the object 70 from the robot hand 20 exceeds the upper limit value. Further, the deformation evaluation unit 33 determines the time-series information of the gripping force calculated based on the relational expression between the gripping force applied to the object 70 and the displacement of the object 70 as a part of the shape deformation information. Output to 36.

Generally, it can be classified into three categories by the method of deformation as a category of flexible objects. Deformations include elastic bodies whose shape returns to their original shape when the gripping force is removed after deformation, rheological bodies whose shape does not completely return, and plastic bodies that deform by the amount of force applied. On the other hand, flexible objects have an upper limit of allowable deformation. If the conditions of deformation are exceeded, an event occurs in which the object 70 is damaged or the commercial value is impaired.

The amount of deformation is calculated by the force and the time to apply the force. The relational expression between the force and the amount of deformation can be expressed mathematically, for example, by the description in Non-Patent Document 1. For example, in a rheological body or a plastic body, since it does not return to its original shape, a physical property model can be simulated by a configuration in which an elastic element and a damping element are connected in series as in the Maxwell model.

FIG. 7 is a physical characteristic model that combines a spring element and a damping element. As shown in FIG. 7, a physical characteristic model is constructed by combining a spring element and a damping element. In FIG. 7, the spring multiplier of the spring element is indicated by K1, and the damping coefficient of the damping element is indicated by C1 and C2. By connecting a plurality of them in series or in parallel, a physical characteristic model as shown in FIG. 7, for example, can be expressed. Here, the calculation of the force and the amount of deformation will be described by taking the "Maxwell model" shown in FIG. 7B as an example.

First, a physical property model is set in which a fixed point is P1 and a point on which a force is applied is P2, and a spring element having a spring multiplier K1 and a damping element having a damping coefficient C2 are arranged between P1 and P2. For each coefficient of the spring element and damping element, the value shall be input in advance according to the target flexible object. Further, by using the time-series position information obtained by inputting known forces (values such as 1N, 2N, 3N, etc.) as the physical characteristics of the object 70 for each coefficient, each of the spring element and the damping element is used. You can also estimate the coefficients. Further, when measurement or the like is difficult, it is possible to use a coefficient for a similar flexible object obtained in advance based on the physical properties of the object 70.

When using a 2-finger hand, two grip points are given to the finger. These are set as gripping points PG1 and PG2. At this time, the gripping point PG2 and the point P2 on which the force is applied are made to coincide with each other. Further, the vector (P1P2) and the vector (PG1PG2) are set to be parallel to each other. Regarding the displacement, the displacement of the joint between the spring element and the damping element is defined by x1, and the displacement of the gripping point P2 is defined by x2. The origins of the displacement x1 and the displacement x2 can both define the state of natural length. At that time, as an initial positional relationship, the length of the spring element of k1 is set to X10, and the length of the damping element is set to X20.

Under such conditions, when the time-series data F (t) of the gripping force is applied from the outside, it can be obtained as the time-series data of the displacement x1 and the displacement x2 by calculating the equation of motion. As described in Non-Patent Document 1, the characteristics (having residual displacement) of the rheological object can be simulated by giving the damping coefficient C2 a non-linear characteristic. However, the definition of the physical property model is not limited to this, and it can be applied to rigid bodies, elastic bodies, rheological objects, and plastic bodies by changing the coefficients and configurations.

As a result of the above calculation, the change in the position of the displacement x2 can be acquired, so it is possible to determine what kind of deformation occurs according to the time-series data F (t) of the appropriate gripping force. In particular, when a damping element is included, the position after unloading may not be the original position (x1 = 0 and x2 = 0).

Actually, there are various variations depending on the characteristics of the object, such as modeling in which the elastic element and the damping element are connected in series and then the damping element is added in parallel with the elastic element. Therefore, in this embodiment, the format of the physical property model is not particularly limited.

As described above, according to the present embodiment, the deformation evaluation unit 33 can extract only the gripping points in consideration of the permissible deformation of the gripping object which is the object 70, so that the gripping fails. The ratio of selecting points is reduced, and a special effect of improving production efficiency can be obtained. Further, when the robot hand 20 grips the object 70, which is a flexible amorphous object, the object 70 is damaged based on the time-series information of the shape of the deformed object 70 and the force up to it. It is possible to select a gripping point having high gripping stability without any problem. In particular, even when the robot hand 20 grips the object 70 with a slightly large gripping force, it can include a case where the deformation of the object 70 is allowed if the gripping force is removed within a predetermined time. Therefore, the failure is reduced, the object 70 can be grasped with a high success rate, the tact time can be shortened, and the production efficiency can be maintained high, which is a special effect.

Next, another modification of the present embodiment will be described. When considering the gripping of food as the object 70, there may be an acceptable amount of deformation because it impairs the commercial value from the viewpoint of appearance. In this case, the number of gripping point candidates may be very small if it is determined only by whether or not a certain upper limit value as mentioned above is exceeded. In this case, it can be utilized that the amount of deformation is within the permissible range as long as the upper limit of time is slightly exceeded.

That is, in the present embodiment, the deformation evaluation unit 33 has a gripping force F (t), which is the magnitude of the force acting on the gripping point, and a time t for applying a force equal to or greater than the allowable load, as a part of the shape deformation information. Is output. In this case, it is possible to evaluate whether or not the amount of deformation finally allowed is reached based on the gripping force F (t) expressed in time series. For example, the gripping point determining unit 36 uses the gripping point, the gripping force, and the gripping time as the shape deformation information to determine whether or not the deformation amount is within the allowable range based on the magnitude relationship between the force and the time threshold. .. As a result, the gripping point determining unit 36 can acquire the gripping point and the gripping force that realize the state in which the shape of the food is kept within a certain range. In the case of this configuration, the gripping point information includes information on the position of the gripping point and the gripping force (acting force) at the gripping point.

Further, the threshold value regarding the gripping force and the time can be obtained by being converted into the gripping force and the gripping time based on the allowable range of the deformation amount of the object 70. For the relationship between displacement and force, the description in Non-Patent Document 1 can be referred to. However, in order to determine whether or not the deformation amount of the object 70 is within the allowable range, the deformation amount may be calculated using the gripping point, the gripping force, and the gripping time, and the upper limit value may be set based on the deformation amount. not. The upper limit of the deformation amount of the object 70 is provided in advance by the user of this system for each food. Also in this case, since the robot control device 30 can extract only the gripping points in consideration of the permissible deformation of the gripping object which is the object 70, the ratio of selecting the gripping points where the gripping fails is reduced and is high. It is possible to grasp the object with the success rate, and it is possible to obtain a special effect that the tact time can be shortened and the production efficiency can be maintained high.

Embodiment 3.
In the present embodiment, the embodiment further includes a gripping stability calculation unit that evaluates gripping stability by mechanical stability against a predetermined external force with respect to the balance of the deformed force near the gripping point. Different from 2. FIG. 8 is a block diagram showing a configuration of a gripping point generation unit according to the third embodiment. In addition to the configuration of the grip point generation unit 31 shown in FIG. 3, the grip point generation unit 31a is composed of a grip stability calculation unit 34 and a result DB (result database) 35.

The gripping stability calculation unit 34 evaluates the mechanical stability against a predetermined external force with respect to the balance of the deformed force of the object 70 in the vicinity of the gripping point of the object 70. Further, the gripping stability calculation unit 34 evaluates the balance of the deformed force of the object 70 in the vicinity of the gripping point of the object 70, and the gripping force of the robot hand 20 with respect to the object 70 is minimized. Extract the gripping point.

The grip stability calculation unit 34 inputs shape deformation information. First, the grip stability calculation unit 34 calculates each point of the finger of the robot hand 20 and the grip target based on the force vector generated in the grip target after the deformation. Next, the grip stability calculation unit 34 evaluates whether or not the object 70 does not move based on the balance of forces at the grip point of the object 70. At this time, if the deformed object 70 and the finger of the robot hand 20 are geometrically constrained (immovable), the object 70 and the finger are affected by the action of a force other than the gripping force of the finger of the robot hand 20. It becomes a state where and is pressed, and it is regarded as a stable state.

The grip stability calculation unit 34 determines whether or not the "stable state" can be maintained. The stable state (stability) will be described. The predetermined external force is defined as Fdis, and the state in which the object 70 is only deformed according to the force of the finger of the robot hand 20 is defined as the external force Fdis = 0. The stability in the state of Fdis = 0 can be evaluated as to whether or not a couple or an accelerating force is generated in the object 70 based on the balance of mechanical forces. Even when a couple or acceleration force is generated on the object 70, the finger of the robot hand 20 and the deformed shape of the object are arranged so as to prevent the object 70 from moving in the direction in which the couple or acceleration force is generated. If it is configured, that is, if the geometrical constraint is established, the couple and the acceleration are canceled, and it is determined that the grip of the object at the designated gripping point is in the "stable state".

Further, the grip stability calculation unit 34 determines whether or not the "stable state" can be maintained even when the external force Fdis is set to a value other than 0. When the external force Fdis is applied, the deformation when the gripping force F (t) and the external force Fdis are added to the shape deformation information is added. Deformation is obtained by the relationship between displacement and force using the above-mentioned physical property model. The "stable state" is determined based on this shape deformation information.

Further, the grip stability calculation unit 34 can also determine the "stable state" under the condition that the robot 10 accelerates or decelerates. First, when accelerating or decelerating while gripping, an inertial force is generated on the object. In the case of an inertial force, the external force Fdis (t) due to the inertial force can also be expressed as in Equation 1 by the mass m of the object 70 and the acceleration α_obj (t) of the object 70. The acceleration α_obj (t) of the object 70 is a function of time t, but is basically obtained based on the command value regarding the finger of the robot 10.
Fdis (t) ＝ m ・ α_obj (t) (Equation 1)

The phenomenon that the object 70 slides down from the finger of the robot hand 20 with respect to the inertial force Finr, that is, the upper limit Flim of the binding force for eliminating the geometrical constraint is set according to the physical properties of the object 70 (elastic modulus K and damping coefficient C). If the upper limit of the binding force is exceeded, it will not be in a stable state even if the geometrical constraint is lost. When the gripping stability calculation unit 34 maintains the "stable state", the gripping stability evaluation value is set high, and the stability evaluation result is output to the DB 35 as a result. Further, when the grip stability calculation unit 34 is no longer in the "stable state", the grip stability evaluation value is set low, and the stability evaluation result is output to the DB 35 as a result.

The upper limit of the binding force from which the stable state disappears can also be defined by the friction coefficient μ between the object 70 and the robot hand 20. Assuming that the pressing force at the gripping point of the robot hand 20 is Fi, the binding force upper limit Flim can also be defined as in Equation 2.
Flim = μ ・ Fi (Equation 2)
In this case, for example, assuming that the "stable state" of the finger of the robot hand 20 with respect to the gripping point i is the gripping stability Si, the gripping stability Si can be defined as in the equation 3.
Si = (Flim-max (Fdis (t))) (Equation 3)
By using the grip stability Si, it becomes possible to compare the grip points with each other. The grip stability calculation unit 34 outputs this as a stability evaluation result to the result DB 35. The stability evaluation results are not limited to this method.

Based on the above rules, the grip stability calculation unit 34 outputs the grip point candidates and the stability evaluation results calculated via the result DB 35 to the grip point determination unit 36. The gripping point determination unit 36 can select the gripping point having the highest stability evaluation result based on the plurality of stored gripping point candidates and the stability evaluation result.

Here, the operation of the robot control device 30 will be described. FIG. 9 is a flowchart showing the operation of the robot control device. Since steps S101 to S103 of FIG. 9 are the same as those of FIG. 6, the description thereof will be omitted. In step S201, the grip stability calculation unit 34 determines whether or not the "stable state" can be maintained. If the "stable state" can be maintained, the process proceeds to step S202, and the grip stability calculation unit 34 sets a high evaluation value of the grip stability. When it is no longer in the "stable state", the process proceeds to step S203, and the grip stability calculation unit 34 sets the evaluation value of the grip stability low. Then, in step S204, the gripping point determination unit 36 selects the gripping point having the highest stability evaluation result based on the plurality of stored gripping point candidates and the stability evaluation result, and determines the gripping point.

With such a configuration, it becomes possible to extract a gripping point that is difficult to fall in consideration of a fall due to a shape deformation while the robot hand 20 is transporting the object 70. In this case as well, since it is possible to extract only the gripping points in consideration of the allowable deformation of the gripping object which is the object 70, the ratio of selecting the gripping points that fail to grip is reduced, and the production efficiency is improved. Effect can be obtained.

Next, another modification of the present embodiment will be described. It is considered that the deformation evaluation unit 33 performs a simulation in which the gripping force Fi (t) at each gripping point is variously changed. When a gripping force Fi (t) is included, according to Equation 2, the binding force upper limit Flim becomes smaller, and as a result, the gripping stability Si tends to become smaller. On the other hand, when the gripping force Fi (t) at each gripping point becomes smaller, the amount of deformation output as the shape deformation information calculated by the deformation evaluation unit 33 becomes smaller. At this time, the gripping stability Si can be defined as in the formula 4 including the index of "minimum deformation" as an index different from the "stable state" as in the formula 3.
Si = w1 * (Flim-max (Fdis (t))) + w2 / max (Fi (t)) (Equation 4)
Here, w1 and w2 are appropriate weighting factors. The weighting factor is designed by the user depending on whether it is easier to maintain a stable state or whether it is gripped with the minimum gripping force.

When the gripping stability calculation unit 34 evaluates the gripping stability based on the gripping stability Si, when the robot hand 20 grips a flexible amorphous object, it is based on the shape of the deformed object 70. It becomes possible to select a gripping point having high gripping stability without damaging the object 70. As a result, gripping failures are reduced, the object can be gripped with a high success rate, the tact time is shortened, and the production efficiency is improved.

Embodiment 4.
In the present embodiment, the gripping stability calculation unit 34 evaluates the resistance to geometrical deviation based on the shape of the finger of the robot hand 20 and the shape deformation information, and outputs the result of the gripping stability evaluation. In the third embodiment, the gripping points are represented by points, but in the present embodiment, the gripping points are given a geometric shape. In this case, a plurality of contact points are generated even for one finger. The grip stability calculation unit 34 evaluates the difficulty of geometrically shifting the object 70 with respect to the robot hand 20 based on the shape of the finger of the robot hand 20 and the shape deformation information.

FIG. 10 is a diagram showing the positional relationship between the finger of the robot hand 20 and the object 70. By introducing a physical characteristic model in which there are a plurality of contact points associated with gripping the object 70, it is possible to replace Equation 2 defining the binding force upper limit Flim regarding geometrical restraint as Equation 5.
Flim = μ ・ A ・ Fi (Equation 5)
Here, A is an effective contact area between the finger of the robot hand 20 and the object 70. As shown in FIG. 10, the effective contact area indicates the contact area when the finger makes surface contact with an object instead of point contact. In general, the coefficient of friction in the state of surface contact is larger than the coefficient of friction in the state of point contact. Modeling to reflect this is the embodiment of the present embodiment. It shows the amount of contact area of the finger that increases with the deformation of the object 70. It is a physical characteristic model in which A = 1 when deformed by a certain amount or more, and 0 <A <1 when the gripping force Fi (t) is small in a lightly pinched state.

That is, it is equivalent to changing the effective coefficient of friction based on the deformed shape of the object 70 and the contact amount with the finger of the robot hand 20. As described above, the grip stability calculation unit 34 defines the friction coefficient based on the effective contact area, and the grip stability is calculated based on the friction coefficient, which is a feature of the present embodiment.

In the configuration in which the above physical property model is introduced, the weaker the gripping force Fi (t), the smaller the effective contact area A, so that the upper limit of the binding force Flim tends to change according to the gripping force Fi (t). As a result, the deformation evaluation of the actual gripping state and the simulated accuracy of the gripping stability calculation unit are improved.

According to the present embodiment, when the robot hand 20 grips a flexible amorphous object, the accuracy of the grip stability calculated based on the deformed shape is improved, and the object is damaged more than before. It becomes possible to select a gripping point having high gripping stability. As a result, failures are reduced, the object can be grasped with a high success rate, the tact time is shortened, and the production efficiency is improved.

Embodiment 5.
This embodiment is different from the third embodiment in that it outputs shape deformation information when the gripping force is removed after a certain period of time after the gripping force is applied. The deformation evaluation unit 33 outputs the shape deformation information after the robot hand 20 applies the gripping force to the object 70 and then unloads the gripping force after a certain period of time. Then, the grip stability calculation unit 34 obtains the difference amount between the original shape of the object 70 and the shape of the object 70 after unloading, and evaluates by comparing the difference amount with the predetermined deformation allowable value. do.

When the robot hand 20 continues to apply the gripping force to the object 70, the gripping force Fi (t) = F0 (constant) can be set. In this case, the shape deformation information is expected to converge to a certain shape. On the other hand, unless it is completely plastically deformed, the shape changes further when the gripping force is removed.

Here, consider a case where, for example, a force called a gripping force F0 is applied from 0 seconds to t0 seconds, and the load is unloaded after the elapse of t0 seconds. After unloading, Fi (t + t0) = 0. At this time, in the present embodiment, the shape after a sufficient time has passed after unloading is regarded as "shape deformation information after unloading", and "shape after unloading" is used as a part of the shape deformation information from the deformation evaluation unit. It adds a configuration to output "transformation information".

The shape deformation information in the state where the gripping force Fi (t) = F0 output from the deformation evaluation unit 33 is applied is used as the first shape deformation information, and the gripping force is unloaded after a certain period of time t0 has elapsed for a sufficient time. The shape deformation information after the lapse of time is called the second shape deformation information. At this time, the difference amount between the original shape of the object 70 and the shape in the second shape deformation information is calculated, and the deformation tolerance value and the magnitude relationship determined in advance are compared for the difference amount, and the deformation tolerance value is exceeded. It is characterized in that the gripping stability is evaluated low for those that do, and the gripping stability is evaluated high for those that do not exceed.

FIG. 11 is a diagram showing the positional relationship between the finger and the object 70. Hereinafter, an example of a method of taking a difference in shape before and after the gripping force is applied will be illustrated with reference to FIG. 11. The gripping stability calculation unit 34 obtains the difference amount between the curvature of the original shape of the object 70 and the curvature of the shape of the object 70 after unloading, and determines the difference amount of the curvature and the predetermined deformation allowable value. Compare and evaluate. The amount of difference in curvature can be obtained as follows. The first shape deformation information and the second shape deformation information are superimposed on the non-deformed point (point far from the gripping point) as a reference. After superimposing, select two points that change from the deformed point to the non-deformed point. That is, the line segments of the curves whose positions have changed before and after the deformation are overlapped because they have not been deformed for the first time (in the case of FIG. 11, they correspond to the discrete points DP3 and DP4). At this time, the length of the curve between these two points is obtained. The curves between the two points are a curve having a length L1 from the discrete point DP3 through the discrete point DP5 to the discrete point DP4 and a curve having a length L2 from the discrete point DP3 through the discrete point DP1 to the discrete point DP4. There are two. Corresponding points are defined for each fixed ratio based on the respective lengths L1 and L2. For example, in each curve, points corresponding to 0.25 × L1 and 0.25 × L2 are defined as corresponding points, and the distance between the corresponding points is obtained. Find each of the distances and define the maximum value as the "difference in curvature". In the case of FIG. 11, the difference amount of the curvature is the distance DC1 between the discrete point DP1 and the discrete point DP5. The deformation evaluation unit 33 evaluates whether the difference amount is larger or smaller than the "deformation allowable value" predetermined by the user, and outputs the evaluation result as a part of the shape deformation information. In the case of FIG. 11, three discrete points are provided between the curves between the two points, but the number is not limited to three, and may be, for example, nine points. When there are 9 discrete points, the points corresponding to 0.1 × L1 and 0.1 × L2 from the end points are defined as corresponding points in each curve.

In the present embodiment, the grip stability calculation unit 34 uses the shape deformation information and outputs it with a label to be rejected as a grip point candidate when it is larger than the "deformation allowable value".

According to the present embodiment, when the robot hand 20 grips a flexible amorphous object, the result is evaluated by evaluating whether or not the shape is acceptable based on the shape after unloading, that is, the final shape after work. It is possible to exclude from the extraction target a gripping force or a gripping point that is treated as a work failure. As a result, failures are reduced, the object can be grasped with a high success rate, the tact time is shortened, and the production efficiency is improved.

Embodiment 6.
In this embodiment, a success / failure label (success / failure information), which is a result label obtained by performing work on a simulated or actual gripping point, and shape deformation information, gripping force, gripping point, and physical properties of an object at that time. The grip point generation unit is different from the first embodiment in that the grip point candidate learning unit for constructing a neural network capable of performing learning by inputting the above and outputting the grip point by inputting the shape of the object is provided.

FIG. 12 is a block diagram showing the configuration of the gripping point generation unit 31b according to the sixth embodiment. In addition to the configuration of the grip point generation unit 31 shown in FIG. 3, the grip point generation unit 31b is composed of a grip stability calculation unit 34 and a result DB (result database) 35. Further, the robot control device 30 includes a gripping point candidate learning unit 37 and a learning DB (learning database) 38. The grip point candidate learning unit 37 has a neural network 40. The gripping point candidate learning unit 37 inputs the gripping point candidate output from the gripping stability calculation unit 34, the result data which is the stability evaluation result, and the result label obtained in the actual work, and the object is obtained from the shape deformation information. Learn the relationship that outputs grip point candidates. As shown in FIG. 12, it is exemplified to learn a network that outputs a gripping point, a gripping force, and a gripping stability by inputting target shape information (before deformation).

The gripping point candidate learning unit 37 will be described with an example using simulation and an actual machine experiment. Consider using simulation (numerical calculation processing) to determine gripping point candidates based on the shape information of the object 70, which is the object to be gripped, obtained immediately before gripping the object 70, and to actually try the gripping operation. .. Based on the configurations described up to the fifth embodiment, it is expected that gripping will be successful with a high probability when working with an actual machine. However, when working on an actual machine, it is assumed that gripping will fail due to factors that could not be modeled.

In this case, there are success and failure result labels for all the grip point candidates designed by the deformation evaluation unit 33 and the grip point determination unit 36. However, in general, it is difficult to formulate the causal relationship between success, the shape of the object (before and after deformation), the gripping point, and the gripping force. Therefore, for example, a framework of a neural network can be used to learn a non-linear relationship and acquire the relationship.

Success / failure label for multiple trials, gripping point, gripping force, physical properties of the object, deformation shape of the object (shape before deformation and shape information after deformation) and grip stability, success / failure label for each trial Is prepared, and the process of learning the neural network is performed.

Here, the grip point generation unit 31b is a physical characteristic model definition unit (not shown) that models the relationship between the force acting on the object 70 and the displacement of the object 70 by a model using a spring multiplier and a damping coefficient. have. The physical property model definition unit applies a force that changes with time to the object 70, and sets the physical property model (spring multiplier) of the model based on the time-series information of the displacement based on the deformation of the object 70 with respect to the applied force. K and the damping coefficient C) are estimated. At this time, the spring multiplier K and the damping coefficient C can be updated with respect to the predetermined spring multiplier K and the damping coefficient C based on the deformation result obtained by the actual machine work. In addition, as another method, the relationship between the force and the displacement can be obtained by learning only from the time-series information of the deformation information and the gripping force actually obtained without assuming the spring multiplier K and the damping coefficient C. can. For example, it is possible to exemplify the construction of a neural network that gives time-series information of deformation information and gripping force and outputs displacement information from the time-series information of gripping force.

Further, the gripping point candidate learning unit 37 has a physical characteristic model learning unit (not shown) that learns by modeling the relationship between the force acting on the object 70 and the displacement of the object 70 by a neural network. The physical characteristic model learning unit applies a force that changes with time to the object 70, and learns a neural network 40 that is set based on time-series information of displacement based on the deformation of the object with respect to the applied force.

The gripping point candidate learning unit 37 performs learning processing based on the gripping point candidate stored in the result DB 35 and the stability evaluation result. For example, learning of the neural network 40 is exemplified. The neural network 40 has a learning unit and an inference unit (not shown). Using the learning parameters in the learning unit, the neural network 41 that reflects the learning parameters is incorporated in the inference unit. Then, the target shape information can be input and the gripping point and the gripping force can be output. The learning parameters are exemplified by the coefficients that define the network structure of the neural network.

FIG. 13 is a block diagram showing a configuration of another grip point generation unit 31c according to the sixth embodiment. As shown in FIG. 13, when the neural network 41 acquired in the above process is applied as the grip point candidate generation unit 32a and the target shape information is input, the grip point candidate generation unit 32a generates a plurality of grip point candidates and grip stability. It is generated and output to the gripping point determination unit 36. The gripping point determination unit 36 selects and outputs one gripping point candidate using the gripping stability.

According to the present embodiment, a gripping point generation algorithm that corrects the modeling error acquired in the actual work can be acquired by learning, and as a result, the calculation cost for calculating the gripping point candidate is reduced, and the gripping point is calculated. Since the time to do is shortened, a special effect of increasing production efficiency can be obtained.

Next, another modification of the present embodiment will be described. A method of performing the same processing as described above from the gripping point candidates obtained by the simulation will be described. In the above description, it is considered to utilize simulation (numerical calculation processing) to determine a gripping point candidate based on the shape information of the gripping object obtained immediately before gripping, and to actually try the gripping operation. At this time, the shape information of the gripping object is also the shape generated by the simulation. Further, the trial itself for grasping the object 70 simulates a physical contact phenomenon and the like, and is carried out on a physical simulation in which the shape is precisely observed. Therefore, when using a physics simulation without disturbance or uncertainties, it is expected that the grip points with the highest success rate are known and all are labeled as successful.

For trials with multiple success labels, the gripping point, gripping force, physical properties of the object 70, the deformed shape of the object 70 (shape before deformation and shape information after deformation), and grip stability, respectively. Prepare the success / failure label for the trial of, and perform the process of training the neural network.

Although it is based on a physics simulation model, it is exemplified to learn a network that outputs a gripping point, a gripping force, and a gripping stability by inputting shape information (before deformation) as in the case of using an actual machine.

When the neural network 41 acquired in the above process is applied as the grip point candidate generation unit 32a and the target shape information is input, a plurality of grip point candidates and grip stability are generated and output to the grip point determination unit 36. The gripping point determination unit 36 selects and outputs one gripping point candidate using the gripping stability.

According to the present embodiment, when the robot hand 20 grips an object 70 which is a flexible amorphous object, it is possible to extract one candidate point having the highest gripping stability simply by inputting shape information. It can be done and the failure of gripping is reduced. Further, the physical property information regarding the deformation of the object 70 can be acquired from the actual object, and the deformation simulation accuracy of the deformation evaluation unit based on the simulation is improved. In this way, for grip points and stable grip points determined based on a complicated physics simulation model, a grip point generation algorithm that automatically outputs grip points when a target shape is input can be acquired by learning, and as a result, Since the calculation cost for calculating the grip point candidate is reduced and the time for calculating the grip point is shortened, the tact time is shortened and the production efficiency is improved, which is a special effect.

Embodiment 7.
In the present embodiment, after the gripping point candidate generation unit defines the first gripping force, evaluates it under the condition of gripping with the first gripping force with respect to the gripping point, and extracts an effective gripping point, It is different from the third embodiment in that the gripping point can be efficiently searched for by gripping the object with a second gripping force smaller than the first gripping force.

FIG. 14 is a block diagram showing the configuration of the gripping point generation unit 31d according to the seventh embodiment. For the configuration of the gripping point generation unit 31a shown in FIG. 8, the gripping point candidate is input to the gripping point candidate generation unit 32 from the result DB 35. As shown in FIG. 14, the gripping point candidate obtained by the gripping stability calculation unit 34 and the result database based on the stability evaluation result are input to the gripping point candidate generation unit 32 again. Then, the gripping point candidate generation unit 32 extracts a finite number of those having a high stability evaluation, and uses the finite number of those extracted gripping point candidates as a second gripping force (however, smaller than the first gripping force). Is specified.

The gripping point generation unit 31d defines a result DB 35 for storing a plurality of gripping point candidates and a first gripping force output by the robot hand 20 to the object 70, and is designated to be gripped by the first gripping force. It has a gripping point candidate generation unit 32 that outputs the first gripping point candidate to the deformation evaluation unit. The gripping stability calculation unit 34 calculates the stability evaluation result for the first gripping point candidate, and outputs the first gripping point candidate and the stability evaluation result to the result DB 35. The gripping point candidate generation unit 32 extracts a plurality of gripping point candidates from the first gripping point candidate stored in the result DB 35 based on the stability evaluation result, and the second gripping force with respect to the plurality of gripping point candidates. Is defined and output to the deformation evaluation unit 33 again.

The gripping point candidate generation unit 32 can repeat the same process three or more times. For example, by repeating the third gripping force, the fourth gripping force, ..., The kth gripping force, and reducing the gripping force to be searched, the object 70 effective with the minimum gripping force. It is possible to extract the gripping point where the deformation is obtained. As a result, it is possible to efficiently search for points that can be stably gripped with the smallest gripping force, and as a result of being able to extract candidate points that are unlikely to fail in gripping in a short time, the work time per robot operation is shortened and tact. The special effect of shortening the time and increasing the production efficiency can be obtained.

Embodiment 8.
In the present embodiment, the gripping stability calculation unit 34 is different from the first embodiment in that the gripping stability calculation unit 34 obtains a combination of gripping point candidates for stably gripping the object from the information on the contour of the object. The grip stability calculation unit 34 acquires information on the contour of the object 70 from the point cloud coordinates of the contour of the object of the object 70, and selects a combination of grip point candidates from above the contour of the object 70. Then, the gripping stability calculation unit 34 obtains an evaluation value when the robot hand 20 grips the object 70 with a predetermined gripping force for each combination, and a gripping point for stably gripping the object 70 based on the evaluation value. Find a combination of candidates.

An evaluation method for searching for a combination of gripping point candidates for the object 70 for stably gripping the object 70 will be described. The grip stability calculation unit 34 derives a combination of stable grip points based on the evaluation of the magnitude of the minimum grip force required to grip the object 70. The minimum gripping force required to grip the object 70 is the minimum fingertip force required to resist the gravity acting on the object 70. From the viewpoint of the difficulty of breaking the object 70, this value should be small. The evaluation is performed using the value of the fingertip force obtained from the gripping force and the frictional force, and the search for the combination of stable gripping points is performed by the following procedure.

First, from the point cloud coordinates of the outline of the object 70, the points arranged in the two-dimensional plane are smoothly connected by the Spline interpolation method, and the information of the outline of the object 70 is acquired. Grip point candidates are taken on the contour of the object 70, and all combinations of the two points are stored.

Next, the gripping force is set to a certain value, evaluation values are obtained for all combinations of gripping point candidates, and from the results, a combination of stable gripping points at that gripping force is obtained. Then, the gripping force is changed, evaluation values are obtained for all combinations of gripping point candidates, and from the results, a combination of stable gripping points at the gripping force is obtained. This operation is repeated to obtain a combination of stable grip points with the optimum grip force.

As described above, the gripping stability calculation unit 34 obtains a combination of gripping point candidates of the object 70 for stably gripping the object 70, whereby an amorphous object such as a flexible object is obtained with respect to the object. , Gripping failure due to gripping the selected gripping point is greatly reduced, the object can be gripped with a high success rate, the tact time can be shortened, and the production efficiency can be improved. Can be done.

Embodiment 9.
In the present embodiment, the gripping stability calculation unit 34 obtains an evaluation value based not only on the shape deformation information of the object 70 after gripping but also on the shape information of the object 70 before gripping. FIG. 15 is a schematic view of the object 70 before gripping according to the ninth embodiment. Specifically, the deformation evaluation unit 33 outputs a plurality of discrete points DPB1, DPB2, ... As shape information of the object 70 before gripping. The plurality of discrete points DPB1, DPB2, ... Are set based on the contour of the object 70. The grip stability calculation unit 34 quantitatively evaluates the degree of depression of the object from the positional relationship of the plurality of discrete points DPB1, DPB2, ..., And outputs it as an evaluation value. In the case of FIG. 15, since the dents are generated at the discrete points DPB1 and DPB2, the gripping stability calculation unit 34 may output the discrete points DPB1 and DPB2 as gripping point candidates having high gripping stability.

As described above, the gripping stability calculation unit 34 can accurately select the gripping point by obtaining the evaluation value based on the shape information of the object 70 before gripping. As a result, it is possible to grasp the object with a high success rate, and it is possible to obtain a special effect that the tact time can be shortened and the production efficiency can be increased.

Here, the hardware configuration of the robot control device 30 will be described. Each function of the robot control device 30 can be realized by a processing circuit. The processing circuit comprises at least one processor and at least one memory.

FIG. 16 is a diagram showing a hardware configuration of the robot control device according to the first to ninth embodiments. The robot control device 30 can be realized by the control circuit shown in FIG. 16A, that is, the processor 81 and the memory 82. An example of the processor 81 is a CPU (Central Processing Unit, central processing unit, processing unit, arithmetic unit, microprocessor, microprocessor, processor, DSP (Digital Signal Processor)) or system LSI (Large Scale Integration).

The memory 82 is, for example, a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable Read Only Memory), an EEPROM (registered trademark), etc. Alternatively, it may be a volatile semiconductor memory, a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, a DVD (Digital Versaille Disc), or the like.

The robot control device 30 is realized by the processor 81 reading and executing a program stored in the memory 82 for executing the operation of the robot control device 30. It can also be said that this program causes a computer to execute the procedure or method of the robot control device 30. The program executed by the robot control device 30 includes a grip point generation unit 31 and a command value generation unit 39, which are loaded on the main storage device and these are generated on the main storage device.

The memory 82 stores obstacle information, target shape information, shape deformation information, and the like. The memory 82 is also used as a temporary memory when the processor 81 executes various processes.

The program executed by the processor 81 may be a file in an installable format or an executable format, stored in a computer-readable storage medium, and provided as a computer program product. Further, the program executed by the processor 81 may be provided to the robot control device 30 via a network such as the Internet.

Further, the robot control device 30 may be realized by dedicated hardware. Further, the functions of the robot control device 30 may be partially realized by dedicated hardware and partially realized by software or firmware.

Further, the robot control device 30 may be realized by the dedicated processing circuit 83 shown in FIG. 16 (b). At least a part of the grip point generation unit 31 and the command value generation unit 39 may be realized by the processing circuit 83. The processing circuit 83 is dedicated hardware. The processing circuit 83 is, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination thereof. Is. A part of the function of the robot control device 30 may be realized by software or firmware, and the other part may be realized by dedicated hardware.

10 robot, 20 robot hand, 30 robot control device, 31, 31a, 31b, 31c, 31d grip point generation unit, 32, 32a grip point candidate generation unit, 33 deformation evaluation unit, 34 grip stability calculation unit, 35 result DB , 36 grip point determination unit, 37 grip point candidate learning unit, 38 learning DB, 39 command value generation unit, 50 measuring device controller, 60 measuring device, 70 object, 81 processor, 82 memory, 83 processing circuit, FP1 to FP6 Finger position, DP1 to DP5, DPB1, DPB2 discrete points.

Claims

A robot control device that controls a robot and the robot hand of the robot in order to grip an object.
A grip point generation unit for generating a grip point of the object to be gripped by the robot hand is provided.
The gripping point generation unit includes a deformation evaluation unit that calculates shape deformation information when the shape of the object is deformed by the gripping operation of the robot hand.
A robot control device comprising a gripping point determining unit that determines a gripping point of the object based on the shape deformation information.
The gripping point determining unit is characterized in that the gripping point of the object is determined based on the deformation amount of the object included in the shape deformation information and the geometrical constraint condition after the deformation of the object. The robot control device according to claim 1.
The deformation evaluation unit outputs a plurality of discrete points as the shape deformation information, and outputs the plurality of discrete points.
The robot control device according to claim 2, wherein the gripping point determining unit determines the geometrical constraint condition based on the positional relationship between the plurality of discrete points and the finger of the robot hand.
The deformation evaluation unit outputs a plurality of discrete points as the shape deformation information, and outputs the plurality of discrete points.
The claim is characterized in that the gripping point determining unit determines the geometric constraint condition based on the amount of change in the positions of the plurality of discrete points when a virtual force is applied to the object. 2. The robot control device according to 2.
The deformation evaluation unit sets an upper limit value of the gripping force applied to the object based on the relational expression between the gripping force applied to the object and the displacement of the object and the amount of deformation of the object that the object can tolerate. The robot control device according to any one of claims 1 to 4, wherein the robot control device is calculated and evaluated whether or not the gripping force applied to the object from the robot hand exceeds the upper limit value.
The deformation evaluation unit is characterized in that it outputs time-series information of the gripping force calculated based on the relational expression between the gripping force applied to the object and the displacement of the object as a part of the shape deformation information. The robot control device according to any one of claims 1 to 5.
The gripping point generation unit has a gripping stability calculation unit that evaluates mechanical stability against a predetermined external force with respect to the balance of the deformed force of the object in the vicinity of the gripping point of the object. The robot control device according to any one of claims 1 to 6, wherein the robot control device is characterized.
The gripping stability calculation unit evaluates the balance of the deformed force of the object in the vicinity of the gripping point of the object, and grips the object so that the gripping force of the robot hand with respect to the object is minimized. The robot control device according to claim 7, wherein points are extracted.
The gripping point generation unit is characterized by having a gripping stability calculation unit that evaluates the resistance to geometrical deviation of the object with respect to the robot hand based on the shape of the fingertip of the robot hand and the shape deformation information. The robot control device according to any one of claims 1 to 6.
The deformation evaluation unit outputs the shape deformation information after the robot hand applies the gripping force to the object and then unloads the gripping force after a certain period of time.
The gripping stability calculation unit obtains a difference amount between the original shape of the object and the shape of the object after unloading, and evaluates the difference amount by comparing it with a predetermined deformation tolerance value. The robot control device according to any one of claims 7 to 9, wherein the robot control device is characterized.
The deformation evaluation unit outputs the shape deformation information after the robot hand applies the gripping force to the object and then unloads the gripping force after a certain period of time.
The gripping stability calculation unit obtains a difference amount between the curvature of the original shape of the object and the curvature of the shape after unloading the object, and sets the difference amount of the curvature and a predetermined deformation tolerance value. The robot control device according to any one of claims 7 to 9, wherein the robot control apparatus is compared and evaluated.
The gripping point generation unit acquires information on the contour of the object from the point cloud coordinates of the contour of the object of the object, and selects a combination of grip point candidates of the object from above the contour of the object. The evaluation value when the robot hand grips the object with a predetermined gripping force is obtained for each combination of gripping point candidates of the object, and the object that stably grips the object based on the evaluation value. The robot control device according to any one of claims 1 to 6, further comprising a gripping stability calculation unit for obtaining a combination of gripping point candidates for an object.
The grip point generation unit includes a result database for storing a plurality of grip point candidates and a result database.
A gripping point that defines a first gripping force output to the object by the robot hand and outputs a first gripping point candidate designated to be gripped by the first gripping force to the deformation evaluation unit. It has a candidate generator and
The gripping stability calculation unit calculates the stability evaluation result for the first gripping point candidate, outputs the first gripping point candidate and the stability evaluation result to the result database, and outputs the result.
The gripping point candidate generation unit extracts a plurality of gripping point candidates from the first gripping point candidate stored in the result database based on the stability evaluation result, and the first gripping point candidate is obtained with respect to the plurality of gripping point candidates. The robot control device according to any one of claims 7 to 12, wherein the second gripping force is defined and output to the deformation evaluation unit again.
The gripping point generation unit inputs the result data output from the gripping stability calculation unit and the result label obtained in the actual work, and learns the relationship of outputting the gripping point candidate of the object from the shape deformation information. The robot control device according to any one of claims 7 to 12, further comprising a gripping point candidate learning unit.
The grip point generation unit has a physical characteristic model definition unit that models the relationship between the force acting on the object and the displacement of the object by a model using a spring multiplier and a damping coefficient.
The physical property model definition unit applies a force that changes with time to the object, and sets the spring multiplier and damping coefficient of the model based on the time-series information of the displacement based on the deformation of the object with respect to the force. The robot control device according to any one of claims 1 to 14, wherein the robot control device is characterized in that.
The gripping point generation unit has a physical characteristic model learning unit that models the relationship between the force acting on the object and the displacement of the object by a neural network.
The physical property model learning unit is characterized by applying a force that changes with time to the object and learning a neural network set based on time-series information of displacement based on the deformation of the object with respect to the force. The robot control device according to any one of claims 1 to 14.
The deformation evaluation unit further outputs a plurality of discrete points as shape information of the object before gripping.
The gripping stability calculation unit is characterized in that it evaluates a dent of the object from the positional relationship of the plurality of discrete points to obtain a dent evaluation value, and outputs the grip point candidate based on the dent evaluation value. 12. The robot control device according to claim 12 or 13.
A robot control method for controlling a robot and the robot hand of the robot in order to grip an object.
A step of calculating shape deformation information when the shape of the object is deformed by the gripping operation of the robot hand, and
A robot control method comprising a step of determining a gripping point of the object based on the shape deformation information and generating a gripping point of the object gripped by the robot hand.