JP6988573B2 - Simulation equipment, simulation method and simulation program - Google Patents

Description

The present invention relates to a simulation apparatus, a simulation method and a simulation program.

There is a pick-and-place device as one of the devices for transporting workpieces such as parts and products in a factory production line. The pick-and-place device is provided with a suction part such as a suction pad that sucks and holds the work, sucks and holds the work at a predetermined place, transports the work in the held state, and releases the suction at the target place. And place the work.

In order to determine the optimum execution conditions of the pick-and-place device, a simulation device using a work model and a robot hand model has been proposed (see, for example, Patent Document 1).

According to the technique described in Patent Document 1, the grippability of the work for each shape of the robot hand is calculated, and the shape of the robot hand is selected based on the grippability to obtain the optimum shape of the robot hand. ..

Japanese Unexamined Patent Publication No. 2015-100866

By the way, there are various types of workpieces that the robot sucks and conveys. For example, there is a work packaged with a packaging material. When the robot sucks the packaged workpiece, random wrinkles are generated on the packaging material. Random wrinkles randomly reduce the suction force of the work by the robot, which may cause the robot to fail to hold the work. That is, when transporting a work whose suction by a suction portion such as a work packaged in a packaging material is randomly reduced, the work falls with a certain probability even if the robot is moved under the same execution conditions. In other words, the fall of the work becomes a stochastic fall.

However, according to the simulation described in Patent Document 1, the grippability is a value calculated in consideration of the shape of the robot hand and the depth of the work, and cannot cope with the stochastic drop of the work.

An object of the present invention is to provide a simulation device, a simulation method, and a simulation program capable of outputting simulation execution conditions in response to a stochastic drop of a work.

The simulation device according to the present invention is a robot for accelerating a robot arm in a state where the work is adsorbed by a reception unit that accepts a permissible value of the probability that the work will fall and an elastic suction unit attached to the robot arm. Based on the correlation between the load moment generated in the suction part due to the difference between the acceleration of the arm and the acceleration of the work and the physical amount indicating the probability that the work will fall from the suction part, the physical amount is determined from the allowable value accepted by the reception part. The corresponding load moment is set as the threshold value, and the execution condition of the simulation in which the load moment when the simulation model including the work, the suction part, and the robot arm is accelerated and moved based on the operation command is equal to or less than the threshold value is output. Including the output section.

The output unit may output the execution condition that minimizes the time for the robot arm to convey the work.

As an execution condition, the reception unit receives input of at least one of the speed and acceleration when the robot arm conveys the work, and the output unit simulates the operation of the robot arm by at least one of the speed and acceleration received by the reception unit. However, the time for the robot arm to convey the work can be output.

The load moment can be derived by subtracting the moment generated by the inertial force from the moment generated in the work.

The simulation method according to the present invention is a robot for accelerating the robot arm in a state where the work is sucked by an elastic suction portion attached to the robot arm and a reception step for accepting an allowable value of the probability that the work will fall. Based on the correlation between the load moment generated in the suction part due to the difference between the acceleration of the arm and the acceleration of the work and the physical amount indicating the probability that the work will fall from the suction part, the physical amount is determined from the allowable value accepted in the reception process. The corresponding load moment is set as the threshold value, and the execution condition of the simulation in which the load moment when the simulation model including the work, the suction part, and the robot arm is accelerated and moved based on the operation command is equal to or less than the threshold value is output. Including the output process.

The simulation program according to the present invention has a reception process for accepting an allowable value of the probability that the work will fall, and a robot for accelerating the robot arm while the work is sucked by an elastic suction portion attached to the robot arm. Based on the correlation between the load moment generated in the suction part due to the difference between the acceleration of the arm and the acceleration of the work and the physical amount indicating the probability that the work will fall from the suction part, the physical amount is determined from the allowable value accepted in the reception process. The corresponding load moment is set as the threshold value, and the execution condition of the simulation in which the load moment when the simulation model including the work, the suction part, and the robot arm is accelerated and moved based on the operation command is equal to or less than the threshold value is output. Let the computer perform the output process.

According to the present invention, the threshold value of the load moment can be obtained from the permissible value of the probability of falling that the user allows by using the correlation between the load moment generated in the suction portion and the physical quantity that represents the probability that the work falls. Since the execution condition in which the load moment is equal to or less than the threshold value can be searched, the execution condition of the simulation, that is, the execution condition of the robot arm can be output in response to the stochastic drop of the work.

It is a block diagram which shows the hardware composition of the simulation apparatus which concerns on 1st Embodiment. It is a block diagram which shows the example of the functional structure of a simulation apparatus. It is a figure which shows the example of the coordinate position in which a robot arm operates. It is a figure which shows the example of a motion program. It is a figure which shows the example of a motion parameter. It is a figure which shows the difference of acceleration of a robot arm and a work. It is a figure which shows the rotational motion of a work. It is a figure which shows the example of the physical model including a robot arm, a suction pad and a work. It is a figure which shows an example of the plot result on the Weibull probability paper. It is a flowchart which shows the flow of the operation executed by the CPU of the simulation apparatus in 1st Embodiment. It is a figure which shows an example of the reception screen which accepts the input of a drop rate. It is a figure which shows the downward force applied to a work at the time of a vertical motion. It is a flowchart which shows the flow of the operation executed by the CPU of the simulation apparatus in 2nd Embodiment. It is a figure which shows an example of the input screen of a fall rate and a motion parameter.

Hereinafter, an example of the embodiment of the present invention will be described with reference to the drawings. The same reference numerals are given to the same or equivalent components and parts in each drawing. In addition, the dimensional ratios in the drawings are exaggerated for convenience of explanation and may differ from the actual ratios.

(First Embodiment)

FIG. 1 is a block diagram showing a hardware configuration of the simulation device according to the first embodiment.

The simulation device 10 of the present embodiment predicts and designs the behavior of the simulation model by using the target device, the work and obstacles handled by the device as the simulation model. Examples of the simulation model include a physical model that expresses a physical phenomenon acting on a device or the like by a mathematical formula, or a dynamic model obtained by system identification or machine learning from measurement data of an actual machine.

In the present embodiment, the device to be simulated is a robot provided with a robot arm. An elastic suction portion is attached to the tip of the robot arm, and the robot arm is a so-called pick-and-place device in which a work is picked up by the suction portion, conveyed, and the work is placed on a destination. In this embodiment, a case where the suction portion is a suction pad and the robot arm sucks and conveys the work will be described in particular.

As shown in FIG. 1, the simulation device 10 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a storage 14, an input unit 15, a monitor 16, and an optical disk drive device 17. And has a communication interface 18. The configurations are connected to each other via the bus 19 so as to be communicable with each other.

In the present embodiment, the ROM 12 or the storage 14 stores a simulation program for executing the simulation. The CPU 11 is a central arithmetic processing unit that executes various programs and controls each configuration. That is, the CPU 11 reads the program from the ROM 12 or the storage 14, and executes the program using the RAM 13 as a work area. The CPU 11 controls each of the above configurations and performs various arithmetic processes according to the program recorded in the ROM 12 or the storage 14.

The ROM 12 stores various programs and various data. The RAM 13 temporarily stores a program or data as a work area. The storage 14 is composed of an HDD (Hard Disk Drive) or an SSD (Solid State Drive), and stores various programs including an operating system and various data.

The input unit 15 includes a pointing device such as a keyboard 151 and a mouse 152, and is used for performing various inputs. The monitor 16 is, for example, a liquid crystal display, and displays various information such as execution conditions of simulation (pick and place operation). The monitor 16 may adopt a touch panel method and function as an input unit 15. The optical disk drive device 17 reads data stored in various recording media (CD-ROM, Blu-ray disc, etc.), writes data to the recording medium, and the like.

The communication interface 18 is an interface for communicating with other devices, and for example, a standard such as Ethernet (registered trademark), FDDI, or Wi-Fi (registered trademark) is used.

Next, the functional configuration of the simulation device will be described.

FIG. 2 is a block diagram showing an example of the functional configuration of the simulation device. FIG. 3 is a diagram showing an example of the coordinate position where the robot arm operates, FIG. 4 is a diagram showing an example of a motion program, and FIG. 5 is a diagram showing an example of a motion parameter.

As shown in FIG. 2, the simulation device 10 has a motion program editing unit 101, a motion parameter editing unit 102, a motion command value calculation unit 103, a dynamics calculation unit 104, a load moment derivation unit 105, and an allowable value reception unit as functional configurations. It has 106, a correlation acquisition unit 107, an execution condition output unit 108, and a 3D display unit 109. Each functional configuration is realized by the CPU 11 reading the simulation program stored in the ROM 12 or the storage 14, expanding it into the RAM 13, and executing the simulation program.

The motion program editing unit 101 accepts the desired motion of the robot arm in the simulation from the user, and accepts the editing of the motion. The user can input or edit a motion program by the input unit 15 while referring to the monitor 16. The motion program editing unit 101 may function as a motion planning unit that automatically generates an operation motion program of the device from the arrangement information of the device and obstacles.

In the motion program editing unit 101, for example, the motion program shown in FIG. 4 is set in order to cause the robot arm to execute the operation as shown in FIG. In the example of the operation shown in FIG. 3, the movement of the robot arm starts from the position p0, and the robot arm reaches the position p3 via the positions p1 and p2. The position p0 represents the coordinates (0,0,0) in the xyz coordinate system, that is, the origin. The position p1 is a coordinate (0,0,30), and is a position moved by 30 in the unit defined by the coordinate in the z direction (height direction) from the position p0. The position p2 is a coordinate (50, 0, 30), and is a position further advanced by 50 in the x direction from the position p1. The position p3 is a coordinate (50, 0, 0), and is a position further advanced by 30 in the −z direction from the position p2. As described above, the motion shown in FIG. 3 is a continuous motion of the robot arm ascending from the origin, a horizontal motion maintaining the height, and a vertical motion descending.

An instruction for such an operation is input to the motion program editing unit 101 as a motion program, for example, as shown in FIG. The motion program shown in FIG. 4 includes three lines of operation commands for executing commands in ascending order of line numbers. In the case of a pick-and-place device, pick-up and place actions may also be included in the motion program. Not only the movement of the robot arm, but also the movements such as holding (suctioning) and releasing the work are instructed by the motion program.

The motion parameter editing unit 102 accepts editing of operation parameters instructed by the motion program from the user. The user can input or edit motion parameters by the input unit 15 while referring to the monitor 16. The motion parameters determine, for example, the maximum speed and the maximum acceleration in the operation by each command, as shown in FIG. The maximum speed and the maximum acceleration can be set as a ratio to the reference speed and the reference acceleration preset in the robot arm. By accepting the input as a ratio to the reference speed, it is possible to prevent the user from inputting a functionally or structurally unreasonable operating speed without knowing the common-sense operating speed of the robot arm. Alternatively, the motion parameter editing unit 102 may accept numerical values of physical quantities of velocity and acceleration instead of the ratio. In addition, in this specification, "acceleration" is used as a term including the concept of both acceleration (positive value) when accelerating and deceleration (negative value) when decelerating. Therefore, the parameter of the maximum acceleration in FIG. 5 includes not only the maximum acceleration during acceleration but also the maximum deceleration during deceleration. However, it is not necessary that the parameters of the maximum acceleration and the maximum deceleration are set uniformly as the same value, and they may be set individually. In this embodiment, the execution condition output unit 108 can determine the optimum parameter. Therefore, the motion parameter editing unit 102 can be used to edit the output result of the execution condition output unit 108.

When the motion command value calculation unit 103 is instructed to start the simulation, the motion command value calculation unit 103 reads the command for the first operation of the motion program obtained by the motion program editing unit 101. Next, the motion command value calculation unit 103 reads the motion parameter corresponding to the command among the motion parameters obtained by the motion parameter editing unit 102. Then, the motion command value calculation unit 103 sequentially calculates and outputs the operation command value used in the simulation from the read command and the motion parameter. Similarly, the motion command value calculation unit 103 reads the subsequent commands and the corresponding motion parameters in order, and outputs the operation command value.

The dynamics calculation unit 104 reads the motion command value output from the motion command value calculation unit 103, the three-dimensional CAD data of the robot, and the simulation model including the work, the suction pad, and the robot arm. The dynamics calculation unit 104 calculates and outputs various data related to the operation of the simulation model considering the dynamics (dynamics) from the read operation command value, CAD data, and the simulation model. The robot's three-dimensional CAD data and simulation model may be stored in advance in the storage 14, may be read from the optical disk drive device 17 and acquired, or may be acquired from an external device via the communication interface 18. May be done. Although the point of using the three-dimensional CAD data is illustrated above, the simulation can be executed without using the three-dimensional CAD data. For example, the simulation may be executed using two-dimensional CAD data, robot dimensional data, robot design data, or the like.

The load moment deriving unit 105 calculates the moment generated in the suction pad according to the difference between the acceleration of the robot arm and the acceleration of the work when the simulation model moves horizontally from the operation of the simulation model calculated by the dynamics calculation unit 104. Derived. In the present specification, the moment generated in the suction pad is referred to as a load moment. The load moment derivation unit 105 may calculate the load moment from the operation of the simulation model, or may derive it from a table registered in advance. The load moment derivation unit 105 repeatedly derives the load moment that changes according to the operation of the simulation model during the simulation. The details of the load moment will be described later.

The permissible value receiving unit 106 accepts the permissible value of the probability that the work will fall. Even if the robot arm is operated by the same operation command, the work will fall with a certain probability depending on the form of the work. That is, the work may fall stochastically. For example, when the robot arm sucks a work packaged by the packaging material, random wrinkles are generated on the packaging material. Random wrinkles randomly reduce the suction force of the work by the robot arm, which may cause the robot arm to fail to hold the work. That is, when transporting a work having a packaging material, even if the robot arm is moved by the same operation command, the work is stochastically dropped. This probability is called the work drop rate. The drop rate tolerance indicates how much drop rate the user can tolerate. For example, when the permissible value is 0.1%, it means that the user allows the work to fall once on average out of 1000 times of transportation.

The correlation acquisition unit 107 acquires data showing the correlation between the load moment and the physical quantity representing the probability that the work will fall from the suction pad. The data acquisition destination is, for example, the storage 14 or an external server. The correlation is obtained by plotting the probability distribution representing the correlation between the load moment and the drop rate obtained by the experiment using a probability paper. For example, the correlation between the load moment and the drop rate is expressed as a Weibull distribution and plotted on the Weibull probability paper to obtain the correlation. The probability distribution is not limited to the Weibull distribution, but a normal distribution, a lognormal distribution, a Gunbel distribution, a Frechet distribution, an exponential function-type minimum value abduction distribution, an exponential distribution, a Rayleigh distribution, or the like can also be applied. In this case as well, the correlation can be obtained by plotting on a probability paper suitable for each. In the following, a case where the correlation between the load moment and the drop rate is obtained by using the Weibull probability paper will be described. The details of the correlation will be described later.

The execution condition output unit 108 determines, as a threshold value, a load moment that is equal to or less than the allowable value received by the allowable value receiving unit 106 from the correlation acquired by the correlation acquiring unit 107. Then, the execution condition output unit 108 outputs the execution condition of the robot arm in which the load moment acting on the work when the simulation using the simulation model is executed is equal to or less than the threshold value.

The 3D display unit 109 displays the operation of the simulation model calculated by the dynamics calculation unit 104 on the monitor 16. For example, the 3D display unit 109 causes the monitor 16 to display a moving image showing how the robot arm conveys the work in three dimensions over time. The display is not limited to three dimensions, and may be two dimensions or numerical values.

Next, the load moment will be described in detail.

6 is a diagram showing the difference in acceleration between the robot arm and the work, FIG. 7 is a diagram showing the rotational movement of the work, and FIG. 8 is a diagram showing an example of a physical model including the robot arm, a suction pad, and the work.

As shown in FIG. 6, a description will be given of a case where the robot arm R is accelerated motion by the acceleration a _r a horizontal direction based on the operation command. Here, an elastic suction pad P is interposed between the robot arm R and the work W. Therefore, the suction pad P is deformed during the acceleration motion, and the work W rotates about the rotation axis assumed for the suction pad P as the center of rotation as shown in FIG. 7, and the acceleration is generated between the work W and the robot arm R. There is a difference. Here, the acceleration of the work _{is represented by aw} , the distance from the center of rotation to the center of gravity of the work W is represented by the symbol L1, and the distance from the center of rotation to the bottom of the work is represented by the symbol L2. Further, the moment of inertia of the work W of the rotating portion of the simulation model is represented by the symbol I. The rotation angle of the work W with respect to the vertical direction is expressed as θ. In the inertial coordinate system in which the work W and the robot arm R move, the moment τ generated in the work W is expressed by the following equation (1).

As shown in the equation (3), the moment τ generated in the work W is expressed by the difference between _{the acceleration a w of the work W and the acceleration a r} of the robot arm R. The moment τ can be regarded as a force applied to the suction pad P as a load, that is, a load moment. Therefore, from the magnitude of the load moment represented by the equation (3), it is possible to determine whether or not the work W is released from being held by the suction pad P and the work W is dropped.

On the other hand, the load moment can also be considered in a translational coordinate system that moves with the robot arm R. In the translational coordinate system, the moment τ generated in the work is expressed by the equation of motion of rotation shown in the following equation (4). M is the load moment.

In translation coordinate system, the moment τ is the horizontal inertia force acting in the horizontal direction by the acceleration motion of the robot arm R (horizontal movement), and the moment ma r _L generated by the inertia force acting on the workpiece W, the suction pad P It can be expressed by the force applied as a load, that is, the sum of the load moment M. Considering the translational coordinate system as in the equation (4), the load moment M applied to the suction pad P is more accurate than the equation (1) because the moment generated by the inertial force acting on the work W is taken into consideration. Can be calculated.

By substituting the equation (2) into the equation (4) and transforming it into the equation of the load moment M, the equation (5) is obtained. Therefore, if you give an acceleration a _r, the acceleration a _w is determined from equation (6) described below, can be determined load moment from equation (5).

_{In the simulation, the acceleration aw} of the work W in the above equations (3) and (5) uses the transfer function G (s) between the suction pad P and the work W, and the following equation (6) is used. It is represented by.

a _w = G (s) × _ar … Equation (6)

Here, the transfer function G (s) is represented by the following equation (7) in which the suction pad P is represented by the physical model shown in FIG. The physical model is constructed by extracting parameters that affect _{the acceleration aw of the} work W from the operation data of the actual machine. For example, in FIG. 10, when brought into accelerated motion of the robot arm R by the acceleration _{a r,} mass _{m Pad} suction pad P that affect acceleration _{a w} of the workpiece W, the rotation damping coefficient _{C Pad,} and torsionally elastic coefficient _{K Pad} Is extracted as a parameter of the suction pad model. In particular, by constructing a suction pad model using only parameters that affect _{the acceleration aw of the work W, external influences other than the acceleration aw} can be reduced, and the acceleration _aw can be obtained with higher accuracy. The transfer function G (s) is expressed, for example, by the equation (7).

Next, the correlation acquired by the correlation acquisition unit 107 will be described in detail. The correlation is obtained by applying various probability distributions as described above. In the following, as an example, the correlation to which the Weibull distribution is applied will be described.

FIG. 9 is a diagram showing an example of plot results on Weibull probability paper.

In the Weibull probability paper, the horizontal axis is represented as the natural logarithm of the load moment M, that is, ln (M). The vertical axis is represented by the following equation (8) using F (M) representing the cumulative frequency of drops.

In an experiment using an actual machine to obtain the experimental value of the above equation (9), a predetermined maximum number of trials is determined, and the work is transferred by the maximum number of trials according to the execution conditions of the pickup place operation in which the same load moment is generated. If the work does not fall in any of the trials for the maximum number of trials, the load moment is increased by changing the execution conditions, and the trials for the maximum number of trials are performed. If the work falls even in one trial, the load moment at that time is recorded and the number of work drops is accumulated. The load moment is increased by the number of trials obtained by subtracting the cumulative number of trials in which the work has fallen from the maximum number of trials. Similarly, the load moment is recorded each time the work falls even in one trial. , Accumulate the number of work drops. This is repeated and the load moment is increased until the number of trials is exhausted. The maximum number of trials is not particularly limited, but for example, a case where the maximum number of trials is 10 will be specifically described below.

First, as a first step, the transfer of the work by the robot arm is tried 10 times for the maximum number of trials under the execution condition in which the same load moment is generated. If the work does not fall even in one of the 10 trials, the execution condition is changed so that a larger load moment is generated, and the work is transferred by the robot arm 10 more times. If the work falls even in one of the 10 trials, the load moment at that time is stored and F (M) is calculated from the equation (9). For example, when the load moment is increased to M1, F (M1) is calculated as 1 / (10 + 1) when one work is dropped. The experimental results with the stored natural logarithm of the load moment M1 and the value obtained by substituting the calculated F (M) into the equation (8) are plotted on the Weibull probability paper.

Subsequently, as the second step, the work is transferred by the robot arm (maximum number of trials-cumulative number of drops of the work so far) by changing to the execution condition in which a larger load moment M is generated. In the first stage described above, since the work is dropped once, the cumulative number of drops is 1. Therefore, in the second stage, 10-1 is set to 9 times as the number of trials. In this way, it is assumed that the load moment is increased while performing 9 trials, and two workpieces are dropped by the load moment M2. In this case, since the cumulative number of drops of the work is 3 in the first stage and the second stage, F (M2) in the equation (9) is calculated as 3 / (10 + 1), and the load moment M2 is stored. To. As in the first stage, the experimental results are plotted on the Weibull probability paper.

In the third stage, the number of transfer trials is set to 10-3 = 7, and the load moment M is increased until even one work is dropped, as in the first and second stages. By repeating the process of increasing the load moment and memorizing the load moment when the work is dropped until the cumulative number of drops reaches 10, the Weibull distribution is generated on the Weibull probability paper as shown in FIG. can get.

That is, in general, when the maximum number of trials is k and the cumulative number of drops of the work up to the previous time is m, the trials are performed a plurality of times as the number of trials (km) at this stage. Weibull distribution will be obtained. Since the number of trials decreases with each stage, the number of trials can be halved and the work efficiency is extremely high as compared with the case where the maximum number of trials is performed at all stages.

From the experimental results plotted on the Weibull probability paper, an approximate straight line can be obtained by the least squares method. Assuming that the slope of the approximate straight line is a and the intercept is b, the approximate straight line is expressed by the following equation (10).

Here, the vertical axis of FIG. 9 (the left side of the equation (10)) is not the probability (fall rate) of the work falling from the robot arm itself. However, when the fall rate p and the cumulative frequency F (M) of the fall are small values, they are almost the same and can be regarded as F (M) = p. Therefore, it can be said that the left side of the equation (10) is a physical quantity representing the fall rate. As a whole, the equation (10) can be said to be an equation showing the correlation between the work drop rate p and the load moment M. When the equation (10) is transformed into the equation of the load moment M, the following equation (11) is obtained.

If the desired drop rate p is known, the load moment M corresponding to the drop rate p can be obtained by substituting the drop rate p into F (M) in the equation (11).

As described above, the above equations (10) and (11) are the contents of the correlation acquired by the correlation acquisition unit 107.

Next, the operation of the simulation device 10 will be described.

FIG. 10 is a flowchart showing a flow of operations executed by the CPU of the simulation apparatus according to the first embodiment. The simulation process is performed by the CPU 11 reading the simulation program from the ROM 12 or the storage 14, deploying it in the RAM 13, and executing it. FIG. 11 is a diagram showing an example of a reception screen that accepts input of the drop rate.

The CPU 11 receives the allowable value for the drop rate of the work W as the allowable value receiving unit 106 (step S101). The allowable value of the accepted drop rate is hereinafter referred to as a drop rate p. For example, the CPU 11 displays a screen as shown in FIG. 11 on the monitor 16 and allows the input box 201 to input the drop rate p. The user can input the drop rate p via the input unit 15.

The CPU 11 refers to the correlation acquired by the correlation acquisition unit 107 as the execution condition output unit 108, and specifies the load moment M corresponding to the drop rate p received in step S101 as the threshold value Tp (step S102). The correlation is as shown in FIGS. 9, 10 and 11 described above, and the load moment M can be calculated by substituting the drop rate p into F (M) of the equation (11). The threshold value Tp is temporarily stored in the storage 14 or the RAM 13.

The CPU 11 receives the motion program and the motion parameter as the motion program editing unit 101 and the motion parameter editing unit 102 (step S103). If the motion program and a part of the motion program are decided in advance, the input of only the remaining part may be accepted. For example, the CPU 11 displays the screen shown in FIG. 11 on the monitor 16, and receives input of the moving distance of the robot arm R, that is, the work W, which is set as a motion program, in the input boxes 202 to 204. Further, although not shown, when accepting a motion parameter, the CPU 11 sets the initial value of the simulation. This is because, as will be described later, the motion parameters can be changed in order to search for the optimum motion parameters. The processing after step S103 may be started, for example, by the user pressing the "optimal parameter calculation execution" button 205 in FIG. 11.

The CPU 11 reads the motion program and the motion parameter as the motion command value calculation unit 103, and calculates the motion command value (step S104).

As the dynamics calculation unit 104, the CPU 11 executes a simulation based on the motion command value and the simulation model (step S105). As the load moment derivation unit 105, the CPU 11 calculates all the load moments M that change during the simulation based on the equations (5) and (6) (step S106).

The CPU 11 determines whether or not the load moment M is equal to or less than the threshold value Tp as the execution condition output unit 108 (step S107). When the load moment M is larger than the threshold value Tp (step S107: NO), the CPU 11 proceeds to the process of step S110. When the load moment M is equal to or less than the threshold value Tp (step S107: YES), the CPU 11 calculates the tact time for the robot arm R to convey the work W in the simulation (step S108). The takt time is the time from when the work W is picked up until it is placed. The CPU 11 associates the motion parameter used in the simulation in step S105 with the calculated takt time and stores it in the storage 14 or the RAM 13 (step S109).

The CPU 11 determines whether or not the search for the tact time at which the load moment M is equal to or less than the threshold value Tp has been completed for all the combinations of motion parameters that can be changed (step S110). If the search for all combinations of motion parameters has not been completed (step S110: NO), the CPU 11 returns to the process of step S104.

When the search for all the combinations of motion parameters is completed (step S110: YES), the CPU 11 displays the combination of the motion parameters that minimizes the tact time in the stored state as the execution condition (step S111). For example, as shown in the calculation result display window 206 of FIG. 11, the CPU 11 displays a motion parameter in each operation of the robot arm R and a takt time at the time of operation according to the motion parameter.

The CPU 11 performs a simulation as the motion command value calculation unit 103 and the dynamics calculation unit 104 according to the motion parameters determined in step S111, and displays the simulation result as the 3D display unit 109 (step S112). The simulation result is displayed, for example, as a moving image of the operation of the simulation model. The CPU 11 ends the simulation process.

As described above, according to the simulation device 10 of the present embodiment, when the work W during transportation falls stochastically, the motion parameter of the robot arm R can be output in response to the stochastic fall. In particular, since the simulation device 10 uses the correlation between the physical quantity representing the drop rate of the work and the load moment, the drop rate allowed by the user is converted into the threshold value Tp of the load moment, and the load moment M calculated during the simulation is used. Can be compared.

Since the simulation device 10 changes the motion parameters and searches for the motion parameters that minimize the tact time while satisfying the drop rate p specified by the user, it is possible to output the optimum motion parameters in terms of work efficiency. ..

The present invention is not limited to the above embodiment, and various modifications can be made.

In the above embodiment, since the Weibull distribution is applied, the formula (8) is used as a physical quantity representing the drop rate of the work. However, as mentioned above, various probability distributions can be applied to the present invention. Therefore, the work drop rate can be used as it is as a physical quantity representing the work drop rate.

Further, in the above embodiment, a case where the optimum maximum speed and maximum acceleration of the robot arm among the motion parameters are output as execution conditions has been described. However, the optimum execution conditions may be output by other motion parameters. For example, execution conditions may be output with motion parameters such as acceleration (deceleration) time and jerk (jerk).

Further, in the above embodiment, when determining the optimum motion parameter in the horizontal direction, a case where a search is performed for all combinations of motion parameters in order to search for a motion parameter that minimizes the tact time has been described. However, the optimum motion parameter does not necessarily have to be the motion parameter that minimizes the takt time. The search method may be changed from the viewpoint of shortening the simulation time. For example, the velocity or acceleration is gradually reduced from a predetermined value or maximum value larger than the median value, and the parameter when the threshold value Tp is first lowered is also set as the optimum motion parameter that minimizes the takt time. good.

Further, in the above embodiment, the motion parameter is output in consideration of the falling rate during the horizontal movement by using the correlation between the load moment generated when the robot arm moves horizontally and the physical quantity representing the falling rate of the work. Was. However, motion parameters may be output in response to a probabilistic fall during vertical motion.

FIG. 12 is a diagram showing a downward force applied to the work during vertical movement.

When the robot arm R accelerates upward and downward in the vertical direction, the downward force acting on the work W is the sum of gravity and inertial force. That is, as shown in FIG. 12, when the robot arm R is accelerated motion by the acceleration a _r vertically, the center of gravity of the workpiece W to be included in the simulation model, and gravity m × g, and the inertial force m × a _z The power of harmony is applied. Here, m is the mass of the work W, g is the gravitational acceleration, and _az is the acceleration of the work W. The sum of the gravity m × g and the inertial force m × _az acts on the work W as a downward force. The correlation between the downward force and the physical quantity representing the drop rate of the work W can be obtained by applying it to the probability distribution in the same manner as the above-mentioned correlation between the load moment and the physical quantity representing the drop rate.

As a result, the simulation device 10 receives the desired drop rate p from the user, derives the downward force corresponding to the drop rate p as a threshold value from the correlation between the downward force and the physical quantity representing the drop rate, and downwards. It is possible to determine the motion parameter in which the force of is less than or equal to the threshold value. Similar to the case of determining the optimum motion parameter in the horizontal direction, the motion parameter having the minimum takt time among the motion parameters in which the downward force is equal to or less than the threshold value is searched for, and the optimum motion parameter is obtained.

(Second Embodiment)

In the first embodiment, the case where the optimum motion parameter is obtained from the drop rate p input by the user has been described. In the second embodiment, a mode in which the user can input a motion parameter together with the drop rate p will be described.

FIG. 13 is a flowchart showing a flow of operations executed by the CPU of the simulation apparatus according to the second embodiment. FIG. 14 is a diagram showing an example of a drop rate and motion parameter input screen.

The CPU 11 accepts the input of the drop rate p and the motion parameter (step S201). The CPU 11 displays, for example, the input screen shown in FIG. 14 on the monitor 16. The input screen is provided with an input box 301 and input windows 302 to 304. The input box 301 accepts the input of the drop rate p. The input windows 302 to 304 indicate the motion section of the robot arm defined by the motion program, and receive input of at least one motion parameter of velocity and acceleration in the motion section.

The CPU 11 specifies a threshold value corresponding to the drop rate p input by the user from the correlation between the load moment and the physical quantity representing the drop rate and the correlation between the downward force and the physical quantity representing the drop rate (step S202). ..

The CPU 11 calculates a motion command value based on a predetermined motion program and a motion parameter input in step S101 (step S203). The CPU 11 executes a simulation (step S204) and calculates a tact time in which the robot arm R conveys the work W (step S205).

The CPU 11 further derives the load moment M and the downward force generated during the simulation (step S206). The CPU 11 determines whether or not the derived load moment M or the downward force is equal to or less than the threshold value specified in step S202 (step S207).

When the load moment M and the downward force are equal to or less than the threshold value (step S207: YES), the CPU 11 displays the tact time calculated in step S205 (step S208). Then, the CPU 11 ends the simulation process.

When at least one of the load moment M and the downward force exceeds the threshold value (step S207: NO), the CPU 11 does not satisfy the tact time calculated in step S205 and the drop rate p in which the motion parameter is input to the monitor 16. The effect is displayed (step S209). For example, the CPU 11 displays the message "The input motion parameter does not satisfy the allowable drop rate. Please change the motion parameter" in the result display window 305 as shown in the screen of FIG. Here, the CPU 11 may not only display a message that the allowable drop rate is not satisfied, but may also display a motion parameter candidate that satisfies the drop rate p input by the user. In this case, the CPU 11 searches for a motion parameter that satisfies the drop rate p input by the user, and displays the search result as a candidate for the motion parameter, as in the first embodiment. Then, the CPU 11 ends the simulation process.

As described above, according to the simulation device 10 of the second embodiment, the user inputs a desired motion parameter, and the tact time when the input motion parameter is used as an execution condition and whether or not the drop rate is satisfied. Can be confirmed.

In each of the above embodiments, various processors other than the CPU may execute the simulation process in which the CPU reads the software (program) and executes the simulation process. In this case, the processor includes a PLD (Programmable Logic Device) whose circuit configuration can be changed after manufacturing an FPGA (Field-Programmable Gate Array), an ASIC (Application Specific Integrated Circuit), and the like. An example is a dedicated electric circuit or the like, which is a processor having a circuit configuration designed exclusively for the purpose. Further, the simulation process may be executed by one of these various processors, or a combination of two or more processors of the same type or different types (for example, a plurality of FPGAs, and a combination of a CPU and an FPGA, etc.). ) May be executed. Further, the hardware-like structure of these various processors is, more specifically, an electric circuit in which circuit elements such as semiconductor elements are combined.

Further, in each of the above embodiments, the embodiment in which the simulation program is stored (installed) in the storage 14 or the ROM 12 in advance has been described, but the present invention is not limited thereto. The program may be provided in a form recorded on a recording medium such as a CD-ROM (Compact Disk Read Only Memory), a DVD-ROM (Digital Versaille Disk Read Only Memory), and a USB (Universal Serial Bus) memory. Further, the program may be downloaded from an external device via a network.

10 Simulation device 101 Motion program editing unit 102 Motion parameter editing unit 103 Motion command value calculation unit 104 Dynamics calculation unit 105 Load moment derivation unit 106 Allowable value reception unit 107 Correlation acquisition unit 108 Execution condition output unit 109 3D display unit

Claims (6)

The reception unit that accepts the allowable value of the probability that the work will fall, and
A load moment generated in the suction portion due to the difference between the acceleration of the robot arm and the acceleration of the work when the robot arm is accelerated while the work is attracted by the elastic suction portion attached to the robot arm. Based on the correlation with the physical quantity representing the probability that the work will fall from the suction unit, the load moment corresponding to the physical quantity determined from the permissible value received by the reception unit is set as the threshold value, and the work and the suction In a simulation in which a simulation model including a unit and the robot arm is accelerated based on an operation command, the acceleration of the work is calculated based on the transmission function between the suction unit and the work and the acceleration of the robot arm. An output unit that calculates a load moment based on the calculated acceleration of the robot arm and the calculated acceleration of the work, and outputs a simulation execution condition that the calculated load moment is equal to or less than the threshold value.
Simulation equipment including. The simulation device according to claim 1, wherein the output unit outputs the execution condition that minimizes the time for the robot arm to convey the work. As the execution condition, the reception unit receives at least one input of speed and acceleration when the robot arm conveys the work.
The first or second aspect of the present invention, wherein the output unit simulates the operation of the robot arm by at least one of the speed and the acceleration received by the reception unit, and outputs the time for the robot arm to convey the work. Simulation equipment. The simulation apparatus according to any one of claims 1 to 3, wherein the load moment is derived by subtracting a moment generated by an inertial force from a moment generated in the work. The reception process that accepts the allowable value of the probability that the work will fall, and
A load moment generated in the suction portion due to the difference between the acceleration of the robot arm and the acceleration of the work when the robot arm is accelerated while the work is attracted by the elastic suction portion attached to the robot arm. Based on the correlation with the physical quantity representing the probability that the work will fall from the suction portion, the load moment corresponding to the physical quantity determined from the permissible value received in the reception step is set as the threshold value, and the work and the suction In a simulation in which a simulation model including a unit and the robot arm is accelerated based on an operation command, the acceleration of the work is calculated based on the transmission function between the suction unit and the work and the acceleration of the robot arm. An output process that calculates a load moment based on the calculated acceleration of the robot arm and the calculated acceleration of the work, and outputs a simulation execution condition that the calculated load moment is equal to or less than the threshold value.
A simulation method in which a computer performs. The reception process that accepts the allowable value of the probability that the work will fall, and
A load moment generated in the suction portion due to the difference between the acceleration of the robot arm and the acceleration of the work when the robot arm is accelerated while the work is attracted by the elastic suction portion attached to the robot arm. Based on the correlation with the physical quantity representing the probability that the work will fall from the suction portion, the load moment corresponding to the physical quantity determined from the permissible value received in the reception step is set as the threshold value, and the work and the suction In a simulation in which a simulation model including a unit and the robot arm is accelerated based on an operation command, the acceleration of the work is calculated based on the transmission function between the suction unit and the work and the acceleration of the robot arm. An output process that calculates a load moment based on the calculated acceleration of the robot arm and the calculated acceleration of the work, and outputs a simulation execution condition that the calculated load moment is equal to or less than the threshold value.
A simulation program that causes a computer to execute.

Images