KR0176456B1 - Path control method of robot system - Google Patents

Abstract

The present invention relates to a path control method performed by a main controller to calculate a target position value of each joint in an articulated robot system. In an articulated robot system, splitting points are divided by dividing a continuous path to be performed when performing a task. Division point setting process to set; A target position value calculating step of calculating target position values of each joint with respect to the split points; An experimental path calculation process of calculating an experimental path by performing a simulation using the target position values; An error calculating process of setting the maximum value of the difference between the continuous path and the experimental path as the interval experiment error in each interval when the division points are referred to as intervals; Calculating the target position value and calculating the experimental path until the interval experimental error newly obtained for the newly set split point becomes less than or equal to the tolerance after setting an additional split point for the interval in which the interval experimental error is larger than the tolerance. Additional splitting point setting process for repeating the process and the error calculation process; And outputting position commands according to the target position values when the interval experiment errors are all less than or equal to the tolerance. Such a method greatly reduces the number of operations, thereby speeding up the operation of the robot body and reducing power consumption.

Description

Path control method of robot system

1 is a block diagram showing the configuration of a robot system.

2 is a block diagram showing the configuration of the position controller and the servo driver.

3 is a view illustrating welding of a sheet iron plate.

4 is a flowchart illustrating a path control method of a conventional robot system.

5A and 5B are diagrams for explaining a case where target position values are set according to a conventional method.

6 is a view showing a two-axis robot body.

7 is a flowchart illustrating a path control method of the robot system according to the present invention.

8A and 8B are diagrams for explaining a case of setting target position values according to the present invention.

* Explanation of symbols for main parts of the drawings

101: teaching box 102: terminal

103: main controller 104: input and output board

105: 2-way RAM 106: Position controller

107: servo driver 108: robot body

The present invention relates to a robot system, and more particularly, to a path fishing method for implementing a continuous path in an articulated robot system.

FIG. 1 is a block diagram showing the configuration of the robot system. The teaching box 101, the terminal 102, the main controller 103, the input / output board 104, the interactive ram 105, the position controller 106, the servo driver 107 and the robot body 108 is configured.

In FIG. 1, the teaching box 101 is composed of a central processing unit (CPU), a monitor, a keyboard, and the like, and teaches an operation performed by the robot body 108. Terminal 102 is used when a user wants to enter a command or enter a program. The main controller 103 is a part that performs overall control of the operation of the robot body 108, and provides a function of receiving a user's command, writing a work program of the robot, and analyzing the robot state and notifying the user of this. In addition, the main controller 103 is responsible for outputting a position command every predetermined time to the position controller 106 in order to perform a task designated by the user. The input / output board 104 performs an operation for exchanging necessary control signals between the robot system and the external environment. The position controller 106 inputs a position command to which the robot body 108 should move from the main controller 103, inputs an actual position value to the robot body 108, and performs feedback control according to the result. The speed command according to the output to the servo driver 107. The servo driver 107 drives the robot body 108 in accordance with the speed command. In this case, when the robot body 108 includes a plurality of joints and a servo motor M corresponding to each joint, the servo driver 107 is configured for each joint. Meanwhile, the interactive RAM 105 performs a function as a buffer in which data to be transmitted and received is temporarily stored so that operations of the main controller 103 and the position controller 106 can be independently performed.

At this time, the main controller 103 outputs a position command to the position controller 106 as described above. The position command is a unit sampling of each joint constituting the robot body 108 to perform a task designated by a user. It means the distance value to move during the time. When such a position command is applied to the position controller 106, the position controller 106 calculates an error position value by subtracting the actual position value of each joint constituting the robot body 108 from the target position value. After that, the control operation is performed using this. Although the actual position value is not shown in the figure, it is detected through position detection means for detecting the current position of each joint constituting the robot body 108.

The robot body 108 is composed of a plurality of joints and connecting members connecting the joints and a plurality of servo motors (M) for driving each joint, and the output of the servo driver 107 is each servo motor (M). To be applied.

There may be various control schemes performed by the position controller 106. For example, the differential control method (D-opration), integral control method (I-operation), proportional control method (P-opertion), and the PI control method, PD control method, PID control method using a combination thereof There may be.

2 shows an embodiment of the configuration of the position controller 106 and the servo driver 107. The position controller 106 includes a first subtractor 201, a first control operation unit 202, and the like. The servo driver 107 includes a second subtractor 203, a second control operation unit 204, a driving circuit 205, a servo amplifier 206, and the like. In addition, an integrator that integrates the outputs of the speed detecting means 209 and the speed detecting means 208 for detecting the speed of the joints constituting the robot body 108 and applies them to the first subtractor 201 as actual position values. 207 and the like are shown in the figure. The position detecting means here comprises a speed detecting means 208 and an integrator 207.

In FIG. 2, the first subtractor 201 subtracts the actual position value from the position command output from the main controller 103. The first control operation unit 202 inputs an error position value from the first subtractor 201 to perform a proportional calculus control (PID) operation. The second subtractor 203 subtracts the actual speed value from the value output from the first control operation unit 202. The actual speed value refers to a value returned after being detected by the speed detecting means. The second control operation unit 204 performs a proportional integral control operation on the error position value output from the second subtractor 203. The driving circuit 205 converts the output of the second control operation unit 204 into a driving signal, and the servo amplifier 206 amplifies the driving signal so that the servo motor (M) of the corresponding joint in the robot body unit 108 is provided. ) To apply it.

Before describing the path control method performed by the main controller 103, the configuration of the robot body 108 will be briefly described with reference to FIG.

FIG. 6 shows the robot body 108 having a two-axis shape, and includes a first joint 601, a second joint 602, a first connection member 603, a second connection member 604, and a gripper. 605).

In FIG. 6, the gripper 605 refers to a part to which a welding rod or the like is attached when welding is to be performed as shown in FIG. The first joint 601 and the second joint 602 are attached with servo motors M for driving them, respectively, and the first connection member 603 connects the gripper 605 and the first joint 601. The second connection member 604 connects the first joint 601 and the second joint 602.

As can be seen from the above figure, the position command is not calculated according to the target position value of the gripper 605 that actually performs the work, but the target that each joint has in order to position the gripper 605 to a suitable place to perform the work. It is calculated according to the position value.

Here, the conventional path control method will be described with reference to FIGS. 4 and 5.

4 is a flowchart illustrating a path control method of a conventional robot system.

First, when the user designates a task, a portion of the part constituting the robot body 108 that actually performs the task, that is, a continuous path through which the gripper 605 passes, is designated from the teaching box 101.

When the continuous path (CP) of the gripper 605 is set as described above, the continuous path is divided into predetermined intervals L in step 401 to calculate coordinate values of the split point.

In step 402, the coordinate values of the split point are used to calculate target position values that the joints of the robot body 108 should have such coordinate values so that the gripper 605 has such coordinate values.

In step 403, a point-to-point simulation is performed using the target position values of the joints. When the simulation is performed, it is possible to obtain an experimental path through which the gripper 605 passes. Step 404 compares each coordinate value of the experiment path with each coordinate value of the continuous path and sets the experiment error εa to be the maximum.

Step 405 determines whether or not the experimental error ε a is within a tolerance ε d. If the experimental error εa is less than or equal to the tolerance εd, the process proceeds to step 407 and outputs position commands corresponding to the target position values calculated in step 402. Otherwise, In step 406, the value of the predetermined interval L is divided in half, and then, in step 401, the above-described process is repeated. That is, the loop from step 401 to step 406 is repeatedly performed until the experimental error ε a is less than or equal to the tolerance ε d.

5A and 5B are diagrams for describing a path control method according to a conventional method.

In FIG. 5A, N1, N2, N3, N4, ..., etc. represent split points set primarily, a solid path represents a continuous path, and a dotted line path is simulated according to the point-to-point method. The experimental path obtained as a result of the experiment is shown. At this time, it can be seen that the maximum error, that is, the experimental error εa (1), occurred in the section determined between the splitting point N3 and the splitting point N4. If the experimental error εa (1) is larger than the tolerance εd, secondary splitting points NO1, NO2, NO3, ... are set as shown in FIG. Position values are calculated. A dotted line in FIG. 5B shows an experimental path obtained as a result of a point-to-point simulation using target position values for the primary dividing points and target position values for the secondary dividing points. In FIG. 5B, the experimental error ε a (2) represents the maximum value of the difference between the experimental path and the continuous path obtained secondarily. If the experimental error obtained at this time also exceeds the tolerance, the above-described method is repeatedly performed.

As can be seen from the above description and drawings, the conventional method reconstructs all sections even when the experimental error εa exceeds the tolerance εd in only one section when the section is divided between the split points and the split point. Because of the partitioning, there is a problem that a lot of unnecessary operations are performed. In other words, even in the section where the experimental error εa is within the tolerance εd, the splitting point is newly calculated, and the point-to-point simulation is performed using the splitting point so that many unnecessary operations are performed. Can be. Performing such an unnecessary operation causes a long time for performing the work of the robot body.

In addition, the conventional method is inefficient because it simply divides all the continuous paths at predetermined intervals without considering the shape of the robot body 108.

Accordingly, an object of the present invention is to provide a path control method of a robot system that can reduce unnecessary operations in calculating a position command for implementing a continuous path in the robot body to solve the above problems. .

In order to achieve the above object, the path control method of the present invention robot system in the articulated robot system,

A split point setting process of setting split points by dividing a continuous path to be performed when performing a work;

A target position value calculating step of calculating target position values of each joint with respect to the split points;

An experimental path calculation process of calculating an experimental path by performing a simulation using the target position values;

An error calculating process of setting the maximum value of the difference between the continuous path and the experimental path as the interval experiment error in each interval when the division points are referred to as intervals;

Calculating the target position value and calculating the experimental route until the interval test error is newly obtained for the newly set split point is less than or equal to the tolerance after setting an additional split point for the section in which the interval test error is greater than the tolerance. An additional split point setting process for repeating the process and the error calculating process;

And if the interval experiment errors are less than or equal to a mode tolerance, outputting position commands according to the target position values.

Next, the path control method of the present invention robot system will be described in more detail with reference to the accompanying drawings.

7 is a flowchart illustrating a path control method of the robot system according to the present invention.

In FIG. 7, in step 701, the continuous path is divided to set a split point. Here, the setting of the split point is made in consideration of the shape of the robot body 108 without setting the split points unequally and at equal intervals unlike the prior art. For example, in the case of having a two-axis robot body 108 as shown in FIG. 6, the split point may be set when the continuous path needs to move both joints more than a certain range. When the split point is set as described above, the length of each split point and the split point, that is, the length of the section, has a nonlinear characteristic unlike in the prior art.

Step 702 is a step of calculating a target position value for the split point. This is performed as in the prior art, and calculates a value indicating a target position where each joint of the robot body 108 should be positioned so that the gripper 605 of the robot body 108 has the coordinate value of each split point. It is a process.

Step 703 performs a point-to-point simulation using the target position values of the joints. When the simulation is performed, it is possible to obtain an experimental path through which the gripper 605 passes.

In operation 704, an experimental error ε a is calculated for each section by comparing each coordinate value of the experiment path with each coordinate value of the continuous path. Here, as an example of a method for calculating the experimental error (ε1, ε2, ε3, ε4, ε5, ...) for each section, the method is to calculate the absolute value of the difference between the continuous path and the experimental path at the midpoint of each section. There is this.

In operation 705, it is determined whether the experimental errors ε1, ε2, ε3, ε4, ε5, ... for each section are all within the tolerance εd or not. If the experimental errors for each section (ε1, ε2, ε3, ε4, ε5, ...) are less than or equal to the tolerance εd, the process proceeds to step 707 and the target calculated for each split point. If the position values are output, otherwise, the flow proceeds to step 706.

In step 706, an additional split point is set. Unlike the conventional method, an additional split point is set only for a section in which the experimental error for each section is larger than the tolerance εd. Here, as in step 701, additional split points may be set nonlinearly, and the number of additional split points may be freely determined, unlike in the prior art. In other words, if there is no adjustment from the outside, a predetermined number of additional splitting points is set as a default value, and if data for the number of additional splitting points is input from the outside, the number of additional splitting points is determined accordingly. If the additional split point is set in this way, the process proceeds to step 702. That is, the loop from step 702 to step 706 is repeatedly performed until the experimental error for each section becomes less than or equal to the tolerance εd. Here, in steps 702, 703, and 704, when the step is performed due to the additional splitting point, only the part of the newly added additional splitting point is performed to reduce the number of operations required.

8A and 8B are views for explaining a case of setting a path control method according to the present invention.

In Figs. 8A and 8B, N1, N2, N3, ..., etc. indicate a split point set primarily, and NO1 indicates an additional split point set secondary. Also, ε1, ε2, ε3, ... represent the difference between the primary experimental path obtained from the point-to-point simulation and the continuous path given from the teaching box, that is, the primary experimental error. , epsilon 02 and the like indicate the difference between the secondary experimental path and the continuous path, that is, the secondary experimental error obtained by using additional target position values obtained as additional split points.

As can be seen from FIGS. 8a and 8b, the present invention is to set an additional split point only for a section in which the experimental error is larger than the tolerance, thereby calculating the additional target position values and the experimental error. The operation to find is greatly reduced.

As described above, the present invention is to greatly reduce the operation performed to calculate the position commands for implementing the continuous path in the robot system, the advantage that the robot body can perform the fast operation. In addition, there is an advantage that the number of operations performed can be reduced to prevent the power consumption.

Claims (4)

CLAIMS 1. A method of driving an articulated robot system, comprising: a split point setting process of setting split points by dividing a continuous path to be performed when performing a work; A target position value calculating step of calculating target position values of each joint with respect to the split points; An experimental path calculation process of calculating an experimental path by performing a simulation using the target position values; An error calculating process of setting the maximum value of the difference between the continuous path and the experimental path as the interval experiment error in each interval when the division points are referred to as intervals; Calculating the target position value and calculating the experimental path until the interval experimental error newly obtained for the newly set split point becomes less than or equal to the tolerance after setting an additional split point for the interval in which the interval experimental error is larger than the tolerance. Additional splitting point setting process for repeating the process and the error calculation process; And outputting position commands according to the target position values when the interval experiment errors are all less than or equal to the tolerance. The method of claim 1, wherein the number of additional split points is set by the user. The method of claim 1, wherein the experiment path calculating process is performed in a point-to-point manner. The method of claim 1, wherein the dividing point setting process and the additional dividing point setting process are set so that the dividing point is set non-linearly in consideration of the shape of the robot body.