JP2778922B2 - How to determine the work order and work sharing of multiple robots - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for determining the work order and work sharing of a plurality of robots, and more particularly, to the work order and the work order when a plurality of robots having a common moving space share work. It relates to a method for determining the work allotment.

[0002]

2. Description of the Related Art Conventionally, when work is shared by a plurality of robots, the work order and work sharing of each robot are determined by a human, and teaching is performed.

[0003]

In the above-mentioned conventional method of manually determining the work order and work sharing of a plurality of robots, when the amount of work increases, the number of possible combinations of the work order and work sharing is extremely large. Therefore, it was difficult to determine the work order and work sharing so as to minimize the total work time. SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and has as its object the work order and the work order of a plurality of robots capable of determining the work order and work sharing so as to minimize the total work time. The purpose of the present invention is to provide a method of determining work sharing.

[0004]

According to a first aspect of the present invention, a first motion simulation is performed on a plurality of robots having a common moving space based on a certain work order. The total work time obtained as a result of the first operation simulation and the work with the longest wait time of the robot with the longest work time are extracted, and the work of the longest wait time of the robot with the longest work time is extracted. By changing the order, the certain work order is updated, a second operation simulation is performed based on the updated certain work order, and the results of the first and second operation simulations are compared to obtain a desired total. The first and second operation simulations are repeated while updating the certain work order until the required work time is obtained or the predetermined number of repetitions is reached. The Ukoto is configured as a method for determining the working order of the plurality of robots made by determining work order for each robot. The second invention performs a third operation simulation on a plurality of robots having a common movement space based on a certain work assignment, and calculates a total operation time obtained as a result of the third operation simulation,
The task with the longest waiting time of the robot with the longest working time is extracted, and one of the tasks within the operating range of the other robot among the tasks of the robot with the longest working time is shared by the other robot. , The fourth task simulation is performed based on the updated certain task assignment, and the results of the third and fourth operation simulations are compared to obtain a desired total. By repeatedly performing the third and fourth operation simulations while updating the certain work sharing until a work time is obtained or a predetermined number of repetitions, a plurality of work shares determined by each robot are determined. This is a method for determining the work sharing of the platform robot.

According to a third aspect of the present invention, a first operation simulation is performed for a plurality of robots having a common movement space based on a certain operation order, and a total operation time obtained as a result of the first operation simulation is obtained. And the work with the longest wait time of the robot with the longest work time is extracted, and the work order of the longest wait time of the robot with the longest work time is changed to update the above-mentioned certain work order. , Based on the updated work order,
Is performed, and the results of the first and second operation simulations are compared to obtain a desired total work time or to update the certain work order until a predetermined number of repetitions is reached. By repeatedly performing the first and second motion simulations, the work order of each robot is determined, and then a third motion simulation is performed on the plurality of robots based on a certain work share. The total work time obtained as a result of the operation simulation of the above and the work with the longest wait time of the robot with the longest work time are extracted, and among the shared work of the robot with the longest work time, the movements of other robots are extracted. The task sharing is updated by changing one of the tasks in the range to the task of the other robot, and based on the updated task sharing. Then, a fourth operation simulation is performed, and the results of the third and fourth operation simulations are compared to obtain a desired total operation time or to update the certain operation sharing until a predetermined number of repetitions is obtained. On the other hand, a method of determining the work order and the work sharing of a plurality of robots, which determines the work sharing of each robot by repeatedly performing the third and fourth motion simulations. Further, there is provided a method for determining the work order and work sharing of a plurality of robots, wherein the work order is a welding order, and the work sharing is a welding work share.

[0006]

According to the first aspect, the first operation simulation is performed on a plurality of robots having a common moving space based on a certain work order. The total work time obtained as a result of the first operation simulation and the work with the longest wait time of the robot with the longest work time are extracted, and the work with the longest wait time of the robot with the longest work time is extracted. By changing the work order, the certain work order is updated. A second operation simulation is performed based on the updated work order. The results of the first and second operation simulations are compared to obtain a desired total work time, or the certain work order is updated until a predetermined number of repetitions is reached.
By repeatedly performing the first and second motion simulations, the work order of each robot is determined. In this way, the work order of each robot can be determined so that the total work time of the plurality of robots is minimized.

According to the second aspect, the third operation simulation is performed on a plurality of robots having a common movement space based on a certain task sharing. The total work time obtained as a result of the third operation simulation and the work with the longest wait time of the robot with the longest work time are extracted, and the other work among the shared work of the robot with the longest work time is extracted. By changing one of the tasks within the operation range of the robot to the assignment of the other robot, the above-mentioned certain task order is updated. A fourth operation simulation is performed based on the updated work sharing. The results of the third and fourth operation simulations are compared to obtain a desired total work time, or the certain work sharing is updated until a predetermined number of repetitions is reached. The repetition of the motion simulation determines the work sharing of each robot. In this way, the work sharing of each robot can be determined so that the total work time of the plurality of robots is minimized. According to a third aspect of the present invention, the work order of each robot is determined according to the first aspect of the invention, and then the task sharing of each robot is determined according to the second aspect of the invention.
This makes it possible to determine the work order and work sharing of each robot so that the total work time of the plurality of robots is minimized. Further, when the above-mentioned work order is the welding order and the above-mentioned work sharing is the welding work sharing, the welding work order and the welding work sharing are determined so that the total welding work time of the plurality of robots is minimized. be able to.

[0008]

Embodiments of the present invention will be described below with reference to the accompanying drawings to provide an understanding of the present invention. The following embodiment is an example embodying the present invention and is not intended to limit the technical scope of the present invention. FIG. 1 is a flowchart showing a schematic configuration of a method for determining a welding order of a plurality of welding robots according to an embodiment (first embodiment) of the first invention, and FIG. 2 is a concept of the plurality of welding robots. FIG. 3 is an explanatory view showing the progress of welding by a plurality of welding robots, FIG. 4 is an explanatory view showing the progress of welding by a plurality of welding robots, and FIG. 5 is an embodiment of the second and third inventions. FIG. 6 is a flowchart showing a schematic configuration of a method of determining welding order and sharing of a plurality of welding robots according to a second embodiment, FIG. 6 is a detailed flowchart of a method of determining welding sharing, and FIG. It is explanatory drawing which shows the change of welding allotment. In the method for determining a work order of a plurality of robots according to the first invention, a first operation simulation is performed on a plurality of robots having a common moving space based on a certain work order, and the first operation simulation is performed. By extracting the resulting total work time and the work with the longest waiting time of the robot with the longest working time, the work order of the longest waiting time of the robot with the longest working time is changed. ,
The operation sequence is updated, a second simulation is performed based on the updated operation sequence, and the results of the first and second operation simulations are compared to obtain a desired total operation time. Alternatively, while updating the certain work order until a predetermined number of repetitions is reached,
The operation sequence of each robot is determined by repeatedly performing the operation simulation. FIG. 2 is a conceptual diagram of a welding system using a plurality of industrial robots applicable to the first invention. In this case, two robots are provided side by side in the portal type robot mounting device, and one welding target is welded by a total of ten robots. Each robot can move in the X, Y, and Z directions.

Hereinafter, an application example (first embodiment) of such a plurality of welding robot systems will be described. However, the present invention can be similarly applied to a system including a plurality of other industrial robots such as assembly and painting. As shown in FIG. 1, in the first embodiment, first, the welding assignment and the welding order of each robot are initially set (S1). The initial welding share and welding order may be set by a human or may be automatically set. As a counter for limiting the number of times of changing the welding order, a non-update counter and an increment counter are set, and each is cleared to 0 (S2). The initial welding time and the total welding time in the welding order are set to the minimum time and the previous welding time (S3). In the first flow, the welding order is not changed (S4). Next, a welding operation simulation (corresponding to a first operation simulation in this case) is performed for the initial welding assignment and the welding order (S5). This welding operation simulation is performed, for example, as follows. As shown in FIG. 3, at the start of welding, a region surrounding the welding start point and the end point is set as an interference region. This interference area is changed with the simulation time. That is, the interference region is defined as the region from the current position of the robot to the welding end point. When the robot moves at a preset speed, the interference region is gradually released. When moving from the welding end point to the next welding start point,
The area from the robot's current position to the next welding start point is defined as an interference area. Then, when the interference area overlaps between the robots, the robot which has created the interference area later in time is made to wait until the interference area no longer overlaps as shown in FIG. The above procedure is performed until all the welding lines are welded in accordance with the preset welding order and welding assignment, and the time until the completion of welding by each robot, the waiting time for each welding and movement, and the total welding time (for each robot) The maximum value until welding is completed).

In the first flow, since the total welding time in the initial welding order and welding sharing is the minimum time and the previous value of the welding time, the non-update counter becomes +1 (S6 to S8) and the increment counter also becomes +1 (S9
To S11), and replaced with the previous welding time value (S1).
2). Next, if the non-update counter and the increment counter are equal to or less than the set values N and M (S13), the welding order is changed (S4). That is, here, the welding order of the welding line having the longest waiting time of the robot having the longest welding time (in the case of movement, the welding line of the movement destination) obtained by the simulation is changed. The method of change is to make the applicable welding line 1) the leading welding line in the robot's assigned welding line. 2) Last. 3) + L (determined by a constant or random number). However, if the current order + L> the number of shared welding lines, the current order + L−the number of shared welding lines. Among them, 1
Choose one. With this change, the initial welding order is updated. Next, a second welding operation simulation (corresponding to a second operation simulation in this case) is performed based on the updated welding order (S5). Next, the total welding time is evaluated based on the results of the two welding operation simulations in the preceding and following loops. That is, the minimum time is compared with the total welding time (S6), and if the total welding time is shorter than the minimum time, the total welding time is replaced with the minimum time, and the welding order and welding share at this time are stored. At the same time, the non-update counter is cleared to 0 (S7). If the total welding time is longer than the minimum time, 1 is added to the non-update counter, and the minimum time is not changed (S8).

Next, the previous value of the welding time is compared with the total welding time (S9). If the previous value is smaller, the increment counter is cleared to 0 (S10). If the previous value is larger, 1 is added to the increment counter. (S11). Next, the previous welding time is replaced with the total welding time obtained by the simulation (S12). Then, the non-update counter and the increment counter are compared with the respective thresholds N and M, and if neither of them exceeds the thresholds N and M, the process returns to step S4 to repeat the welding order change (S13). . If either of them exceeds the threshold value N or M, it is determined that the minimum time has been obtained and the welding order at that time is determined as the final value (S14). In this way, the welding order of each robot can be determined so that the total welding time of a plurality of robots is the shortest. The increment counter is a condition set to reduce the number of loops, and is set to N> M. Then, even if the non-update counter is not> N, if the welding time continues to increase, the increase counter> M, and the loop can be exited. Thereby, the welding order of each robot can be quickly determined. However, a threshold value for the total welding time is provided in place of or in addition to the increment counter and the non-update counter, and the loop exits when the total welding time becomes smaller than this threshold value. You may do so.

Next, the second and third inventions will be described. In the method for determining the work sharing of a plurality of robots according to the second invention, a third motion simulation is performed on a plurality of robots having a common moving space based on a certain work sharing, and the third simulation is performed. The total work time obtained as a result and the work with the longest waiting time of the robot with the longest working time are extracted, and the tasks within the operating range of the other robots among the shared tasks of the robot with the longest working time are extracted. Is changed to one of the other robots, thereby updating the certain work sharing, performing a fourth operation simulation based on the updated certain work sharing, and performing the third and fourth operations. Is it possible to obtain the desired total work time by comparing the simulation results,
Alternatively, the third and fourth motion simulations are repeatedly performed while updating the certain work sharing until a predetermined repetition number is reached, thereby determining the work sharing of each robot. In the third invention, the work order of the plurality of robots is determined according to the first invention, and then the work sharing of each robot is determined according to the second invention. These second and third inventions are also applicable to a plurality of welding robot systems similar to the first embodiment, and an application example (second embodiment) will be described below. However, the present invention can be similarly applied to a system including a plurality of other industrial robots such as assembly and painting. As shown in FIG. 5, first, a retry counter indicating the number of allotment changes is set and cleared to 0 (S20).

In the second invention, the initially set welding order and welding sharing are used as they are, but in the third invention, the initial welding order and welding sharing are used and the welding order in the first embodiment is used. The welding order is determined by the determination method (S
21). This initial or determined welding order,
The welding sharing and the total welding time at that time are stored as the welding time minimum welding order sharing (S22), and 1 is added to the retry counter (S23). Next, the assignment of one of the welding lines is changed (S24), and the welding order in this assignment is determined by the method of the first embodiment. As a result, if the total welding time is shorter than the stored total welding time, the memory is updated with the determined welding order and sharing and the total welding time at that time as the welding time minimum welding order sharing. Then, add 1 to the retry counter. If the retry counter exceeds the preset threshold value L (S25), the welding order and the welding share stored at this time are determined as the time minimum welding order sharing (S26). Here, a method of changing the welding allocation will be described with reference to the detailed flowchart in FIG. 6 and the example of the operation range of the robot in FIG. First, a robot having the longest welding time is extracted from the result of the minimum total welding time obtained in the welding order determination processing (S31). In the example of FIG. 7, the robot 3 is extracted. Next, a robot sharing an operation range with the robot obtained in step S31 is extracted (S32).
In this example, robots 1, 2, 4, 5, and 6 are extracted. Further, from the robots extracted in step S32, a robot that is currently assigned to the robot extracted in step S31 but has a welding line that can be assigned in step S32 is extracted (S33). In this example,
Robots 1, 2, 4, and 5 are extracted. Next, one robot having the shortest welding time is selected from the robots extracted in step S33 (S34). In this example, the robot 4 is extracted.

Further, a welding line currently assigned to the robot extracted in step S31 is extracted from the welding lines included in the shared operation range of the robot extracted in step S31 and the robot selected in step S34. (S
35). Next, the distance between the center position (Xi, Yi) of the welding line extracted in step S35 and each center position (x1, y1), (x2, y2) of the operation range of the two robots is represented by D1j (step S1). The distance between the robot and the welding line extracted in S31) and D2j (the distance between the robot and the welding line extracted in step S34) are obtained (S3).
6). D1j = √ ((Xi−x1) ² + (Yi−y1) ² ) D2j = √ ((Xi−x2) ² + (Yi−y2) ² ) where i is the robot number and j is the welding line number. Show. Finally, the welding line having the largest distance ratio D1j / D2j is selected, and based on the sharing of the robot extracted in step S31,
It changes to the sharing of the robot extracted in step S34 (S37). In this example, the welding line 1 is assigned to the robot 4. However, the selection criterion of the welding line may be the one with the largest distance D1j or the smallest with the distance D2j. In this way, it is possible to automatically and quickly determine the welding order and the welding share of each robot that greatly reduces the total welding time. It should be noted that a threshold value of the total welding time may be provided instead of or in addition to the retry counter, and the loop may be exited when the total welding time becomes smaller than this threshold value. . In each of the first and second embodiments, the simulation was performed by obtaining an interference area on a two-dimensional plane. However, in actual use, the interference was determined using a three-dimensional robot position and posture. There is no hindrance in seeking the area.

[0015]

According to the method for determining the work order and the work sharing of a plurality of robots according to the present invention, the work order and the work sharing of each robot can be reduced in a short time since the work time is extremely reduced. And automatically.

[Brief description of the drawings]

FIG. 1 is a flowchart showing a schematic configuration of a method for determining a welding order of a plurality of welding robots according to an embodiment (first embodiment) of the first invention.

FIG. 2 is a conceptual diagram of a plurality of welding robots.

FIG. 3 is an explanatory diagram showing the progress of welding by a plurality of welding robots.

FIG. 4 is an explanatory diagram showing the progress of welding by a plurality of welding robots.

FIG. 5 is a flowchart showing a schematic configuration of a method for determining welding order and sharing of a plurality of welding robots according to an embodiment (second embodiment) of the second and third inventions.

FIG. 6 is a detailed flowchart of a method of determining welding sharing.

FIG. 7 is an explanatory diagram showing a change in welding allocation of a plurality of welding robots.

[Explanation of symbols]

S4: welding order changing step (corresponding to the work order updating step) S5: simulation step (corresponding to the first and second operation simulation steps) S6 to 14: total welding time evaluation step (corresponding to the work order determining step) )

──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Kiyoshi Hashimoto 1-2 Nakahara, Miya-machi, Toyohashi-shi, Aichi Prefecture, Japan Kobe Steel, Ltd. Toyohashi FA Robot Center (56) References JP-A-6-131021 ( JP, A) JP-A-62-73310 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) B25J 9/22

Claims (4)

(57) [Claims] 1. A first operation simulation is performed on a plurality of robots having a common moving space based on a certain operation order, and a total operation time obtained as a result of the first operation simulation is determined to be the maximum operation time. The above-mentioned work order is updated by extracting the longest-waiting work of the longest robot and changing the work order of the longest-waiting work of the longest robot,
A second operation simulation is performed based on the updated operation order, and the results of the first and second operation simulations are compared to obtain a desired total operation time or a predetermined number of repetitions. A method for determining the work order of a plurality of robots, wherein the work order of each robot is determined by repeatedly performing the first and second motion simulations while updating the certain work order up to the above. 2. A third operation simulation is performed on a plurality of robots having a common movement space based on a certain work assignment, and the total operation time obtained as a result of the third operation simulation is most significantly reduced. The task with the longest waiting time of the robot with the longest working time is extracted, and one of the tasks within the operating range of the other robot among the tasks of the robot with the longest working time is assigned to the other robot. The above-mentioned certain work sharing is updated by changing, and a fourth operation simulation is performed based on the updated certain work sharing, and the results of the third and fourth operation simulations are compared to obtain a desired total work. Repeating the third and fourth operation simulations while updating a certain task sharing until time is obtained or a predetermined number of repetitions is reached. Is a method of determining the work sharing of multiple robots by determining the work sharing of each robot. 3. A first operation simulation is performed on a plurality of robots having a common moving space based on a certain operation order, and the total operation time obtained as a result of the first operation simulation is determined by the maximum operation time. The above-mentioned work order is updated by extracting the longest-waiting work of the longest robot and changing the work order of the longest-waiting work of the longest robot,
A second operation simulation is performed based on the updated operation order, and the results of the first and second operation simulations are compared to obtain a desired total operation time or a predetermined number of repetitions. By repeating the first and second motion simulations while updating the certain work order up to the above, the work order of each robot is determined, and based on a certain work assignment to the plurality of robots. A third operation simulation is performed, and the total operation time obtained as a result of the third operation simulation and the operation with the longest waiting time of the robot with the longest operation time are extracted, and the robot with the longest operation time is extracted. By changing one of the tasks within the operation range of the other robot to the task of the other robot, the task sharing described above is performed. A fourth operation simulation is performed based on the new and updated work allocation, and the results of the third and fourth operation simulations are compared to obtain a desired total operation time or a predetermined operation time. By repeating the third and fourth motion simulations while updating the certain work sharing until the number of repetitions is reached, the work order and work sharing of the plurality of robots are determined by determining the work sharing of each robot. Method. 4. The method according to claim 3, wherein the work order is a welding order, and the work share is a welding work share.