The embodiment discussed herein is directed to a robot system and an article transfer method.
Japanese Patent Application No. 3834088 discloses a technique in which articles on a conveyance path are distributed to each robot in advance as a transfer target on the basis of the order of the articles conveyed on the conveyance path and each robot transfers the distributed articles from the conveyance path.
A robot system according to an aspect of embodiments includes an information acquisition unit, a plurality of robots, and a distribution unit. The information acquisition unit acquires information indicating a two-dimensional position on a surface of a conveyance path of an article conveyed on the conveyance path. The plurality of robots transfers the article from the conveyance path. The distribution unit distributes the articles on the surface of the conveyance path to the plurality of robots as transfer targets on the basis of the information acquired by the information acquisition unit.
A more complete appreciation of the invention and many of the attendant advantages thereof will be readily obtained as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings, wherein:
Hereinafter, embodiments of the robot system and the article transfer method disclosed by the present application will be described in detail with reference to the accompanying drawings. The present invention is not limited by the embodiments described below.
Here, the conveyance path 2 according to the present embodiment is a conveyor device such as a belt conveyor and a roller conveyor. The conveyor device includes a servo motor 2B that operates the conveyor device and an encoder 2C that detects a rotational position (that is, a conveyance position of the conveyor device) of the servo motor 2B. The servo motor 2B and the encoder 2C are connected to a control device (a robot controller 5 in the present embodiment).
The servo motor 2B operates according to an instruction of the robot controller 5. Thereby, the conveyance path 2 sequentially conveys a plurality of articles W from an upstream side to a downstream side in a conveyance direction indicated by void arrows in
The conveyance path 2 may convey the articles W at a constant conveyance speed set in advance. The conveyance path 2 is not limited to a conveyor, but may be a shooter or the like which is formed so that a conveyance surface forms a downslope from the upstream side to the downstream side and conveys the articles W from the upstream side to the downstream side.
The robot system 1 includes a plurality of robots 3-1 and 3-2 that take out the article W from the conveyance path 2 and transfer the article W to the place 30. Although, in
In the description below, when a specific robot is indicated from among the n robots included in the robot system 1, the robots are referred to as a first robot 3-1, a second robot 3-2, . . . , and nth robot 3-n in an order from the robot of which takeout position of the article W on the conveyance path 2 is the most upstream. When a specific robot is not indicated, each robot is referred to as simply a robot.
The robot system 1 further includes a two-dimensional position information acquisition unit 4 that acquires information indicating a two-dimensional position on the surface of the conveyance path 2 of the article W conveyed on the conveyance path 2 and the robot controller 5 that controls operations of the conveyance path 2 and the first robot 3-1 to the nth robot 3-n.
Here, the two-dimensional position information acquisition unit 4 is, for example, an image capturing device. The two-dimensional position information acquisition unit 4 captures an image of the article W conveyed to an image capturing area 20 defined on the immediately upstream side of a first takeout area 2-1 on the conveyance path 2 from which the first robot 3-1 takes out the article W, and outputs image information of the captured image to the robot controller 5.
The first robot 3-1 includes a single-arm serial link robot arm 3-1A and an end effector 3-1B provided at the tip of the robot arm 3-1A. Here, the end effector 3-1B includes, for example, a sucker which adsorbs and holds an article by applying negative pressure by a vacuum pump and releases the article by releasing the negative pressure. The end effector 3-1B is not limited to a device that adsorbs and holds an article as long as the end effector 3-1B has a mechanism that can hold and convey an article.
The first robot 3-1 takes out the article W from the first takeout area 2-1 of the conveyance path 2 according to control of the robot controller 5 and transfers the article W to a place 30 provided at a predetermined position.
The second robot 3-2 to the nth robot 3-n (see
Although
The robot controller 5 is a control device that controls operations of the conveyance path 2 and the first robot 3-1 to the nth robot 3-n (see
Then, the robot controller 5 distributes the articles W on the surface of the conveyance path 2 to each robot as transfer targets so that the transfer load of the articles W of each robot becomes even on the basis of the detected two-dimensional positions of the articles W in the image capturing area 20 and the rotational position of the servo motor B inputted from the encoder 2C of the conveyance path 2.
Thereby, in the robot system 1, it is possible to prevent the workload of each robot from becoming uneven. A specific example of the method for the robot controller 5 to distribute the articles W will be described later with reference to
Next, a configuration of the robot system 1 will be described with reference to
The robot controller 5 includes a distribution unit 51 and a storage unit 52. Although omitted here in
The storage unit 52 stores predetermined evaluation function information 53. The evaluation function information 53 is a function which calculates the workload of the robot as a score when the robot transfers the article W by substituting the two-dimensional position of the article W in the image capturing area 20 in the function. In the present embodiment, the following Formula (1) is used as the evaluation function information 53.
R=A|X|+BY Formula (1)
Here, R in Formula (1) is a score indicating the workload of the robot, and X and Y are variables where the X coordinate and the Y coordinate of the article W in an XY orthogonal coordinate system defined in the image capturing area 20 (see
A and B are predetermined coefficients. Specifically, A is a coefficient for converting a moving distance of the end effector 3-1B moved by the robot in a direction in parallel with the X axis in the XY orthogonal coordinate system into the workload of the robot.
B is a coefficient for converting a moving distance of the end effector 3-1B moved by the robot in a direction in parallel with the Y axis in the XY orthogonal coordinate system into the workload of the robot. In Formula (1), the larger the distance from the robot to the position of the article W on the conveyance path 2 when the article W is transferred by the robot is, the higher the calculated score becomes.
The distribution unit 51 is a processing unit that distributes the articles W to each robot to cause the robot to transfer the articles W. The distribution unit 51 detects the two-dimensional position of the article W in the image capturing area 200 (see
Subsequently, the distribution unit 51 reads the evaluation function information 53 from the storage unit 52 and substitutes the detected two-dimensional position of each article W in the aforementioned Formula (1), which is the evaluation function information 53, to calculate the score indicating the workload of the robot when the robot transfers the article W for each article W.
Then, the distribution unit 51 determines a robot to be a distribution destination of each article W on the conveyance path 2 so that the sum of the scores of the articles W distributed to the robot as transfer targets is even for each robot.
Thereafter, the distribution unit 51 sequentially calculates the position of each article W on the conveyance path 2 (hereinafter referred to as an “article position”) on the basis of the two-dimensional position of each article W in the image capturing area 20 and the rotational position of the servo motor B inputted from the encoder 2C of the conveyance path 2.
When each article W is conveyed to the takeout area of the robot determined to be a distribution destination, the distribution unit 51 distributes the article W to the robot as a transfer target by providing the article position of the article W conveyed to the takeout area to the robot that is the distribution destination.
For example, when an article W of which the distribution destination is determined to be the first robot 3-1 is transferred to the first takeout area 2-1 (see
Next, a specific example of the method for the distribution unit 51 to distribute the articles W will be described with reference to
Here, a case will be described in which, when six articles W1 to W6 are conveyed to the image capturing area 20, the distribution unit 51 distributes the articles W1 to W6 to the first robot 3-1 and the second robot 3-2 as transfer targets.
As illustrated in
Specifically, when the captured image is assumed to be captured in the first takeout area 2-1, the position closest to the first robot 3-1 in the image capturing area 20 is defined as the origin O, and when the captured image is assumed to be captured in the second takeout area 2-2, the position closest to the second robot 3-2 in the image capturing area 20 is defined as the origin O.
Further, the distribution unit 51 defines the XY orthogonal coordinate system by defining a direction which passes through the origin O (0, 0) and which is in parallel with the conveyance direction of the articles W1 to W6 conveyed by the conveyance path 2 as the X axis and defining a direction which passes through the origin O (0, 0) and which crosses the image capturing area 20 in a direction perpendicular to the X axis as the Y axis.
Subsequently, the distribution unit 51 calculates the X coordinate and the Y coordinate of the articles W1 to W6 in the XY orthogonal coordinate system, respectively. Then, the distribution unit 51 calculates the score of each of the articles W1 to W6 by substituting the calculated X coordinate and Y coordinate of the articles W1 to W6 in the aforementioned Formula (1).
As a result, for example, as illustrated in
In this case, the distribution unit 51 determines the first robot 3-1 as the distribution destination of the article W1, the article W3, and the article W6 as indicated by the solid line arrows in
Thereby, the total score of the scores of the article W1, the article W3, and the article W6 which are distributed to the first robot 3-1 as the transfer targets is “10”, which is the same as the total score of the scores of the article W2, the article W4, and the article W5 which are distributed to the second robot 3-2 as the transfer targets.
In this way, in the robot system 1, the distribution destinations of the articles W are determined so that the total score of the articles W distributed to each robot is even for each robot, that is, the workload of each robot when the robot transfers the articles W is even. Therefore, according to the robot system 1, it is possible to prevent the workload of each robot from becoming uneven.
Next, a process performed by the distribution unit 51 will be described with reference to
Specifically, as illustrated in
Then, the distribution unit 51 determines robots to be distribution destinations so that the total score of the articles W distributed to one robot as transfer targets is even for each robot (step S103). Thereafter, the distribution unit 51 provides the article positions on the conveyance path 2 of the articles W, which are transfer targets, to the determined distribution destination robots (step S104) and ends the process.
As described above, in the robot system according to the first embodiment, the articles on the conveyance path surface are distributed to a plurality of robots as transfer targets on the basis of the information indicating the two-dimensional positions of the articles on the conveyance path surface. Thereby, the robot system according to the first embodiment can prevent the workload of each robot from becoming uneven.
In the present embodiment, the score of each article W is calculated by substituting the two-dimensional position of the article W in the image capturing area 20 into Formula (1). However, the formula used to calculate the score is not limited to Formula (1).
For example, the score of each article W may be calculated by using the following Formula (2):
R=A|X|+BY+Cα Formula (2)
Here, Formula (2) is a formula in which a term Cα for converting a factor to be a workload of a robot when the robot transfers the article W other than the two-dimensional position of the article W in the image capturing area 20 into a score is added to the right side of Formula (1).
The distribution unit 51 calculates the score of each article W by using Formula (2), so that the distribution unit 51 can distribute the articles W on the conveyance path 2 to each robot so that the workload of each robot is more accurately even.
Specifically, when an article W is transferred from the conveyance path 2 by the robot, the workload of the robot when the robot transfers the article W varies due to a difference of type of the article W on the conveyance path 2. For example, regarding a relatively heavy article W and a relatively light article W, the workload for transferring the former is greater than that for transferring the latter.
Further, regarding an article W that can be transferred relatively roughly and an article W that needs to be transferred relatively carefully, the time required to transfer the latter is longer than that required to transfer the former, so that the workload for transferring the latter is greater than that for transferring the former. The workload of the robot when the robot transfers the article W varies due to, for example, a difference between face up and face down of the conveyed article W and an orientation (angle) of the conveyed article W in addition to the type of the article W.
When there is a factor to be a workload of the robot when the robot transfers the article W other than the two-dimensional position of the article W in the image capturing area 20, the distribution unit 51 calculates the score of each article W by using the aforementioned Formula (2).
In this case, the distribution unit 51 identifies the type, orientation, face-up/face-down, and the like of the article W in the captured image by analyzing the image of the image information inputted from the two-dimensional position information acquisition unit 4, calculates a value indicating the workload of the robot according to the identification result, and substitutes the value into Cα in Formula (2). Further, the distribution unit 51 substitutes the two-dimensional position of the article W, of which type, orientation, face-up/face-down, and the like are identified, in the image capturing area 20 into the aforementioned Formula (2) and calculates the score of each article W.
Thereby, the distribution unit 51 can distribute the articles W on the conveyance path 2 to each robot so that the workload of each robot is more accurately even on the basis of the score of each article W calculated by using Formula (2).
Next, a robot system 1a according to a second embodiment will be described with reference to
As illustrated in
The transfer result acquisition unit 41 is a processing unit that counts the number of articles W that are not transferred from the conveyance path 2 by any of the first robot 3-1 to the nth robot 3-n, for example, the number of articles W which the robots failed to transfer, and outputs the counted number of articles W to the coefficient correction unit 54 as a transfer result.
For example, the transfer result acquisition unit 41 counts the number of articles W that were not transferred by any of the robots by counting the number of articles W that pass through the takeout area on the conveyance path 2 from which the nth robot 3-n takes out the article W without being transferred by the nth robot 3-n.
The counting method of the number of articles W performed by the transfer result acquisition unit 41 is not limited to this. For example, the transfer result acquisition unit 41 may be configured to count the number of articles W that were not transferred by any of the robots by subtracting the sum of the numbers of articles W that are transferred to each place 30 from the total number of articles W that are conveyed by the conveyance path 2.
The coefficient correction unit 54 is a processing unit that corrects the coefficients A and B in Formula (1) used to calculate the score of each article W by a distribution unit 51a. In the coefficient correction unit 54, the coefficients A and B in Formula (1) are inputted from the distribution unit 51a every predetermined time (for example, every hour) and a transfer result is inputted from the transfer result acquisition unit 41.
Then, the coefficient correction unit 54 associates the transfer result inputted from the transfer result acquisition unit 41 with the coefficients A and B inputted from the distribution unit 51a and stores the transfer result and the coefficients A and B in the storage unit 52a as the transfer result history information 53a.
Further, when the transfer result is inputted from the transfer result acquisition unit 41 and the coefficients A and B are inputted from the distribution unit 51a, the coefficient correction unit 54 compares the transfer result and the coefficients A and B which are inputted at this time with the transfer result history information 53a.
Then, the coefficient correction unit 54 corrects the coefficients A and B in Formula (1) of the evaluation function information 53 so that the transfer result to be inputted next time is improved, that is, the number of articles W that were not transferred by any of the robots decreases, on the basis of the comparison result.
As described above, in the robot system according to the second embodiment, the coefficient correction unit corrects the coefficients in the evaluation function by feedback control so that the next transfer result is improved on the basis of a history of the transfer result at this time and the transfer result in the past.
Therefore, according to the robot system of the second embodiment, it is possible to prevent the workload of each robot from becoming uneven as well as to reduce the number of articles that are not transferred from the conveyance path by the robots.
Next, a robot system 1b according to a third embodiment will be described with reference to
As illustrated in
The distribution unit of the robot controller 5b according to the present embodiment distributes the articles W on the surface of the conveyance path 2 to a plurality of robots as transfer targets on the basis of the relative position between each robot and the conveyance path 2. More specifically, the distribution unit distributes the articles W to any of a plurality of robots as transfer targets on the basis of the relative position between a robot and the two-dimensional position of the article W on the conveyance path 2 when the article W is conveyed to a position from which the article W is transferred by the robot.
For example, it is assumed that the distance from the article W of which image is captured by the two-dimensional position information acquisition unit 4 and which is conveyed to the first takeout area 2-1 to the first robot 3-1 is L1, and the distance from the article W of which image is captured by the two-dimensional position information acquisition unit 4 and which is conveyed to the second takeout area 2-2 to the second robot 3-2 is L2.
Here, it is assumed that the distances from each place 30 (see
In this case, the workloads from when the first robot 3-1 and the second robot 3-2 hold the article W on the conveyance path 2 to when the first robot 3-1 and the second robot 3-2 transfer the article W to the places 30 are the same. However, the workloads to when the first robot 3-1 and the second robot 3-2 hold the article W on the conveyance path 2 are not the same.
Therefore, when the distribution unit calculates the score of each article W, the distribution unit calculates the score of one article W for each robot by using a different formula of evaluation function according to the difference of the distance from each robot to the conveyance path 2.
Then, the distribution unit determines robots to be distribution destinations so that the total score of the articles W distributed to one robot as transfer targets is even for each robot, and provides article positions on the conveyance path 2 of the articles W, which are transfer targets, to the determined distribution destination robots.
For example, when calculating the score of the article W of which image is captured in the image capturing area 20 illustrated in
On the other hand, the distribution unit calculates the score when the second robot 3-2 transfers the article W by using the following Formula (3):
R=A|X|+B(Y+(L2−L1)) Formula (3)
The distribution unit also calculates the score when another robot transfers the article W by using a formula of a different evaluation function according to the difference of the distance from the robot to the conveyance path 2. Thereby, the distribution unit of the robot system 1b can calculate the score accurately reflecting the workload when each robot transfers the article W.
Then, the distribution unit calculates the score for each robot for each of a plurality of articles W conveyed to the image capturing area 20 and distributes the articles W, which are transfer targets, to each robot so that the total score of the articles W distributed to one robot as transfer targets is even for each robot.
As described above, the distribution unit of the robot system according to the third embodiment distributes the articles to any of a plurality of robots as transfer targets on the basis of the relative position between each robot and the two-dimensional position of the article on the conveyance path when the article is conveyed to a position from which the article is transferred by each robot.
Thereby, according to the robot system of the third embodiment, even when the distance from the conveyance path to each robot is different from each other, it is possible to prevent the workload of each robot from becoming uneven.
Here, a case in which the robots are arranged on the same side of the conveyance path is described. However, a plurality of robots may be arranged on both sides of the conveyance path. In this case, the distribution unit calculates the score of one article for each robot by using a different formula of evaluation function according to a positional relationship between each robot and the conveyance path.
Then, the distribution unit distributes the articles, which are transfer targets, to each robot so that the total score of the articles distributed to one robot as transfer targets is even for each robot. Thereby, even when a plurality of robots is arranged on both sides of the conveyance path, it is possible to prevent the workload of each robot from becoming uneven.
Next, a robot system 1c according to a fourth embodiment will be described with reference to
The robot system 1c illustrated in
The first robot controller 5-1 to the nth robot controller 5-n function as instruction units that issue various operation instructions to the first robot 3-1 to the nth robot 3-n, respectively.
Further, the robot system 1c includes an integrated controller 50 connected to the first robot controller 5-1 to the nth robot controller 5-n and the two-dimensional position information acquisition unit 4. The integrated controller 50 has the same function as that of the distribution unit 51 illustrated in
In other words, the integrated controller 50 includes a distribution unit 51c and the storage unit 52. In the same manner as the distribution unit 51 illustrated in
Then, the distribution unit 51c determines robots to be distribution destinations so that the total score of the articles W distributed to one robot as transfer targets is even for each robot, and provides article positions on the conveyance path 2 of the articles W, which are transfer targets, to the determined distribution destination robots. Thereby, in the robot system 1c, it is possible to prevent the workload of each robot from becoming uneven.
Further, in the robot system 1c, it is possible to decentralize the function of the robot controller 5 illustrated in
Therefore, according to the robot system 1c, it is possible to use an inexpensive controller having a processing capacity lower than that of the robot controller 5 illustrated in
Further, according to the robot system 1c, it is possible to add a robot by only adding the robot and a robot controller that issues operation instructions to the robot without largely changing the control content of the integrated controller 50. Therefore, according to the robot system 1c, it is possible to easily add robots and further improve transfer efficiency of the articles W.
As described above, according to the fourth embodiment, robots can be easily added, so that it is possible to achieve a robot system that can reduce articles that are not transferred by any of robots at lower cost.
In the same manner, also in the robot systems 1a and 1b illustrated in
Additional advantages and modifications will readily occur to those skilled in the art. Therefore, the invention in its broader aspects is not limited to the specific details and representative embodiments shown and described herein. Accordingly, various modifications may be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalents.
This application is a continuation of International Application No. PCT/JP2012/068457, filed on Jul. 20, 2012, the entire contents of which are incorporated herein by reference.
Number | Date | Country | |
---|---|---|---|
Parent | PCT/JP2012/068457 | Jul 2012 | US |
Child | 14598233 | US |