JP2018020423A - Robot system and picking method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To take an article in storage means in and out without contacting with an obstacle.SOLUTION: A robot system includes a robot which takes an article out from storage means, and an operation part which is arranged outside or inside the robot and controls the robot. The robot is provided with a gripper which can hold the article, an arm which can change a position posture of the gripper, and a travel part which can change a position posture of the robot. The operation part calculates a stop position posture of the robot and a target joint angle of the arm based on a shape of the storage means, a shape of the robot, a shape of an object article to be taken out, and a position posture of the object article in the storage means. The robot moves to the stop position posture, changes a joint angle of the arm to the target joint angle, and takes the object article out from the storage means.SELECTED DRAWING: Figure 5

Description

  The present invention relates to a robot system for taking in and out articles from storage means (shelf, container, etc.).

  In recent years, in order to flexibly respond to diversifying consumer needs, there are increasing numbers of warehouses that handle many items such as mail order warehouses and factories that handle many items such as high-mix low-volume production factories. Such a picking operation in a warehouse or a factory is an operation for taking out an ordered product or a part necessary for production of a product from a shelf or a container. Conventionally, picking work has been performed manually. However, picking work by a robot is necessary from the viewpoint of labor shortage and productivity improvement.

  There are the following prior arts as background art of this technical field. Patent Document 1 (Japanese Patent Laid-Open No. 2015-215651) discloses a robot equipped with an arm, which moves to a shelf position and takes out a product. Patent Document 2 (Japanese Patent Publication No. 2015-535787) discloses a system in which a transport vehicle transports a shelf to a human work area.

JP2015-215651A Special table 2015-535787 gazette

  The robot shown in Patent Document 1 moves to the front of the shelf where the corresponding article is stored, and moves the arm so as to grip the article at the arrival point. In such a procedure, reach reaches a high range. A long arm needs to be mounted, and a lot of freedom is needed to operate the long reach so that the long arm does not touch the shelf or other obstacles, making the system expensive.

  Moreover, in the system shown in Patent Document 2, although the operation of the person who goes around the shelf is automated, the operation of taking out the product from the shelf requires a manual operation.

  It is an object of the present invention to provide a robot system that can load and unload articles in a storage means without contacting an obstacle even with a robot having a short reach and a low degree of joint freedom.

  A typical example of the invention disclosed in the present application is as follows. That is, the robot system includes a robot that takes out an article from a storage unit, and a calculation unit that is provided outside or inside the robot and controls the robot. The robot includes a gripper that can hold the article. An arm that can change the position and orientation of the gripper and a traveling unit that can change the position and orientation of the robot, and the computing unit is configured to extract the shape of the storage means, the shape of the robot, and the object to be extracted. Based on the shape of the article and the position and orientation of the target article in the storage means, the stop position and orientation of the robot and the target joint angle of the arm are calculated, and the robot moves to the stop position and orientation, The joint angle of the arm is changed to the target joint angle, and the target article is taken out from the storage means.

  According to one aspect of the present invention, the article in the storage means can be taken in and out without contacting an obstacle. Problems, configurations, and effects other than those described above will become apparent from the description of the following embodiments.

It is a figure which shows the structure of the robot system of Example 1. FIG. It is a figure which shows the robot system of Example 1 which picks the articles | goods in a storage shelf. It is a figure which shows the robot system of Example 1 which picks the articles | goods in a container. 3 is a flowchart of a process for controlling the operation of the robot according to the first embodiment. It is a figure which shows the stop position attitude | position of the robot of Example 1. FIG. It is a figure which shows the structure of the robot system of Example 2. FIG. It is a figure which shows the stop position attitude | position of the robot of Example 2. FIG. 10 is a flowchart of processing for controlling an operation until the robot according to the second embodiment holds an article. It is a figure which shows operation | movement until the robot of Example 2 hold | maintains articles | goods. It is a figure which shows the operation | movement which the robot of Example 2 drive | works along an interference avoidance movement locus. 6 is a flowchart of processing for controlling an operation after the robot of the second embodiment holds an article. It is a figure which shows the operation | movement after the robot of Example 2 hold | maintains articles | goods. It is a figure which shows the structure of the robot system of Example 3. FIG. It is a figure which shows the robot system of Example 3 which picks the articles | goods in a storage shelf. It is a figure which shows the robot system of Example 3 which picks the articles | goods in a container. 10 is a flowchart of a process for controlling the operation of the robot according to the third embodiment. It is a figure which shows the stop position attitude | position of the storage shelf of Example 3. FIG. FIG. 10 is a diagram illustrating a configuration of a robot system according to a fourth embodiment. It is a figure which shows the stop position attitude | position of the storage shelf of Example 4. FIG. 14 is a flowchart of processing for controlling an operation until the robot according to the fourth embodiment holds an article. It is a figure which shows operation | movement until the robot of Example 4 hold | maintains articles | goods. It is a figure which shows the operation | movement which the robot of Example 4 drive | works along an interference avoidance movement locus. 12 is a flowchart of processing for controlling an operation after the robot according to the fourth embodiment holds an article. It is a figure which shows the operation | movement after the robot of Example 4 hold | maintains articles | goods.

<Example 1>
FIG. 1 is a diagram illustrating a configuration of the robot system according to the first embodiment.

  The robot system according to the first embodiment includes a robot 10a and an overall management apparatus 200.

  The robot 10a includes an arm 12 having a gripper 11 at the tip, and an article measurement sensor 18a that recognizes an article held by the gripper 11. In addition, a traveling unit 13 a, an ambient environment measurement sensor 17 a, a calculation unit 14, a suction force generation unit 15, and a power supply unit 16 are provided in the base unit of the robot 10 a.

  The arm 12 extends in the lateral direction from a column standing on the base portion, and holds the article with the gripper 11 provided at the tip. Further, the arm 12 has a plurality of joints, the arm 12 is bent at the joints, the arm 12 is retracted and extended, and the position of the gripper 11 is moved. The arm 12 of the present embodiment can move the gripper 11 back and forth and up and down at three joints to change the orientation of the gripper 11, but can perform more complicated movement (for example, a 6-axis arm, a lateral direction) Horizontal articulated arm).

  The gripper 11 is provided at the tip of the arm 12, and the arm 12 holds the article by being in close contact with and adsorbing the article. The gripper 11 is not an adsorption gripper, and various types of grippers such as a multi-finger hand gripper having a plurality of fingers, a two-nail single joint gripper, and a two-nail parallel gripper can be employed.

  The article measurement sensor 18 a is provided at the tip of the arm 12 and is configured by a distance image camera for recognizing an article held by the gripper 11. The article measurement sensor 18a may be a gray or color camera (imaging device) or an infrared camera for photographing an article, or a combination of a distance image camera and a camera. In this embodiment, the article measurement sensor 18a has one configuration of the robot 10a, but may be configured separately from the robot 10a. For example, the article measurement sensor 18a may be provided on the storage shelf 100a or fixedly provided at a plurality of locations in a warehouse where the storage shelf 100a is installed.

  The traveling unit 13a provided in the base unit of the robot 10a includes a rotating tire and a motor that rotationally drives the tire. Two tires are provided facing each other, and by controlling the rotation direction and the number of rotations of each tire, the robot 10a can travel, change direction, and rotate on the spot. An auxiliary wheel (not shown) may be provided below the base portion of the robot 10a.

  The ambient environment measurement sensor 17a is constituted by a two-dimensional or three-dimensional laser scanner having a laser light irradiation unit and a laser light receiving unit, and irradiates the surroundings with laser light and measures reflected light. The distance to the surrounding obstacles (walls, storage shelves, etc.) can be calculated by the laser reflected light measured by the surrounding environment measurement sensor 17a, and the calculated distance is connected to determine the position and size of the surrounding obstacles. It can be calculated.

  The calculation unit 14 is configured by a computer having a processor (CPU), a memory, an auxiliary storage device, and a communication interface. In the present embodiment, the calculation unit 14 has one configuration of the robot 10a, but may be provided separately from the robot 10a so as to be communicable with the robot 10a.

  The CPU executes a program stored in the memory. The memory includes a ROM that is a nonvolatile storage device and a RAM that is a volatile storage device. The ROM stores an immutable program (for example, BIOS). The RAM is a high-speed and volatile storage device such as a DRAM (Dynamic Random Access Memory), and temporarily stores a program executed by the CPU and data used when the program is executed.

  The auxiliary storage device is configured by a large-capacity and nonvolatile storage device such as a magnetic storage device (HDD) or a flash memory (SSD), for example, and stores a program executed by the CPU and data used when the program is executed. . That is, the program is read from the auxiliary storage device, loaded into the memory, and executed by the CPU. The auxiliary storage device also stores the shape data of the storage shelf 100, the data of the shape and mechanism (operation range) of the robot 10, the data of the shape and size of the article 101, the position and orientation data, and the article 101 when the article 101 is held. The relative position and orientation data of the gripper 11, the position and orientation data of the robot 10 and the storage shelf 100, the warehouse shape and passage data are stored.

  The communication interface is a network interface device that controls communication with another device (for example, the overall management device 200) according to a predetermined protocol.

  A program executed by the CPU is provided to the computing unit 14 via a removable medium (flash memory, CD-ROM, or the like) or a network, and is stored in a nonvolatile auxiliary storage device that is a non-temporary storage medium. For this reason, the calculating part 14 is good to have an interface (for example, USB port) which reads data from a removable medium. Note that in the calculation unit 14, all or a part of the functional blocks implemented by the program may be configured by a physical integrated circuit (for example, a Field-Programmable Gate Array).

  The suction force generation unit 15 includes a compressor (or pump), generates compressed air (or low-pressure air), and supplies the gripper 11 with the suction force or gripping force of the gripper 11.

  The power supply unit 16 is a battery, and supplies power to each unit of the robot 10a such as the traveling unit 13a, the calculation unit 14, the ambient environment measurement sensor 17a, and the article measurement sensor 18a.

  The overall management apparatus 200 is configured by a computer having a processor (CPU), a memory, an auxiliary storage device, and a communication interface. The overall management apparatus 200 and the robot 10a are connected by wireless communication.

  The CPU executes a program stored in the memory. The memory includes a ROM that is a nonvolatile storage device and a RAM that is a volatile storage device. The ROM stores an immutable program (for example, BIOS). The RAM is a high-speed and volatile storage device such as a DRAM (Dynamic Random Access Memory), and temporarily stores a program executed by the CPU and data used when the program is executed.

  The auxiliary storage device is configured by a large-capacity and nonvolatile storage device such as a magnetic storage device (HDD) or a flash memory (SSD), for example, and stores a program executed by the CPU and data used when the program is executed. . That is, the program is read from the auxiliary storage device, loaded into the memory, and executed by the CPU. The auxiliary storage device also stores the shape data of the storage shelf 100, the data of the shape and mechanism (operation range) of the robot 10, the data of the shape and size of the article 101, the position and orientation data, and the article 101 when the article 101 is held. Data on the relative position and orientation of the gripper 11, data on the position and orientation of the robot 10 and the storage shelf 100, and data on the shape and passage of the warehouse may be stored.

  A program executed by the CPU is provided to the overall management apparatus 200 via a removable medium (CD-ROM, flash memory, etc.) or a network, and is stored in a nonvolatile auxiliary storage device that is a non-temporary storage medium. Therefore, the overall management apparatus 200 may have an interface for reading data from a removable medium.

  The overall management apparatus 200 is a computer system configured on a single computer or a plurality of computers that are logically or physically configured, and operates on separate threads on the same computer. It may be possible to operate on a virtual machine constructed on a plurality of physical computer resources.

  In the overall management apparatus 200, all or a part of the functional blocks implemented by the program may be configured by a physical integrated circuit (for example, Field-Programmable Gate Array).

  Next, picking of an article by the robot 10a will be described with reference to FIGS.

  FIG. 2 is a diagram illustrating the robot system according to the first embodiment that picks articles in a storage shelf.

  The robot system shown in FIG. 2 includes a robot 10a, a storage shelf 100a, and an overall management apparatus 200 (not shown). The robot 10a moves to the front of the article 101a taken out from the storage shelf 100a, adsorbs the gripper 11 to the side surface of the article 101a, and bends the joint of the arm 12, thereby taking out the article 101a from the storage shelf 100a.

  FIG. 3 is a diagram illustrating the robot system according to the first embodiment that picks an article in a container.

  The robot system shown in FIG. 3 includes a robot 10a, a storage container 150a, and an overall management apparatus 200 (not shown). The robot 10a moves to the front of the article 101a taken out from the storage container 150a, adsorbs the gripper 11 to the upper surface of the article 101a, and bends the joint of the arm 12, thereby taking out the article 101a from the storage container 150a.

  In addition, although the storage shelf 100a and the storage container 150a have been described as storage means for storing articles in the first embodiment, any form may be used as long as the articles can be stored and stored.

  FIG. 4 is a flowchart of a process for controlling the operation of the robot 10a according to the first embodiment. The program for performing the process shown in FIG. 4 is executed by the CPU of the calculation unit 14 of the robot 10a.

  First, when the robot 10a receives a command to take out the article 101a from the storage shelf 100a from the overall management apparatus 200 (1001), the robot 10a autonomously reaches a position in front of the storage shelf 100a, that is, a position where the article 101 stored in the storage shelf 100a can be measured. (1002). In this specification, to move autonomously, the calculation unit 14 estimates the position and orientation (self-position) of the robot 10a, and the robot 10a moves along the path from the positional relationship between the planned path and the self-position. This means that the traveling unit 13a is controlled. The self-position is estimated by collating the position and size of surrounding obstacles measured by the surrounding environment measurement sensor 17a with the shape data of the warehouse stored in the auxiliary storage device of the calculation unit 14. The route is divided into a long-distance route and a short-distance route, and a route 1002 corresponds to a long-distance route. The long-distance route is planned by connecting along the passage from the current location of the robot 10a to the destination (here, the position where the article 101 stored in the storage shelf 100a can be measured) using the passage data of the warehouse. Even if the calculation unit 14 of the robot 10a plans, the overall management apparatus 200 may plan and include it in the command to the robot 10a. Further, when entering an area such as an intersection where there is a high possibility of approaching another robot 10, communication with the overall management apparatus 200 may be made to inquire about whether or not to enter the intersection. On the other hand, the short distance route is planned by simply connecting the current location to the destination by a straight line or as an interference avoidance movement locus 904 by the calculation unit 14 of the robot 10a. The interference avoidance movement locus 904 will be described later.

  Then, the robot 10a measures the article 101 stored in the storage shelf 100a by the article measurement sensor 18a (1003). And the calculating part 14 calculates the detailed position and orientation of each article | item 101 in the storage shelf 100a using the shape and magnitude | size of the article | item 101 registered previously with respect to the measurement result by the article | item measurement sensor 18a ( 1004). For example, the article measurement sensor 18a acquires a distance image including the article 101, and searches the distance image for a portion that matches the shape and size of the article 101 registered in advance. A simple position and orientation can be calculated. In addition, when the detailed position and orientation of the article 101 in the storage shelf 100a can be acquired from the overall management apparatus 200, the processing of steps 1003 and 1004 can be omitted.

  Then, the calculation unit 14 calculates a stop position / posture 900a of the robot 10a for holding the article 101a (1005). This calculation is realized, for example, by simulation by the calculation unit 14. First, the absolute position of the gripper 11 for holding the article 101a from the data of the relative position and orientation of the gripper 11 viewed from the article 101a for holding the predetermined article 101a and the position and orientation of the article 101a calculated in 1004. Calculate the position and orientation. Subsequently, in the simulation, the temporary stop position / posture of the robot 10a is set, and the robot 10a calculates the absolute position / posture of the gripper 11 calculated in the temporary stop position / posture using the shape and mechanism data of the robot 10a. Check if it is possible or impossible to take. This confirmation can be realized by solving the inverse kinematics problem. If the result of this confirmation is possible, the temporary stop position / posture is output as a stop position / posture 900a. If the result of the confirmation is impossible, the temporary stop position / posture is slightly changed, and confirmation of whether or not the gripper 11 can realize the absolute position / posture is executed again. This is repeated until a confirmation result is possible. In this embodiment, since the relative position and orientation of the gripper 11 as viewed from the article 101a for holding the article 101a is perpendicular to the center of the front surface of the article 101a, the robot 10a for taking out the article 101a is stopped. The position is a position facing the front of the article 101a (the surface facing the robot 10a).

  Thereafter, the robot 10a autonomously moves to the calculated stop position / posture 900a (1006). The route at this time corresponds to a short-distance route, and the robot 10a turns on the spot toward the position of the stop position / posture 900a at the current location, moves linearly to the position of the stop position / posture 900a, and stops. The robot 10a is finally set to the stop position / posture 900a by turning on the spot at the position / posture 900a to match the posture.

  The angle of the joint of the arm 12 is changed to the target joint angle 901a so that the gripper 11 moves to a position and posture capable of holding the article 101a (1007). The result of solving the inverse kinematics problem in 1005 is the target joint angle 901a of the arm 12 for the robot 10a to take the absolute position and orientation of the gripper 11. For example, there is a method of gradually changing the joint angle from the current joint angle to the target joint angle 901a at a constant angular velocity. As a result, the gripper 11 moves so that the article 101a can be adsorbed to face the front of the article 101a to be taken out.

  Next, the gripper 11 holds the article 101a (1008). For example, the gripper 11 in surface contact with the article 101 a is decompressed, and the article 101 a is adsorbed to the gripper 11.

  Then, the joint angle of the arm 12 is changed to the end joint angle 902 so that the article 101a is pulled out from the storage shelf 100a while holding the article 101a, and the gripper 11 is moved (1009). For example, without changing the posture of the gripper 11 and the posture of the article 101a in the state of 1008, the gripper 11 is moved vertically to the front of the storage shelf 100a to the position where the article 101a is completely separated from the storage shelf 100a. The joint angle of the arm 12 when the article 101a is completely separated from the storage shelf 100a is the end joint angle 902.

  Thereafter, the calculation unit 14 confirms that the joint angle of the arm 12 has been changed to the end joint angle 902, and transmits a work completion report to the overall management apparatus 200 (1010).

  FIG. 5 is a diagram illustrating a stop position / posture 900 of the robot 10a corresponding to the position / posture of the article 101 according to the first embodiment.

  As described above, in the first embodiment, the robot 10a stops at a position where the gripper 11 can hold the article 101a. That is, the position and orientation at which the robot stops depends on the position and orientation of the article 101 to be taken out. For example, when taking out the article 101a in the upper row in the middle row under the storage shelf 100a, the robot 10a stops at the stop position / posture 900a. Further, when the article 101e in the upper left column of the front opening below the storage shelf 100a is taken out, the robot 10a stops at the stop position / posture 900e. Furthermore, when taking out the article 101i in the upper left column of the front opening below the storage shelf 100a, the robot 10a stops at the stop position / posture 900i.

  The stop position / posture 900 and the target joint angle 901 differ depending on the type of the gripper 11, the attachment position / posture with respect to the arm 12, and the shape and size of the article 101. Therefore, the calculation unit 14 calculates the stop position / posture 900 and the target joint angle 901 using information on the direction in which the gripper 11 accesses the article 101.

  So far, in the first embodiment, the case where the article stored in the storage shelf 100a shown in FIG. 2 is taken out has been described, but the same applies to the case where the article 151a stored in the storage container 150a shown in FIG. 3 is taken out. . That is, the robot 10a moves to a position where the gripper 11 faces the upper surface of the article and can attract the article 151a.

  As described above, in the first embodiment, the gripper 11 holds the article 101a after calculating the stop position / posture 900a of the robot 10a for holding the article 101a and moving to the stop position / posture 900a. The article 101a can be accessed from a position where the gripper 11 can correctly contact the article 101a, and mistakes in taking out a different article 101 or missing the article 101a can be reduced.

<Example 2>
In this embodiment, an example of an article picking robot system in the storage shelf that can take out the article 101 placed in the back of the storage shelf 100 will be described.

  FIG. 6 is a diagram illustrating a configuration of the robot system according to the second embodiment.

  The robot system according to the second embodiment includes a robot 10a, a storage shelf 100M, and an overall management apparatus 200. The robot 10a moves to the front of the article 101M taken out from the storage shelf 100M, picks up the gripper 11 on the side surface of the article 101M, and bends the joint of the arm 12, thereby taking out the article 101M from the storage shelf 100M.

  In the second embodiment, an example in which the robot 10a takes out the article 101M placed in the back of the storage shelf 100M will be described. When the article 101M placed behind the storage shelf 100M is taken out, the arm 10 is changed to the target joint angle 901M to attract the article 101M while the robot 10a is approaching the storage shelf 100M. It may abut against a shelf plate that divides the frontage. Further, if the arm 12 is immediately changed to the end joint angle 902 after the robot 10a holds the article 101M, the robot 10a may come into contact with a shelf plate that divides the front of the storage shelf 100M. For this reason, a special procedure is required when taking out the article 101M placed behind the storage shelf 100M.

  FIG. 7 is a diagram illustrating a stop position / posture 900M of the robot 10a for taking out the article 101M placed behind the storage shelf 100M according to the second embodiment.

  As described above, in the second embodiment, the robot 10a stops at a position where the gripper 11 can hold the article 101M. That is, the stop position and posture of the robot vary depending on the article 101 to be taken out. For example, when the article 101M in the upper left column of the front shelf under the storage shelf 100M is taken out, the robot 10a stops at the stop position / posture 900M. The stop position / posture 900M is closer to the storage shelf 100M than the stop position / posture 900e for taking out the foremost article 101 of the storage shelf 100M.

  Next, with reference to FIG. 8 and FIG. 9, an operation in which the robot 10a according to the second embodiment takes out the article 101M placed behind the storage shelf 100M will be described. FIG. 8 is a flowchart of a process for controlling the operation until the robot 10a according to the second embodiment holds the article 101M. FIG. 9 is a diagram illustrating an operation until the robot 10a according to the second embodiment holds the article 101M.

  First, the robot 10a executes steps 1001 to 1004 in FIG. 4 (1101). That is, the robot 10a receives a command to take out the article 101M from the storage shelf 100M from the overall management apparatus 200 (1001), autonomously moves to a position where the article 101 stored in the storage shelf 100M can be measured (1002), The article 101 stored in the storage shelf 100M is measured by the article measurement sensor 18a (1003), and the detailed position and orientation of each article 101 in the storage shelf 100M is calculated (1004). In this state, as shown in FIG. 9A, the robot 10a is located at a position away from the storage shelf 100M overlooking the article 101M to be taken out and the surrounding situation.

  Then, the calculation unit 14 calculates the stop position / posture 900M of the robot 10a for holding the article 101M and the target joint angle 901M of the arm 12 (1102). This calculation is realized, for example, by simulation by the calculation unit 14. First, the absolute position of the gripper 11 for holding the article 101M from the data of the relative position and orientation of the gripper 11 viewed from the article 101M for holding the predetermined article 101M and the position and orientation of the article 101M calculated in 1004. Calculate the position and orientation. Subsequently, in the simulation, a temporary stop position / posture of the robot 10a is set, and it is confirmed whether or not the robot 10a can take the absolute position / posture of the gripper 11 calculated in the temporary stop position / posture. This confirmation can be realized by solving the inverse kinematics problem, and as a result, the temporary target joint angle of the arm 12 can be obtained simultaneously. If the result of this confirmation is possible, the robot 10a further uses the distance data measured at 1003 in 1101 using the shape and mechanism data of the robot 10a, the shape data of the storage rack 100, and the temporary stop position / posture of the robot 10a. When the temporary target joint angle is taken, it is confirmed whether or not the robot 10a comes into contact with the storage shelf 100 or another article. In the case of no contact, the temporary stop position / posture is output as the stop position / posture 900M, and the temporary target joint angle is output as the target joint angle 901M. If the result of the confirmation is not possible or if it is “contact”, the temporary stop position / posture is slightly changed, whether the absolute position / posture of the gripper 11 can be realized, and whether the robot 10a comes into contact. Check again. These are repeated until a confirmation result is possible and “no contact”. In the present embodiment, since the relative position and orientation of the gripper 11 viewed from the article 101M for holding the article 101M is a direction perpendicular to the center of the front surface of the article 101M, the robot 10a for taking out the article 101M is stopped. The position is a position facing the front of the article 101M (the surface facing the robot 10a). Further, a state where the arm 12 is extended so that the gripper 11 reaches the article 101M in the back of the storage shelf 100M is a target joint angle 901M (FIG. 9B).

  Next, when the robot 10a is in the stop position / posture 900M, the calculation unit 14 determines whether the arm 12 can be changed from the current joint angle to the target joint angle 901M (1103). For example, referring to the warehouse in which the storage shelf 100M is installed and the shape data of the storage shelf 100M (eg, the position and size of the walls and pillars of the warehouse, the CAD data representing the shape and position of the storage shelf 100M), In the space where the tip (gripper 11) can move, the arm 12 does not interfere with obstacles (storage rack 100M, other robot 10, warehouse pillars, walls, etc.) and the target joint angle from the current joint angle. A trajectory that can transition to 901M is searched. This search is realized by, for example, simulation by the calculation unit 14. First, in the simulation, for the joint angles of a large number of arms 12, the robot uses the stop position / posture 900 </ b> M, the shape and mechanism data of the robot 10 a, the shape data of the storage shelf 100, and the distance image measured at 1003 in 1101. It is confirmed whether or not 10a is in contact with the storage shelf 100M and other articles, and the joint angles that are not in contact are stored. Next, arbitrarily select two stored joint angles, determine whether or not they are close using a predetermined threshold value, and create a joint angle transition graph connecting the close joint angles. Thereafter, the joint angle transition graph is traced to search for a path that reaches the target joint angle 901M from the current joint angle. If the path can be found within a predetermined time, it is determined that the arm 12 can transition from the current joint angle to the target joint angle 901M. For example, as shown in FIG. 9C, when the robot 10a tries to move the arm 12 to the target joint angle 901M in the stop position / posture 900M, the robot 10a always comes into contact with the upper shelf board within a predetermined time. It is determined that the path cannot be found because the path cannot be found.

  As a result, if the arm 12 cannot shift to the target joint angle 901M, the process proceeds to step 1107. On the other hand, if the arm 12 can transition to the target joint angle 901M, the calculation unit 14 causes the interference avoidance joint angle to transition the arm 12 from the current joint angle to the target joint angle 901M when the robot 10a is in the stop position / posture 900M. A trajectory is determined (1104). Specifically, the shortest path (or the path that satisfies a predetermined condition) that transitions from the current state to the target joint angle 901M is searched using the result of the simulation in step 1103, and determined as an interference avoidance joint angle locus. To do.

  Then, the robot 10a autonomously moves to the stop position / posture 900M (1105), controls the arm 12 according to the interference avoidance joint angle locus, and moves the gripper 11 (1106). A route 1105 corresponds to a short-distance route. Thereafter, the gripper 11 holds the article 101M (1111). For example, the gripper 11 in surface contact with the article 101M is depressurized, and the article 101M is adsorbed to the gripper 11.

  On the other hand, if it is determined in step 1103 that the arm 12 interferes with the obstacle and cannot be shifted to the target joint angle 901M, the calculation unit 14 determines in step 1107 that the arm 12 has the target joint angle 901M. Then, it is determined whether the robot 10a can be moved from the current position / posture to the stop position / posture 900M. This determination is realized by, for example, simulation by the calculation unit 14. First, in the simulation, using the target joint angle 901M, the shape data of the warehouse, the shape and mechanism data of the robot 10a, the shape data of the storage shelf 100, and the distance image measured in step 1003 in step 1101, the storage shelf 100M In the positions and postures of a large number of robots 10a in the vicinity, it is confirmed whether or not the robots 10a are in contact with the storage rack 100 or other articles, and the positions and postures that are not in contact are stored. Next, two arbitrarily stored positions and orientations are selected, it is determined whether or not they are close using a predetermined threshold value, and a movement graph connecting the close positions and orientations is created. Thereafter, the movement graph is traced to search for a route from the current position / posture to the stop position / posture 900M. If the path can be found within a predetermined time, it is determined that the robot 10a can be moved from the current position / posture to the stop position / posture 900M.

  As a result, if the robot 10a cannot move to the stop position / posture 900M, the process returns to Step 1102, and another combination of the stop position / posture 900M and the target joint angle 901M is calculated. At this time, one of the stop position / posture 900M and the target joint angle 901M may be new, or both the stop position / posture 900M and the target joint angle 901M may be new.

  On the other hand, if the robot 10a can move to the stop position / posture 900M without interfering with the obstacle, the calculation unit 14 determines whether the robot 10a is moved from the current position / posture when the arm 12 has the target joint angle 901M. The interference avoidance movement locus 904M that moves to the stop position / posture 900M is determined (1108). Specifically, the shortest path (or the path that satisfies a predetermined condition) that moves from the current position / posture to the stop position / posture 900M is searched using the simulation result in Step 1107, and the interference avoidance movement trajectory 904M is obtained. It decides (Drawing 9 (d)).

  Thereafter, the arm 12 changes the joint angle to the target joint angle 901M (1109), the computing unit 14 controls the traveling unit 13a according to the interference avoidance movement locus 904M, and the robot 10a moves (1110, see FIG. 9 (e)). ). Thereafter, the gripper 11 holds the article 101M (1111, see FIG. 9F).

  As shown in FIG. 10, the robot 10a according to the second embodiment has the joint angle of the arm 12 set to the target joint angle 901M in order to hold the article 101M placed behind the storage shelf 100M (1109). Then, it approaches the storage shelf 100M according to the interference avoidance movement trajectory 904M.

  FIG. 11 is a flowchart of a process for controlling the operation after the robot 10a according to the second embodiment holds the article 101M. That is, the process shown in FIG. 11 is executed after the process shown in FIG. FIG. 12 is a diagram illustrating an operation after the robot 10a according to the second embodiment holds the article 101M.

  First, the calculation unit 14 calculates a stop position / posture 900Q of the robot 10a and a target joint angle 901Q of the arm 12 for pulling out the storage shelf 100M while holding the article 101M (1112). The stop position / posture 900Q and the target joint angle 901Q calculated here are the position / posture at which the robot 10a can move away from the storage shelf 100M and the robot 10a can move to the next operation and the target joint angle of the arm 12 (FIG. 12B). )).

  Next, the calculation unit 14 determines whether or not the arm 12 can be changed from the current joint angle to the target joint angle 901Q when the robot 10a is in the current position and posture (1113). For example, the shape data of the warehouse in which the storage shelf 100M is installed and the storage shelf 100M (for example, CAD data representing the position and size of the walls and pillars of the warehouse and the shape and position of the storage shelf 100M, as in Step 1103). Referring to the trajectory in which the distal end of the arm 12 (gripper 11) can move is searched for a trajectory that allows the arm 12 to transition from the current joint angle to the target joint angle 901Q without interfering with an obstacle. This search is realized by, for example, simulation by the calculation unit 14. First, in the simulation, the current position and orientation, the shape and mechanism data of the robot 10a, the shape data of the storage shelf 100, and the distance image measured in step 1003 in step 1101, It is confirmed whether or not the robot 10a comes into contact with the storage shelf 100 or other articles, and the joint angles that do not come in contact are stored. Next, arbitrarily select two stored joint angles, determine whether or not they are close using a predetermined threshold value, and create a joint angle transition graph connecting the close joint angles. Thereafter, the joint angle transition graph is traced to search for a path that reaches the target joint angle 901Q from the current joint angle. If a path can be found within a predetermined time, it is determined that the arm 12 can transition from the current joint angle to the target joint angle 901Q. For example, as shown in FIG. 12 (c), when the robot 10a tries to move the arm 12 to the target joint angle 901Q in the current position and posture, if the robot 10a always comes into contact with the upper shelf board, the path is determined within a predetermined time. Cannot be found and it is determined that the transition is impossible.

  As a result, if the arm 12 cannot shift to the target joint angle 901Q, the process proceeds to step 1117. On the other hand, if the arm 12 can transition to the target joint angle 901Q, the computing unit 14 causes the interference avoidance joint angle trajectory to transition the arm 12 from the current joint angle to the target joint angle 901Q when the robot 10a is in the current position / posture. Is determined (1114). Specifically, using the result of the simulation in step 1113, the shortest path (or the path satisfying a predetermined condition) that transitions from the current state to the target joint angle is searched and determined as an interference avoidance joint angle trajectory. .

  Then, the robot 10a controls the arm 12 according to the interference avoidance joint angle locus, moves the gripper 11 (1115), and autonomously moves to the stop position / posture 900Q (1116). The route 1116 corresponds to a short-distance route. Thereafter, the operation unit 14 confirms that the robot 10a has moved to the stop position / posture 900Q and the joint angle of the arm 12 has been changed to the target joint angle 901Q, and then transmits a work completion report to the overall management apparatus 200 ( 1121).

  On the other hand, when it is determined in step 1113 that the arm 12 cannot be deformed to the target joint angle 901Q without interfering with the obstacle, the calculation unit 14 determines that the robot 10a has the arm 12 at the current joint angle. Can be moved from the current position / posture to the stop position / posture 900Q (1117). This determination is realized by, for example, simulation by the calculation unit 14. First, in the simulation, storage is performed using the current joint angle of the arm 12, the shape data of the warehouse, the shape and mechanism data of the robot 10a, the shape data of the storage shelf 100, and the distance image measured in step 1003 in step 1101. For the positions and orientations of a large number of robots 10a in the vicinity of the shelf 100M, it is confirmed whether or not the robots 10a are in contact with the storage shelf 100 and other articles, and the positions and orientations that are not in contact are stored. Next, two arbitrarily stored positions and orientations are selected, it is determined whether or not they are close using a predetermined threshold value, and a movement graph connecting the close positions and orientations is created. Thereafter, the movement graph is traced to search for a route from the current position / posture to the stop position / posture 900Q. If a path can be found within a predetermined time, it is determined that the robot 10a can be moved from the current position / posture to the stop position / posture 900Q (FIG. 12D).

  As a result, if the robot 10a cannot move to the stop position / posture 900Q, the process returns to Step 1112 to calculate another combination of the stop position / posture 900Q and the target joint angle 901Q. At this time, one of the stop position / posture 900Q and the target joint angle 901Q may be new, or both the stop position / posture 900Q and the target joint angle 901Q may be new.

  On the other hand, if the robot 10a can move to the stop position / posture 900Q without interfering with the obstacle, the calculation unit 14 determines whether the robot 10a is moved from the current position / posture when the arm 12 has the target joint angle 901Q. The interference avoidance movement trajectory 904Q that moves to the stop position / posture 900Q is determined (1118). Specifically, using the result of the simulation in step 1107, the shortest path (or the path that satisfies a predetermined condition) that moves from the current position / posture to the stop position / posture 900Q is searched, and the interference avoidance movement trajectory 904Q is obtained. decide.

  Thereafter, the computing unit 14 controls the traveling unit 13a according to the interference avoidance movement locus 904Q, the robot 10a moves (1119, FIG. 12 (e)), and the arm 12 changes the joint angle to the target joint angle 901Q (1120, FIG. 12 (f)). Thereafter, the calculation unit 14 confirms that the robot 10a has moved to the stop position / posture 900Q and the joint angle of the arm 12 has been changed to the target joint angle 901Q, and transmits a work completion report to the overall management apparatus 200 (1121). ).

  So far, in the second embodiment, the case where the article stored in the storage shelf 100M shown in FIG. 6 is taken out has been described, but the same applies to the case where the article 151a stored in the storage container 150a shown in FIG. 3 is taken out. . That is, the robot 10a moves to a position where the gripper 11 faces the upper surface of the article and can attract the article 151a.

  As described above, in the second embodiment, the stop position / posture 900M and the target joint angle 901M of the robot 10a for holding the article 101M are calculated, and the robot 10a stops while the arm 12 is at the target joint angle 901M. After moving to the position / orientation 900M, the gripper 11 holds the article 101M. Further, the stop position / posture 900Q and the target joint angle 901Q in a state where the article 101M is taken out from the storage shelf 100M are calculated, and the robot 10a is in the stop position / posture in a state where the arm 12 holds the article 101M (901M). After moving to 900Q, the arm 12 changes to the target joint angle 901Q. With this series of operations, even if the article 101M is stored in the back of the storage shelf 100M, the article 101M can be moved from the position where the gripper 11 can correctly contact the article 101M without the robot 10a coming into contact with the storage shelf 100M or other articles. And the mistake of taking out a different article 101 or losing the article 101M can be reduced.

<Example 3>
FIG. 13 is a diagram illustrating the configuration of the robot system according to the third embodiment.

  The robot system according to the third embodiment includes a robot 10b, a transport vehicle 90b, and an overall management apparatus 200.

  The robot 10 b includes an arm 12 having a gripper 11 at the tip, an article measurement sensor 18 b that recognizes an article held by the gripper 11, a calculation unit 14, a suction force generation unit 15, and a power supply unit 16. Unlike the robot 10a according to the first embodiment, the robot 10b according to the third embodiment does not include a traveling unit and an ambient environment measurement sensor. Since the configuration of the arm 12, the article measurement sensor 18b, the calculation unit 14, the suction force generation unit 15 and the power supply unit 16 is the same as that of the robot 10a of the first embodiment described above, description thereof is omitted.

  The transport vehicle 90b includes a loading unit 91, a traveling unit 13a, and an ambient environment measurement sensor 17b.

  The loading unit 91 is provided in the upper part of the transport vehicle 90b, and ascends and descends when it reaches a predetermined position and posture or according to a command signal from the overall management apparatus 200. For example, as illustrated in FIG. 14, as the stacking unit 91 rises, the transport vehicle 90 b engages with the lower portion of the storage shelf 100 b and transports the storage shelf 100 b with the storage shelf 100 b lifted. Further, as shown in FIG. 15, the storage container 150 b placed on the stacking unit 91 is transported.

  The traveling unit 13a includes a rotating tire and a motor that rotationally drives the tire. Two tires are provided opposite to each other, and by controlling the rotation direction and the rotation speed of each tire, the transport vehicle 90b travels, changes its direction, and can rotate on the spot. An auxiliary wheel (not shown) may be provided in the lower part of the transport vehicle 90b.

  The ambient environment measurement sensor 17b is composed of a two-dimensional or three-dimensional laser scanner having a laser light irradiation unit and a laser light receiving unit, and irradiates the surrounding with laser light and measures reflected light. The distance to the surrounding obstacles (walls, storage shelves, etc.) can be calculated from the laser reflected light measured by the surrounding environment measurement sensor 17b, and the calculated distance is connected to determine the position and size of the surrounding obstacles. It can be calculated.

  The overall management apparatus 200 is connected to the robot 10b by wired or wireless communication, and is connected to the transport vehicle 90b by wireless communication. Since the configuration of the overall management apparatus 200 is the same as that of the first embodiment described above, description thereof is omitted.

  In the present embodiment, the calculation unit 14 has one configuration of the robot 10b, but may be configured separately from the robot 10b. For example, the transport vehicle 90b may include the calculation unit 14 so that the transport vehicle 90b and the robot 10a can communicate with each other.

  Next, with reference to FIGS. 14 and 15, picking of an article by the robot 10b will be described.

  FIG. 14 is a diagram illustrating the robot system according to the third embodiment that picks articles in a storage shelf.

  The robot system shown in FIG. 14 includes a robot 10b, a transport vehicle 90b, a storage shelf 100b, and an overall management device 200 (not shown). The transport vehicle 90b loaded with the storage shelf 100b moves so that the robot 10b is positioned in front of the article 101b taken out from the storage shelf 100b. The robot 10b picks up the gripper 11 on the side surface of the article 101b and bends the joint of the arm 12 to take out the article 101b from the storage shelf 100b.

  FIG. 15 is a diagram illustrating a robot system according to a third embodiment that picks an article in a container.

  The robot system shown in FIG. 15 includes a robot 10b, a transport vehicle 90b, a storage container 150b, and an overall management device 200 (not shown). The transport vehicle 90b loaded with the storage container 150b moves so that the robot 10b is positioned in front of the article 101b taken out from the storage container 150b. The robot 10b picks up the gripper 11 on the upper surface of the article 101b and bends the joint of the arm 12 to take out the article 101b from the storage container 150b.

  Although the storage shelf 100b and the storage container 150b have been described as storage means for storing articles in the third embodiment, any form may be used as long as the articles can be stored and stored.

  FIG. 16 is a flowchart of a process for controlling the operation of the robot 10b according to the third embodiment. The program for performing the process shown in FIG. 16 is executed by the CPU of the calculation unit 14 of the robot 10b.

  First, the robot 10b receives from the overall management apparatus 200 a command to take out the article 101b from the storage shelf 100b. Further, when the robot 10b receives a command to take out the article 101b from the storage shelf 100b from the overall management apparatus 200 (2001), the transport vehicle 90b autonomously moves to a position where the storage shelf 100b can be loaded (2002) and stores. The shelf 100b is loaded (2003), and moves autonomously to the position before the robot 10b, that is, the position where the robot 10b can measure the article 101 stored in the storage shelf 100b, and transports the storage shelf 100b (2004). The routes 2002 and 2004 correspond to long distance routes.

  Then, the robot 10b measures the article 101 stored in the storage shelf 100b by the article measurement sensor 18b (2005). And the calculating part 14 calculates the detailed position and orientation of each article | item 101 in the storage shelf 100b using the shape and magnitude | size of the article | item 101 registered previously with respect to the measurement result by the article | item measurement sensor 18b. 2006). For example, the article measurement sensor 18b acquires a distance image including the article 101, and searches the distance image for a portion that matches the shape and size of the article 101 registered in advance. A simple position and orientation can be calculated. In addition, when the detailed position and orientation of the article 101 in the storage shelf 100b can be acquired from the overall management apparatus 200, the processes in steps 2005 and 2006 can be omitted.

  And the calculating part 14 calculates the stop position attitude | position 900b of the storage shelf 100b for hold | maintaining the articles | goods 101b (2007). This calculation is realized, for example, by simulation by the calculation unit 14. First, the absolute position of the gripper 11 for holding the article 101b from the data of the relative position and orientation of the gripper 11 viewed from the article 101b for holding the predetermined article 101b and the position and orientation of the article 101b calculated in 2006. Calculate the position and orientation. Subsequently, in the simulation, a temporary stop position / posture of the storage shelf 100b is set, and the absolute position / posture of the gripper 11 calculated in the temporary stop position / posture is calculated using the shape and mechanism data of the robot 10b. To see if it is possible or impossible to take. This confirmation can be realized by solving the inverse kinematics problem. If the result of this confirmation is possible, the temporary stop position / posture is output as the stop position / posture 900b. If the result of the confirmation is impossible, the temporary stop position / posture is slightly changed, and confirmation of whether or not the gripper 11 can realize the absolute position / posture is executed again. This is repeated until a confirmation result is possible. In the present embodiment, the relative position and orientation of the gripper 11 viewed from the article 101b for holding the article 101b is perpendicular to the center of the front surface of the article 101b, so that the storage shelf 100b for taking out the article 101b The stop position is a position where the robot 10b faces the front of the article 101b (the surface facing the robot 10b).

  Thereafter, the transport vehicle 90b autonomously moves to the calculated stop position / posture 900b and transports the storage shelf 100b (2008). The route at this time corresponds to a short-distance route, and the transport vehicle 90b turns on the spot toward the position of the stop position / posture 900b at the current location, and moves linearly to the position of the stop position / posture 900b, The posture is adjusted by turning on the spot at the position of the stop position / posture 900b, and the storage shelf 100b finally becomes the stop position / posture 900b.

  The angle of the joint of the arm 12 is changed to the target joint angle 901b so that the gripper 11 moves to a position and posture capable of holding the article 101b (2009). The result of solving the inverse kinematics problem in 2007 is the target joint angle 901b of the arm 12 for the robot 10a to take the absolute position and orientation of the gripper 11. For example, there is a method of gradually changing the joint angle from the current joint angle to the target joint angle 901b at a constant angular velocity. As a result, the gripper 11 moves so that the article 101b can be adsorbed in front of the article 101b to be taken out.

  Next, the gripper 11 holds the article 101b (2010). For example, the gripper 11 that is in surface contact with the article 101 b is decompressed, and the article 101 b is adsorbed to the gripper 11.

  Then, the joint angle of the arm 12 is changed to the end joint angle 902 so that the article 101b is pulled out from the storage shelf 100b while holding the article 101b, and the gripper 11 is moved (2011). For example, without changing the posture of the gripper 11 and the posture of the article 101b in the state of 2010, the gripper 11 is moved vertically to the front of the storage shelf 100b to a position where the article 101b is completely separated from the storage shelf 100b. The joint angle of the arm 12 when the article 101b is completely separated from the storage shelf 100b is the end joint angle 902.

  Thereafter, after the joint angle of the arm 12 is changed to the end joint angle 902, the transport vehicle 90b autonomously moves to the installation position and posture, transports the storage shelf 100b (2012), and the storage shelf 100b is determined. Install in position (2013). Thereafter, the transport vehicle 90b transmits a work completion report to the overall management apparatus 200 (2014). The route 2012 corresponds to a long distance route.

  FIG. 17 is a diagram illustrating a stop position / posture 900 of the storage shelf 100b corresponding to the position / posture of the article 101 according to the third embodiment.

  As described above, in the third embodiment, the transport vehicle 90b (storage shelf 100b) stops at a position where the gripper 11 of the robot 10b can hold the article 101b. That is, the position and orientation at which the transport vehicle 90b stops differs depending on the position and orientation of the article 101 to be taken out.

  For example, when taking out the article 101b in the upper row of the front row under the storage shelf 100b, the transport vehicle 90b (storage shelf 100b) stops at the stop position / posture 900b. Further, when taking out the article 101f in the upper left column of the front opening under the storage shelf 100b, the transport vehicle 90b (storage shelf 100b) stops at the stop position / posture 900f. Furthermore, when taking out the article 101j in the upper left column of the front opening below the storage shelf 100b, the transport vehicle 90b (storage shelf 100b) stops at the stop position / posture 900j.

  The stop position / posture 900 and the target joint angle 901 differ depending on the type of the gripper 11, the attachment position / posture with respect to the arm 12, and the shape and size of the article 101. Therefore, the calculation unit 14 calculates the stop position / posture 900 and the target joint angle 901 using information on the direction in which the gripper 11 accesses the article 101.

  So far, in the third embodiment, the case where the article stored in the storage shelf 100b shown in FIG. 14 is taken out has been described, but the same applies to the case where the article 151b stored in the storage container 150b shown in FIG. 15 is taken out. is there. That is, the robot 10b moves to a position where the gripper 11 faces the upper surface of the article and can attract the article 151b.

  As described above, in the third embodiment, the stop position / posture 900b of the transport vehicle 90b (storage shelf 100b) for the robot 10b to hold the article 101b is calculated, and the storage shelf 100b is transported to the stop position / posture 900b. After that, since the gripper 11 holds the article 101b, it is possible to access the article 101b from a position where the gripper 11 can correctly contact the article 101b, and it is possible to reduce mistakes that a different article 101 is taken out or the article 101b is missed.

<Example 4>
In this embodiment, an example of an article picking robot system in the storage shelf that can take out the article 101 placed in the back of the storage shelf 100 will be described.

  FIG. 18 is a diagram illustrating a configuration of the robot system according to the fourth embodiment.

  The robot system according to the fourth embodiment includes a robot 10b, a transport vehicle 90b, a storage shelf 100n, and an overall management apparatus 200. The robot 10b moves to the front of the article 101n taken out from the storage shelf 100n, sucks the gripper 11 on the side surface of the article 101n, and bends the joint of the arm 12, thereby taking out the article 101n from the storage shelf 100n.

  In the fourth embodiment, an example in which the robot 10b takes out an article 101n placed in the back of the storage shelf 100n will be described. When taking out the article 101n placed in the back of the storage shelf 100n, if the arm 12 is changed to the target joint angle 901n to attract the article 101n while the storage shelf 100n is close to the robot 10b, the storage shelf 100n It may abut against a shelf plate that divides the frontage. Further, if the arm 12 is immediately changed to the end joint angle 902 after the robot 10b has adsorbed the article 101n, the robot 10b may come into contact with a shelf plate that divides the storage shelf 100n. For this reason, a special procedure is required when taking out the article 101n placed behind the storage shelf 100n.

  FIG. 19 is a diagram illustrating a stop position / posture 900n of the robot 10b for taking out the article 101n placed in the back of the storage shelf 100n according to the fourth embodiment.

  As described above, in the fourth embodiment, the transport vehicle 90b (storage shelf 100n) stops at a position where the gripper 11 of the robot 10b can hold the article 101n. That is, the stop position and posture of the robot vary depending on the article 101 to be taken out. For example, when taking out the article 101n in the upper left column of the front opening below the storage shelf 100n, the transport vehicle 90b transports the storage shelf 100n to a position in front of the robot 10b, and then approaches the robot 10b to the stop position posture 900n. Stop. The stop position / posture 900n is closer to the robot 10b than the stop position / posture 900f for taking out the foremost article 101 of the storage shelf 100n.

  Next, with reference to FIG. 20 and FIG. 21, an operation in which the robot 10b according to the fourth embodiment takes out the article 101n placed behind the storage shelf 100n will be described. FIG. 20 is a flowchart of a process for controlling the operation until the robot 10b according to the fourth embodiment holds the article 101n. FIG. 21 is a diagram illustrating an operation until the robot 10b according to the fourth embodiment holds the article 101n.

  First, the robot 10b executes steps 2001 to 2006 in FIG. 16 (2101). That is, the transport vehicle 90b receives an instruction from the overall management apparatus 200 for the robot 10b to take out the article 101n from the storage shelf 100n (2001), autonomously moves to a position where the storage shelf 100n can be loaded (2002), and stores it. The rack 100n is loaded (2003), and the robot 10b autonomously moves to a position where the article 101 stored in the storage rack 100n can be measured (2004). Further, the robot 10b receives a command for taking out the article 101n from the storage shelf 100n from the overall management apparatus 200 (2001), measures the article 101 stored in the storage shelf 100n by the article measurement sensor 18b (2005), and stores the storage shelf. The detailed position and orientation of each article 101 in 100n are calculated (2006). In this state, as shown in FIG. 21A, the transport vehicle 90b is at a position away from the robot 10b so that the robot 101b can overlook the article 101n to be taken out and the surrounding situation.

  Then, the calculation unit 14 calculates the stop position / posture 900n of the storage shelf 100n for holding the article 101n and the target joint angle 901n of the arm 12 (2102). This calculation is realized, for example, by simulation by the calculation unit 14. First, the absolute position and orientation of the gripper 11 for holding the article 101n from the data of the relative position and orientation of the gripper 11 viewed from the article 101n for holding the predetermined article 101n and the position and orientation of the article 101n calculated in 2006. Calculate the position and orientation. Subsequently, in the simulation, a temporary stop position / posture of the storage shelf 100n is set, and it is confirmed whether or not the robot 10b can take the absolute position / posture of the gripper 11 calculated at the temporary stop position / posture. . This confirmation can be realized by solving the inverse kinematics problem, and as a result, the temporary target joint angle of the arm 12 can be obtained simultaneously. If the result of this confirmation is possible, the robot 10a is further temporarily stopped using the shape and mechanism data of the robot 10b, the shape data of the storage shelf 100, and the distance image measured in step 2005 in step 2101. When the position and orientation and the provisional target joint angle are taken, it is confirmed whether or not the robot 10a comes into contact with the storage shelf 100 or another article. In the case of no contact, the temporary stop position / posture is output as the stop position / posture 900n, and the temporary target joint angle is output as the target joint angle 901n. If the result of the confirmation is not possible, or if it is “contact”, the temporary stop position / posture is changed slightly to confirm whether the absolute position / posture of the gripper 11 can be realized and whether the robot 10b comes into contact. Check again. These are repeated until a confirmation result is possible and “no contact”. In this embodiment, the relative position and orientation of the gripper 11 viewed from the article 101n for holding the article 101n is a vertical direction at the center of the front surface of the article 101n. The stop position of the transport vehicle 90b) is a position where the robot 10b faces the front surface of the article 101n (the surface facing the robot 10b). Further, a state in which the arm 12 is extended so that the gripper 11 reaches the article 101n at the back of the storage shelf 100n is a target joint angle 901n (FIG. 21B).

  Next, the calculation unit 14 determines whether the arm 12 can be changed from the current joint angle to the target joint angle 901n when the storage shelf 100n (and the transport vehicle 90b) is in the stop position / posture 900n (2103). . For example, referring to the warehouse in which the storage shelf 100n is installed and the shape data of the storage shelf 100n (for example, the position and size of the walls and pillars of the warehouse, the CAD data representing the shape and position of the storage shelf 100n), In the space where the tip (gripper 11) can move, the arm 12 does not interfere with obstacles (storage rack 100n, transport vehicle 90b, other robots 10, warehouse columns, walls, etc.) and the current joint angle. To search for a trajectory that can be shifted to the target joint angle 901n. This search is realized by, for example, simulation by the calculation unit 14. First, in the simulation, using the stop position / posture 900n, the robot 10b shape / mechanism data, the storage shelf 100 shape data, and the distance image measured in step 2005 in step 2101, the joint angles of a large number of arms 12 are obtained. Then, it is confirmed whether or not the robot 10b comes into contact with the storage shelf 100n or another article, and the joint angle that does not come in contact is stored. Next, arbitrarily select two stored joint angles, determine whether or not they are close using a predetermined threshold value, and create a joint angle transition graph connecting the close joint angles. Thereafter, the joint angle transition graph is traced to search for a path that reaches the target joint angle 901n from the current joint angle. If a path can be found within a predetermined time, it is determined that the arm 12 can transition from the current joint angle to the target joint angle 901n. For example, as shown in FIG. 21 (c), when the storage shelf 100n (and the transport vehicle 90b) is in the stop position / posture 900n and the arm 12 always tries to change to the target joint angle 901n, A path cannot be found within a predetermined time, and it is determined that the transition is impossible.

  As a result, if the arm 12 cannot shift to the target joint angle 901n, the process proceeds to step 2107. On the other hand, if the arm 12 can shift to the target joint angle 901n, the calculation unit 14 moves the arm 12 from the current joint angle to the target joint angle 901n when the storage shelf 100n (and the transport vehicle 90b) is in the stop position / posture 900n. The interference avoidance joint angle trajectory that transitions to is determined (2104). Specifically, using the result of the simulation in step 2103, the shortest path (or the path satisfying a predetermined condition) that transitions from the current state to the target joint angle 901n is searched and determined as an interference avoidance joint angle trajectory. To do.

  The transport vehicle 90b autonomously transports the storage rack 100n to the stop position / posture 900n (2105), controls the arm 12 according to the interference avoidance joint angle locus, and moves the gripper 11 (2106). The route 2105 corresponds to a short-distance route. Thereafter, the gripper 11 holds the article 101n (2111). For example, the gripper 11 in surface contact with the article 101n is decompressed, and the article 101n is adsorbed to the gripper 11.

  On the other hand, when it is determined in step 2103 that the arm 12 interferes with the obstacle and cannot be shifted to the target joint angle 901n, the calculation unit 14 determines in step 2107 that the arm 12 has the target joint angle 901n. Then, it is determined whether the storage shelf 100n (the transport vehicle 90b) can be moved from the current position / posture to the stop position / posture 900n. This determination is realized by, for example, simulation by the calculation unit 14. First, in the simulation, using the current joint angle of the arm 12, the shape data of the warehouse, the shape and mechanism data of the robot 10b, the shape data of the storage shelf 100, and the distance image measured in step 2005 in step 2101, In the position and orientation of a large number of storage shelves 100n in the vicinity of the robot 10b, it is confirmed whether or not the storage shelf 100n and the articles in the storage shelf 100n are in contact with the robot 10b and equipment in other warehouses, and the position and orientation that is not in contact are stored. Keep it. Next, two arbitrarily stored positions and orientations are selected, it is determined whether or not they are close using a predetermined threshold value, and a movement graph connecting the close positions and orientations is created. Thereafter, the movement graph is traced to search for a route from the current position / posture to the stop position / posture 900n. When a path can be found within a predetermined time, it is determined that the storage shelf 100n (the transport vehicle 90b) can be moved from the current position / posture to the stop position / posture 900n.

  As a result, if the transport vehicle 90b cannot move to the stop position / posture 900n, the process returns to Step 2102, and another combination of the stop position / posture 900n and the target joint angle 901n is calculated. At this time, one of the stop position / posture 900n and the target joint angle 901n may be new, or both the stop position / posture 900n and the target joint angle 901n may be new.

  On the other hand, if the transport vehicle 90b can move to the stop position / posture 900n without interfering with the obstacle, the calculation unit 14 determines that the transport vehicle 90b is in the current position / posture when the arm 12 has the target joint angle 901n. The interference avoidance movement trajectory 904n for conveying the storage shelf 100n to the stop position / posture 900n is determined (2108). Specifically, using the result of the simulation in step 2107, the shortest path (or the path that satisfies a predetermined condition) that moves from the current position / posture to the stop position / posture 900n is searched, and the interference avoidance movement trajectory 904n is obtained. Determine (FIG. 21 (d)).

  Thereafter, the arm 12 changes the joint angle to the target joint angle 901n (2109), and the transport vehicle 90b moves autonomously according to the interference avoidance movement trajectory 904n to transport the storage shelf 100n (2110, FIG. 21 (e)). reference). Thereafter, the gripper 11 holds the article 101n (2111, see FIG. 21 (f)).

  As shown in FIG. 22, the transport vehicle 90b (storage shelf 100n) of the fourth embodiment sets the joint angle of the arm 12 to the target joint so that the robot 10b holds the article 101n placed in the back of the storage shelf 100n. In a state where the angle is 901n (2109), the robot 10b is approached according to the interference avoidance movement locus 904n.

  FIG. 23 is a flowchart of a process for controlling the operation after the robot 10b according to the fourth embodiment holds the article 101n. FIG. 24 is a diagram illustrating an operation after the robot 10b according to the fourth embodiment holds the article 101n. That is, the process shown in FIG. 23 is executed after the process shown in FIG.

  First, the calculation unit 14 calculates the stop position / posture 900r of the storage shelf 100n (the transport vehicle 90b) and the target joint angle 901r of the arm 12 for the robot 10b to pull out the storage shelf 100n while holding the article 101n (2112). ). The stop position / posture 900r and the target joint angle 901r calculated here are the position / posture and arm of the storage shelf 100n (the transport vehicle 90b) in a state where the storage shelf 100n is separated from the robot 10b and the robot 10b can move to the next operation. 12 target joint angles (FIG. 24B).

  Next, the computing unit 14 determines whether the arm 12 can be changed from the current joint angle to the target joint angle 901r when the storage shelf 100n (the transport vehicle 90b) is in the current position and posture (2113). For example, the shape data of the warehouse in which the storage shelf 100n is installed and the storage shelf 100n (for example, CAD data representing the position and size of the walls and pillars of the warehouse and the shape and position of the storage shelf 100n as in Step 2103). Referring to the trajectory that can move from the current joint angle to the target joint angle 901r without the arm 12 interfering with an obstacle in a space where the tip of the arm 12 (gripper 11) can move. This search is realized by, for example, simulation by the calculation unit 14. First, in the simulation, using the current position and orientation, the shape and mechanism data of the robot 10b, the shape data of the storage shelf 100, and the distance image measured in step 2005 in step 2101, It is confirmed whether or not the robot 10b comes into contact with the storage shelf 100 or other articles, and the joint angles that do not come in contact are stored. Next, arbitrarily select two stored joint angles, determine whether or not they are close using a predetermined threshold value, and create a joint angle transition graph connecting the close joint angles. Thereafter, the joint angle transition graph is traced to search for a path that reaches the target joint angle 901r from the current joint angle. If the path can be found within a predetermined time, it is determined that the arm 12 can transition from the current joint angle to the target joint angle 901r. For example, as shown in FIG. 24 (c), when the storage rack 100n (the transport vehicle 90b) is in the current position and posture and the arm 12 always tries to shift to the target joint angle 901r, it is determined in advance. The route is not found within the given time, and it is determined that the transition is impossible.

  As a result, if the arm 12 cannot shift to the target joint angle 901r, the process proceeds to step 2117. On the other hand, if the arm 12 can be shifted to the target joint angle 901r, the calculation unit 14 causes the interference avoidance joint angle to shift the arm 12 from the current joint angle to the target joint angle 901r when the transport vehicle 90b is in the current position and posture. A trajectory is determined (2114). Specifically, using the result of the simulation in step 2113, the shortest path (or the path satisfying a predetermined condition) that transitions from the current state to the target joint angle is searched and determined as an interference avoidance joint angle trajectory. .

  Then, the robot 10b controls the arm 12 according to the interference avoidance joint angle locus, moves the gripper 11 (2115), and the transport vehicle 90b autonomously transports the storage shelf 100n to the stop position / posture 900r (2116). The route 2116 corresponds to a short-distance route. Thereafter, steps 2011 to 2014 are executed (2121). That is, with the gripper 11 holding the article 101n, the joint angle of the arm 12 is changed to the end joint angle 902, the article 101n is pulled out from the storage shelf 100n (2011), and the transport vehicle 90b is autonomously stored in the installation position / posture. The shelf 100n is transported (2012), the storage shelf 100n is installed at a predetermined position (2013), and the transport vehicle 90b transmits a work completion report to the overall management apparatus 200 (2014).

  On the other hand, if it is determined in step 2113 that the arm 12 cannot move to the target joint angle 901r without interfering with an obstacle, the computing unit 14 determines that the storage shelf It is determined whether 100n (the transport vehicle 90b) can be moved from the current position / posture to the stop position / posture 900r (2117). This determination is realized by, for example, simulation by the calculation unit 14. First, in the simulation, the robot uses the current joint angle of the arm 12, the shape data of the warehouse, the shape and mechanism data of the robot 10a, the shape data of the storage shelf 100, and the distance image measured in step 2005 in step 2101. For the positions and orientations of a large number of storage shelves 100n in the vicinity of 10b, it is confirmed whether or not the storage shelves 100n and the articles in the storage shelf 100n are in contact with the robot 10b or equipment in other warehouses, and the positions and orientations that are not in contact are stored. Keep it. Next, two arbitrarily stored positions and orientations are selected, it is determined whether or not they are close using a predetermined threshold value, and a movement graph connecting the close positions and orientations is created. Thereafter, the movement graph is traced to search for a path from the current position / posture to the stop position / posture 900r. If a path can be found within a predetermined time, it is determined that the storage shelf 100n (the transport vehicle 90b) can be moved from the current position / posture to the stop position / posture 900r (FIG. 24 (d)).

  As a result, if the storage shelf 100n (the transport vehicle 90b) cannot move to the stop position / posture 900r, the process returns to Step 2112 to calculate another combination of the stop position / posture 900r and the target joint angle 901r. At this time, one of the stop position / posture 900r and the target joint angle 901r may be new, or both the stop position / posture 900r and the target joint angle 901r may be new.

  On the other hand, if the storage rack 100n (the transport vehicle 90b) can move to the stop position / posture 900r without interfering with an obstacle, the calculation unit 14 can perform the transport vehicle when the arm 12 has the target joint angle 901r. 90b determines an interference avoidance movement trajectory 904r for conveying the storage shelf 100n from the current position / posture to the stop position / posture 900r (2118). Specifically, the shortest route (or a route satisfying a predetermined condition) that moves from the current position / posture to the stop position / posture 900r is searched using the result of the simulation in Step 2107, and the interference avoidance movement locus 904r is obtained. decide.

  Thereafter, the transport vehicle 90b travels according to the interference avoidance movement locus 904r, transports the storage shelf 100n (2119, FIG. 24 (e)), and the arm 12 changes the joint angle to the target joint angle 901r (2120, FIG. 24 (f)). Thereafter, steps 2011 to 2014 are executed (2121), and finally, the transport vehicle 90b transmits a work completion report to the overall management apparatus 200.

  So far, in the fourth embodiment, the case where the article stored in the storage shelf 100n shown in FIG. 18 is taken out has been described, but the same applies to the case where the article 151b stored in the storage container 150b shown in FIG. 15 is taken out. . That is, the conveyance vehicle 90b conveys the storage container 150b to a position where the gripper 11 faces the upper surface of the article and can attract the article 151b.

  As described above, in the fourth embodiment, the stop position / posture 900n of the storage shelf 100n (the transport vehicle 90b) for holding the article 101n by the robot 10b and the target joint angle 901n of the arm 12 are calculated, and the arm 12 After the storage shelf 100n is transported to the stop position / posture 900n with the target joint angle 901n, the gripper 11 holds the article 101n. Further, the stop position / posture 900r and the target joint angle 901r in a state where the article 101n is taken out from the storage shelf 100n are calculated, and the carrier vehicle 90b is in the state of the joint angle (901n) where the arm 12 holds the article 101n. After moving 100n to the stop position / posture 900r, the arm 12 changes to the target joint angle 901r. By this series of operations, even if the article 101n is stored in the back of the storage shelf 100n, the article 101n can be moved from the position where the gripper 11 can correctly contact the article 101n without the robot 10a coming into contact with the storage shelf 100n or other articles. And the mistake of taking out a different article 101 or losing the article 101n can be reduced.

  In the embodiment described above, the case where the article 101 is taken out from the storage means (storage shelf 100, storage container 150) has been described. However, the present invention can also be applied to the case where the robot 10 stores the article 101 in the storage means. .

  As described above, according to the embodiment of the present invention, the calculation unit 14 has the shape of the storage shelf 100, the shape of the robot 10, the shape of the target article 101 to be taken out, and the target article 101 in the storage shelf 100. The robot 10 calculates the stop position / posture 900 of the robot 10 and the target joint angle 901 of the arm 12, and the robot 10 moves to the stop position / posture 900 to change the joint angle of the arm 12 to the target joint angle 901. Since the target article 101 is taken out from the storage shelf 100 and changed, the robot having an arm with a short reach and a low degree of joint freedom does not come in contact with an obstacle (the storage shelf 100, other articles, etc.) and does not touch the storage shelf 100. The article 101 can be taken out from the container. For this reason, the cost of a robot can be reduced and the restriction | limiting of the shelf shape in an inexpensive robot can be made low.

  Further, the calculation unit 14 realizes the first stop position / posture 900M and the first target joint angle 901M in a state where the robot 10 approaches the storage shelf 100 (a state in which the article 101 in the storage shelf 100 can be accessed). In order to do this, in a state where the joint angle of the arm 12 is set to the first target joint angle 901M, a path 904M on which the robot 10 moves from the current position and posture to the first stop position and posture 900M without interfering with an obstacle is provided. Since the robot 10 approaches the storage shelf 100 according to the calculated path 904M, the article 101 placed in the back of the storage shelf 100 can be held without contacting the obstacle.

  In addition, based on the position and orientation of the target article 101 in the storage shelf 100, the calculation unit 14 combines the stop position and orientation 900 in which the robot 10 holds the target article 101 without interfering with an obstacle and the target joint angle 901. Therefore, a series of operations for taking out the article 101 placed in the back of the storage shelf 100 can be appropriately planned.

  In addition, the calculation unit 14 includes the second stop position / posture 900Q and the second target in a state where the robot 10 has moved away from the storage shelf 100 (the robot 10 has moved away from the storage shelf 100 and can move to another location). In order to realize the joint angle 901Q, a path 904Q is calculated in which the robot 10 moves from the current position and orientation to the second stop position and orientation 900Q without interfering with an obstacle while the joint angle of the arm 12 is in the current state. Thus, since the robot 10 moves away from the storage shelf 100 according to the calculated path 904Q, it is possible to leave the storage shelf 100 without contacting the obstacle after holding the article 101 placed behind the storage shelf 100.

  In addition, the calculation unit 14 calculates a combination of the stop position / posture 900 and the target joint angle 901 that are separated from the storage shelf 100 while the robot 10 holds the target article 101 from a state in which the robot 10 and the storage shelf 100 are close to each other. Therefore, a series of operations for taking out the article 101 placed in the back of the storage shelf 100 can be appropriately planned.

  Further, the calculation unit 14 controls the movement of the robot 10 to the stop position / posture 900 based on the shape of the storage shelf 100, the shape of the robot 10, the shape of the target article 101, and the position / posture of the target article 101. The order of the control for changing the joint angle of the arm 12 to the target joint angle 901 is selected. Although it is desirable for safety that the robot 10 moves with the arm bent, it is necessary to move with the arm extended when the purpose cannot be achieved. Thus, appropriate control can be performed by selecting the order of control.

  Further, the article measurement sensor 18 moves to a position and orientation where the article 101 stored in the storage shelf 100 can be measured, and the article measurement sensor 18 stores the robot after the robot 10 has moved to a measurable position and orientation. Since the article 101 stored in the shelf 100 is measured and the calculation unit 14 recognizes the position and orientation of the target article 101 in the storage shelf 100 based on the measurement result, the central control device (overall management device 200). Thus, the management of the article 101 is not necessary, and the system load can be distributed.

  In addition, the calculation unit 14 stops the storage shelf 100 based on the shape of the storage shelf 100, the shape of the robot 10, the shape of the target article 101 to be taken out, and the position and orientation of the target article 101 in the storage shelf 100. The position / posture 900 and the target joint angle 901 of the arm 12 are calculated, the transport vehicle 90 transports the storage shelf 100 to the stop position / posture 900, and the robot 10 changes the joint angle of the arm 12 to the target joint angle 901. Since the target article 101 is taken out from the storage shelf 100, even a robot having an arm with a short reach and a low degree of joint freedom does not come in contact with an obstacle (the storage shelf 100, another article, etc.) and the article 101 from the storage shelf 100. Can be taken out. For this reason, the cost of a robot can be reduced and the restriction | limiting of the shelf shape in an inexpensive robot can be made low.

  In addition, the calculation unit 14 includes the first stop position / posture 900n and the first target joint angle in a state in which the storage shelf 100 approaches the robot 10 (a state in which the robot 10 can access the article 101 in the storage shelf 100). In order to realize 901n, the storage rack 100 and the stored article do not interfere with an obstacle (such as the robot 10) in the state where the joint angle of the arm 12 is the first target joint angle 901n, and the current position. The path 904n that moves from the attitude to the first stop position / posture 900n is calculated, and the storage shelf 100 approaches the robot 10 along the calculated path 904n. Therefore, the robot 10 is placed in the back of the storage rack 100. 101 can be held without contact with an obstacle (such as the storage shelf 100 or another article).

  In addition, based on the position and orientation of the target article 101 in the storage shelf 100, the calculation unit 14 causes the robot 10 to move the target article without causing the storage shelf 100 and the robot 10 to interfere with obstacles (such as each other or other articles). Since the combination of the stop position / posture 900 that holds 101 and the target joint angle 901 is calculated, a series of cooperative operations for taking out the article 101 placed in the back of the storage shelf 100 can be appropriately planned.

  In addition, the calculation unit 14 includes the second stop position / posture 900r and the second target in a state where the storage shelf 100 is away from the robot 10 (the storage shelf 100 is separated from the robot 10 and can be moved to another location). In order to realize the joint angle 901r, the joint angle of the arm 12 is the current state, and the storage shelf 100 and the stored article do not interfere with an obstacle (such as the robot 10) from the current position and orientation. Since the path 904r conveyed to the stop position / posture 900r is calculated and the storage shelf 100 moves away from the robot 10 along the calculated path 904r, the obstacle 101 after holding the article 101 placed behind the storage rack 100 The storage shelf 100 can be detached from the robot 10 without contact with the robot 10 or the like.

  In addition, the calculation unit 14 stops the posture 900 when the storage shelf 100 and the stored article are separated from the robot 10 in a state where the robot 10 holds the target article 101 from a state where the robot 10 and the storage shelf 100 are close to each other. Since the target joint angle 901 is calculated as a combination, a series of cooperative operations for taking out the article 101 placed in the back of the storage shelf 100 can be appropriately planned.

  In addition, the calculation unit 14 causes the transport vehicle 90 to stop the storage shelf 100 based on the shape of the storage shelf 100, the shape of the robot 10, the shape of the target article 101, and the position and orientation of the target article 101. The order of the control to transfer to the target joint angle and the control to change the joint angle of the arm 12 to the target joint angle 901 is selected. Although it is desirable from the viewpoint of work throughput that the storage shelf 100 travels toward the robot 10 is minimized, if the purpose cannot be achieved, the storage shelf 100 is extended so as to avoid the arm 12. 100 needs to be moved. Thus, appropriate control can be performed by selecting the order of control.

  Further, the transport vehicle 90 transports the storage shelf 100 to a position and orientation where the article measurement sensor 18 can measure the article 101 stored in the storage shelf 100, and the robot 10 moves the storage shelf 100 to the measurable position and orientation. After being conveyed, the article 101 stored in the storage shelf 100 is measured, and the calculation unit 14 recognizes the position and orientation of the target article 101 based on the measurement result. ), The management of the article 101 becomes unnecessary, and the system load can be distributed.

  The present invention is not limited to the above-described embodiments, and includes various modifications and equivalent configurations within the scope of the appended claims. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and the present invention is not necessarily limited to those having all the configurations described. A part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Moreover, you may add the structure of another Example to the structure of a certain Example. In addition, for a part of the configuration of each embodiment, another configuration may be added, deleted, or replaced.

  In addition, each of the above-described configurations, functions, processing units, processing means, etc. may be realized in hardware by designing a part or all of them, for example, with an integrated circuit, and the processor realizes each function. It may be realized by software by interpreting and executing the program to be executed.

  Information such as programs, tables, and files that realize each function can be stored in a storage device such as a memory, a hard disk, and an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, and a DVD.

  Further, the control lines and the information lines are those that are considered necessary for the explanation, and not all the control lines and the information lines that are necessary for the mounting are shown. In practice, it can be considered that almost all the components are connected to each other.

DESCRIPTION OF SYMBOLS 1 Shelf display article picking robot system 2 In-container article picking robot system 10 Robot 11 Gripper 12 Arm 13 Running part 14 Calculation part 15 Suction force generator 16 Power supply part 17 Ambient environment measurement sensor 18 Article measurement sensor 90 Transport vehicle 91 Loading part 100 Storage shelf 101 Shelf display article 150 Storage container 151 In-container article 200 Overall management apparatus 900 Stop position and orientation 901 Target joint angle 902 End joint angle 904 Interference avoidance movement locus

Claims (15)

A robot system,
A robot for taking out articles from the storage means;
A calculation unit that is provided outside or inside the robot and controls the robot;
The robot is
A gripper capable of holding the article;
An arm capable of changing the position and orientation of the gripper;
A traveling unit capable of changing the position and orientation of the robot,
The computing unit is configured to determine the stop position and posture of the robot and the shape of the robot, the shape of the robot, the shape of the target article to be taken out, and the position and orientation of the target article in the storage means. Calculate the target joint angle of the arm,
The robot is
Move to the stop position and posture,
Changing the joint angle of the arm to the target joint angle;
A robot system, wherein the target article is taken out from the storage means. The robot system according to claim 1,
The arithmetic unit calculates the joint angle of the arm to the first target joint in order to realize the first stop position / posture and the first target joint angle in which the robot approaches the storage unit. A robot system that calculates a path in which the robot moves from the current position and orientation to the first stop position and orientation without interfering with an obstacle in an angled state. The robot system according to claim 2,
The arithmetic unit calculates a combination of the first stop position and posture at which the robot holds the target article without interfering with the obstacle and the first target joint angle. . The robot system according to claim 1,
In order to realize the second stop position / posture and the second target joint angle in which the robot is away from the storage unit, the arithmetic unit is configured so that the joint angle of the arm is in a current state, A robot system for calculating a path from a current position and orientation to the second stop position and orientation without interfering with an obstacle. The robot system according to claim 4,
The calculation unit calculates a combination of the second stop position and posture separated from the storage unit and the second target joint angle while the robot holds the target article. . The robot system according to claim 1,
The arithmetic unit controls the arm to move the robot to the stop position and posture based on the shape of the storage means, the shape of the robot, the shape of the target article, and the position and orientation of the target article. The robot system is characterized in that the order of the control to change the joint angle to the target joint angle is selected. The robot system according to claim 1,
Having an article measurement sensor,
The robot moves to a position and orientation where the article measurement sensor can measure the article stored in the storage means,
The article measurement sensor measures an article stored in the storage unit after the robot has moved to the measurable position and orientation.
The robot system is characterized in that the calculation unit recognizes the position and orientation of the target article in the storage unit based on the measurement result. A robot system,
A robot for taking out articles from the storage means;
A transport vehicle for transporting the storage means;
A calculation unit that is provided outside or inside the robot and controls the robot;
The robot is
A gripper capable of holding the article;
An arm capable of changing the position and orientation of the gripper;
The computing unit is configured to determine the stop position and orientation of the storage means based on the shape of the storage means, the shape of the robot, the shape of the target article to be taken out, and the position and orientation of the target article in the storage means. Calculating a target joint angle of the arm;
The transport vehicle transports the storage means to the stop position and posture,
The robot is
Changing the joint angle of the arm to the target joint angle;
A robot system, wherein the target article is taken out from the storage means. The robot system according to claim 8, wherein
The arithmetic unit calculates the joint angle of the arm to the first target joint in order to realize the first stop position and posture and the first target joint angle in which the storage means approaches the robot. A robot system that calculates a path in which the storage means moves from the current position and orientation to the first stop position and orientation without interfering with an obstacle in an angled state. The robot system according to claim 9, wherein
The calculation unit calculates a combination of the first stop position and posture at which the robot holds the target article and the first target joint angle without the storage unit interfering with the obstacle. Characteristic robot system. The robot system according to claim 8, wherein
In order to achieve the second stop position / posture and the second target joint angle in which the storage means is away from the robot, the arithmetic unit is configured so that the joint angle of the arm is in a current state, A robot system characterized in that a storage means calculates a route transported from the current position and orientation to the second stop position and orientation without interfering with an obstacle. The robot system according to claim 11, wherein
The calculation unit calculates a combination of the second stop position and posture separated from the robot and the second target joint angle by the storage unit in a state where the robot holds the target article. A robot system. The robot system according to claim 8, wherein
The computing unit conveys the storage means to the stop position and orientation based on the shape of the storage means, the shape of the robot, the shape of the target article, and the position and orientation of the target article. The robot system is characterized in that the order of the control to perform and the control to change the joint angle of the arm to the target joint angle is selected. The robot system according to claim 8, wherein
Having an article measurement sensor,
The transport vehicle transports the storage means to a position and orientation where the article measurement sensor can measure the articles stored in the storage means,
The robot measures an article stored in the storage unit after the storage unit is transported to the measurable position and orientation;
The robot system is characterized in that the calculation unit recognizes the position and orientation of the target article in the storage unit based on the measurement result. A method for picking an article by a robot system,
A robot for taking out articles from the storage means;
A calculation unit provided outside or inside the robot, for controlling the robot;
Provided in a transport vehicle for transporting the robot or the storage means, and a traveling unit capable of changing the relative position and orientation of the robot and the storage means;
The robot has a gripper capable of holding the article, and an arm capable of changing the position and orientation of the gripper,
The method
Based on the shape of the storage means, the shape of the robot, the shape of the target article to be taken out, and the position and orientation of the target article in the storage means, Calculating a stop position / posture of the traveling unit and a target joint angle of the arm to determine a relative position / posture;
The traveling unit moves to the stop position and posture;
The robot changes the joint angle of the arm to the target joint angle;
The picking method, wherein the robot takes out the target article from the storage means.

Images