WO2021006306A1 - Automatic dishwashing system, automatic dishwashing method, automatic dishwashing program, and storage medium - Google Patents

Abstract

Provided is an automatic dishwashing system which may determine a type and a state of each dish, allow a robot to safely pick the dish up, and store the dish at a prescribed storage position, in a state where dishes are conveyed on a conveyor at three-dimensionally different angles.　An automatic dishwashing system according to an embodiment of the present invention is characterized by being provided with: a dish insertion part; a dish washing part; a dish conveying part; a dish information acquisition part which identifies a dish and acquires information about the dish; a dish placing part; and a dish pickup unit, wherein the acquisition information acquired by the dish information acquisition part includes at least one among the type, size, shape, color, position, orientation, direction, marker, or the code of the dish. When the dish information acquisition part identifies a dish which is being conveyed in the dish conveying part, the dish pickup part picks the dish up in correspondence to the state of the dish on the basis of the information acquired by the dish information acquisition part.

Description

Automatic dishwashing system, automatic dishwashing method, automatic dishwashing program and storage medium

The present invention relates to an automatic dishwashing system for automatically washing dishes and the like, an automatic dishwashing method, an automatic dishwashing program, and a storage medium.

In recent years, due to the shortage of employees at restaurants, there has been a demand to automate the processes of operations and services at restaurants. There is also a demand for automation of dishwashing, and conventionally, a technique as described in Patent Document 1 has been proposed.

In the dishwashing system described in Patent Document 1, a human-identifiable symbol or character is provided in a hill of the pottery in order to identify the pottery, and an image on the washing conveyor is acquired by a camera. By reading the symbols or characters provided on the pottery by template matching from the acquired image, the pottery is recognized and the pottery is picked by the robot arm mechanism.

The dishwashing system described in Patent Document 2 has an identifier for identifying the type of tableware (size, bowl, bowl, dish, etc.) as well as being able to identify the tableware on the hill, and the position, range and orientation of the identifier. The symbol for identifying the dish is written, and the image on the washing conveyor is acquired by the camera, the identifier provided on the ceramic is recognized by pattern matching from the acquired image, and the tableware is picked by the robot arm mechanism. .. Further, in Patent Document 2, the identifier is one that can be identified by a human being, and a barcode, a two-dimensional code, a magnetic code, and the like are excluded.

The dishwashing system described in Patent Document 3 is a type of tableware that is taken out from a dishwasher, moved from a waiting area to a picking area, and stationary, and processed with an image acquired by a picking area camera. The position of the center of gravity of the tableware is calculated, picked by the robot arm mechanism, and stacked in the release area for each type of tableware. In Patent Document 3, pattern matching processing is performed as image processing, and the type of tableware, the position of the center of gravity, and the orientation of the plate in the longitudinal direction when viewed from above (the orientation of the plate on the rectangle) are determined.

In the dishwashing system described in Patent Document 4, the washing rack is carried into the washing room, the door of the dishwashing machine is opened and closed, the washing switch is pressed, and the washing rack is carried out from the washing room to the unloading table. The robot arm is responsible for picking the dishes from the unloading table and transferring the dishes to the stock rack. With the washing rack stationary, the picking area on the unloading table corresponding to the position of the washing rack is pattern-matched with the image taken by the picking camera to determine the type of tableware, the position of the center of gravity of the tableware, and the dish. Is determined in the longitudinal direction (the orientation of the dish on the rectangle) when viewed from above.

JP-A-2018-126272 JP-A-2019-037349 International Publication No. 2018/034251 International Publication No. 2018/034252

In the dishwashing system described in Patent Document 1, the ceramics are recognized by the robot arm mechanism by template matching the human-identifiable symbols or characters provided in the hill of the ceramics from the image acquired by the camera. Is picking. However, it is premised that all the pottery is used in a state where it is lined up horizontally with respect to the conveyor (a state where the hill faces directly above), and it is described that even the type of pottery is identified. Absent. When the ceramics are conveyed on the conveyor at three-dimensionally different angles, the dishwashing system of Patent Document 1 cannot handle it. In addition, the symbols or characters provided on the hill of ceramics are limited to those that can be identified by humans. Furthermore, it has not been possible to automate the process of arranging ceramics on a cleaning conveyor.

In the dishwashing system described in Patent Document 2, an identifier for determining the type of tableware and a symbol for specifying the position, range and orientation of the identifier are provided in the tableware hill, and the dishwashing is performed by pattern matching processing. The system determines the type of tableware. However, it is premised that all the tableware is used in a state where it is lined up horizontally with respect to the conveyor (the state where the hill faces directly upward), and the tableware is transported on the conveyor at three-dimensionally different angles. In that case, the dishwashing system of Patent Document 2 cannot handle this. In addition, the symbols or characters provided on the hill of ceramics are limited to those that can be identified by humans. Furthermore, it has not been possible to automate the process of arranging dishes on the washing conveyor.

In the dishwashing system described in Patent Document 3, the type of tableware is determined by processing the image acquired by the picking area camera for the tableware that is stationary in the picking area. Therefore, it is necessary to separate the picking conveyor and the washing conveyor in the picking area and further provide a relay conveyor between them, which complicates the structure of the conveyor. In addition, it is assumed that all tableware is used in a state where it is lined up horizontally with respect to the conveyor (a state in which the hill faces directly upward), and the tableware is transported on the conveyor at three-dimensionally different angles. In that case, the dishwashing system of Patent Document 3 cannot handle this. Furthermore, it has not been possible to automate the process of arranging dishes on the washing conveyor.

In the dishwashing system described in Patent Document 4, in a state where the washing rack is stationary, pattern matching processing is performed on an image taken by a picking camera of a picking area on a carry-out table corresponding to the position of the washing rack. The type of tableware is determined by. Therefore, it cannot be used for picking up tableware moving on a conveyor. In addition, it is assumed that all tableware is used in a state where it is lined up horizontally with respect to the conveyor (a state in which the hill faces directly upward), and the tableware is transported on the conveyor at three-dimensionally different angles. In that case, the dishwashing system of Patent Document 4 cannot handle this. Moreover, it has not been possible to automate the process of arranging dishes in the washing rack.

In the above Patent Documents 1 to 4, it is not considered that the robot picks up the tableware in a state where the plurality of tableware is conveyed on the conveyor at three-dimensionally different angles. In order to realize a fully automatic automatic tableware washing system, the present inventor, for example, automates the stage of arranging the dishes before washing on the conveyor, and practically, the angle of arranging the dishes is not limited to the state where the hill faces directly upward. Fully automatic tableware through static and dynamic analysis of tableware and robots under the conditions such as picking up tableware moving on the conveyor and restrictions on the posture of the robot. As a result of advancing research and development on the automatic cleaning system, the embodiment of the present invention has been completed. That is, an object of the present invention is to determine the type and state of each tableware in a state where the tableware is conveyed on the conveyor at three-dimensionally different angles, the robot safely picks up the tableware, and a predetermined storage position. The purpose is to provide an automatic dishwashing system that can be placed on a conveyor.

The above object of the present invention can be achieved by the following configuration. That is, the automatic dishwashing system according to the first aspect of the present invention is
A dishwashing section that cleans the dishes that are thrown in from the tableware throwing section
A tableware transporting unit that transports dishes washed by the dishwashing unit,
A tableware information acquisition unit that identifies the tableware to be transported by the tableware transport unit and acquires information on the tableware.
A tableware storage unit on which the tableware transported by the tableware transport unit is placed,
The tableware is picked up from the tableware transport section, and the tableware is placed on the tableware storage section.
In an automatic dishwashing system equipped with
The information acquired by the tableware information acquisition unit includes the type of tableware, the size of the tableware, the shape of the tableware, the color of the tableware, the position of the tableware, the posture of the tableware, the orientation of the tableware, the marker attached to the tableware, or Includes at least one of the codes attached to the tableware,
When the tableware information acquisition unit identifies the tableware being transported by the tableware transport unit, the tableware collector picks up the tableware according to the state of the tableware based on the information acquired by the tableware information acquisition unit. It is characterized by.

According to the embodiment of the present invention, in a state where the tableware is conveyed on the conveyor at three-dimensionally different angles, the type and state of each tableware is determined, and the robot safely picks up the tableware and stores it in a predetermined state. An automatic dishwashing system that can be placed in place can be provided.

It is a process diagram of the tableware automatic washing system. It is the whole block diagram of the tableware automatic washing system. It is a block diagram of a tableware taking-out part. FIG. 4A is a plan view of the tableware taking-out part, and FIG. 4B is a perspective view of the tableware taking-out part. It is explanatory drawing of the use example of a marker. It is a perspective view of a hand. FIG. 7A is a perspective view of a modified example of the hand, and FIG. 7B is a side view of the modified example of the hand. FIG. 8A is a perspective view of another modified example of the hand, and FIG. 8B is a side view of another modified example of the hand. FIG. 9A is an explanatory view of the stacking type tableware placing portion, and FIG. 9B is an explanatory view of the stacking type tableware placing portion. FIG. 10 is an explanatory diagram of an AR marker engraved on the back surface of the tableware. FIG. 11 is an explanatory view of the back surface of the tableware after the AR marker is printed and then coated. It is explanatory drawing of the image recognition of the tableware which does not use the tableware marker. It is a top view of the tableware taking out part of Embodiment 4. 14A is a top view of the automatic dishwashing system, and FIG. 14B is a perspective view of the automatic dishwashing system. It is a figure which shows the area around the delivery table of the automatic dishwashing system.

Hereinafter, the tableware automatic washing system, the tableware automatic washing method, the tableware automatic washing program, and the storage medium according to the embodiment of the present invention will be described with reference to the drawings. However, the embodiments shown below exemplify an automatic dishwashing system for embodying the technical idea of the present invention, and do not specify the present invention as these, but are included in the scope of claims. It is equally applicable to those of the embodiment of.

[First Embodiment]
The tableware automatic washing system, the tableware automatic washing method, the tableware automatic washing program, and the storage medium according to the first embodiment of the present invention will be described with reference to FIGS. 1 to 9.

FIG. 1 is an example of a process diagram of the automatic dishwashing system according to the present embodiment. The tableware automatic washing system of the present embodiment includes a tableware input unit 1, a tableware cleaning unit 2, and a tableware extraction unit 3. Further, in the tableware automatic washing system of the present embodiment, a step S1 of removing residues and dust from the tableware, a step S2 of arranging the tableware side by side on a conveyor, a step S3 of washing with a water stream and a detergent, a step S4 of drying, and the tableware. Includes step S5 of moving the dish from the tableware to the storage location. In the tableware loading section 1, a step S1 for removing residues and dust from the dishes and a step S2 for placing the dishes side by side on a conveyor are executed, and in the dishwashing section 2, a step S3 for washing with a stream of water and a detergent and a step for drying. S4 is executed, and the tableware taking-out unit 3 executes the step S5 of moving the tableware from the conveyor to the storage place.

FIG. 2 is an example of an overall configuration diagram of the tableware automatic washing system according to the present embodiment. The tableware 11 includes, for example, Western tableware, Japanese tableware, shape, etc., for example, plates, balls, bowls, bowls, bowls, etc. for serving food, cups, cups, glasses, tea bowls, liquor for containing beverages. Bowls, chopsticks that handle food, forks, knives, spoons, madler, rangers, etc., cooking utensils or kitchen utensils such as pots, frying pans, cutting boards, kitchen utensils, knives, balls, cups, containers, frying bars, spatulas, ladle, etc. Includes all tools and utensils used for meals. The tableware 11 is transported from the tableware input unit 1 to the tableware take-out unit 3 via the tableware washing unit 2 by the tableware transport unit 10. The tableware transport unit 10 may be, for example, a single conveyor for transporting tableware from the tableware input unit 1 through the tableware washing unit 2 to the tableware take-out unit 3, but for example, a conveyor for the tableware input unit 1 and tableware. It is possible to divide the conveyor for the cleaning unit 2 and the conveyor for the tableware taking-out unit 3 into a plurality of conveyors, and further, it is possible to provide a plurality of conveyors in each unit. It is also possible to use one conveyor instead of a plurality of conveyors.

The tableware transport unit 10 is not particularly limited, but is, for example, a conveyor capable of transporting tableware. The types of conveyors include, for example, a) a flat conveyor that has a flat surface and the tableware is laid down on the conveyor, and b) a space under the conveyor surface where tableware can be inserted, starting from thin tableware. An underflight conveyor that can handle up to a washing rack, c) A flight conveyor that can handle dishes from thin dishes to a washing rack by leaning the dishes against the protrusions on the conveyor, d) A rack used to wash the dishes stored in the washing rack. It can be used for all types of conveyors such as conveyors.

The tableware take-out unit 3 is transported by an articulated robot 12 as the upper part of the tableware picker, a tableware storage part 13 for collectively placing the tableware picked up from the tableware transporter 10 at the upper part of the tableware picker, and a tableware transporter 10. A tableware information acquisition unit 14 is provided to identify tableware and acquire tableware information. In the tableware taking out unit 3, based on the tableware information from the tableware information acquisition unit 14, the tableware is picked up by the articulated robot 12 as the upper part of the tableware collecting, and the picked up tableware is put together in the tableware storage unit 13. To do. Here, the articulated robot 12 is illustrated as the upper part of the tableware picking, but the present embodiment is not limited to this, and for example, an industrial robot including a robot having a deformable arm, a lifting robot, and the like. If a mechanism for picking up tableware is provided, all means for picking up tableware are included. In the tableware placing section 13, the tableware picked up by the upper part of the tableware picker is placed together, for example, by type. As a method of placing the tableware together in the tableware placing section 13, for example, the same type of tableware is placed face down and stacked in the same direction. Since the tableware extraction unit 3 can grasp the type and posture of the tableware from the information from the tableware information acquisition unit 14, for example, the tableware is in a state where the long side and the short side of the rectangular tableware in a plan view are aligned. Can be placed on the tableware placing portion 13 in an overlapping manner. The type of tableware includes the size, shape, color, etc. of the tableware. Further, the articulated robot 12 can pick up the tableware from the tableware transport unit 10 even when the tableware is being transported by the tableware transport unit 10, that is, when the tableware is moving.

In the present embodiment, the explanation is given from the viewpoint of fully automating the tableware automatic washing system, but for example, it is also possible to manually perform a part of the tableware loading unit 1.

The tableware input unit 1 includes a tableware immersion unit, a tableware raising unit, and a washing pump, the tableware immersion unit has a tableware immersion sink and a jet nozzle, and the tableware raising unit is a tableware raising conveyor and a tableware alignment. It has a part and a strainer. Water is stored in the tableware immersion sink, and the water flow from the jet nozzle removes residues and debris from the tableware. The water filtered by the strainer is pressurized by the washing pump and ejected from the jet nozzle. The dishes from which residues and dust have been removed by the water flow from the jet nozzle are conveyed to the dishwashing unit 2 by the dish-raising conveyor. The washing pump is drawn at the bottom of the dishwashing section, the strainer is drawn at the bottom of the dishwashing conveyor, and the jet nozzle is drawn at the front of the dishwashing conveyor, but this is just an example, and the dish immersion unit. The arrangement of the washing pump, the strainer, and the jet nozzle can be changed as appropriate according to the shape and arrangement of the dishwashing unit and the dishwashing unit.

The tableware aligning unit is a means for aligning the types and postures of the tableware transported to the tableware washing unit on the tableware raising conveyor, and for example, an industrial robot such as an articulated robot can be used. As for the dishes that are scooped up by the tableware scooping conveyor, the dishes may overlap or the posture of the dishes may be unsuitable for washing the dishwashing unit 2. By arranging the arrangement and posture of the tableware, the tableware arranging unit can make it easier for the tableware washing unit 2 to wash the tableware and the tableware taking-out unit 3 for taking out the tableware easily. For example, when a large plate is transported face down on a tableware raising conveyor, a small plate, a bowl, or other relatively small tableware is placed under the large plate, and the dishwashing unit 2 is the subsequent process. Rough residue may occur in. Further, if the tableware is overlapped, the tableware information acquisition unit 14 may not acquire the tableware information, and in this case, the tableware extraction unit 3 may not take out the tableware. For this reason, if the platter is laid down on the tableware-raising conveyor, for example, the platter is arranged so that it stands against the protrusion on the tableware-raising conveyor so that the platter does not overlap with other dishes. ..

Further, if the tableware is aligned with respect to the transport direction of the conveyor, for example, if the tableware is arranged so that the back surface (thread bottom) of the tableware faces the traveling direction of the conveyor, each tableware is appropriately arranged in the dishwashing unit 2. In addition to being able to wash the dishes, the tableware taking-out unit 3 can easily take out each tableware. Further, as will be described later, by loading the tableware in the tableware loading section 1 according to a predetermined rule for the type, position, and posture of the tableware, the image recognition of the tableware type, position, and posture in the tableware taking-out section 3 It can improve the accuracy of tableware, reduce the failure of tableware picking, and increase the speed of tableware picking.

The arrangement of the tableware input unit 1 and the tableware extraction unit 3 is adjusted according to the type of the tableware washing unit 2, that is, according to the arrangement of the tableware input position and the tableware removal position. Therefore, the overall configuration diagram of the automatic dishwashing system shown in FIG. 2 is an example of the present embodiment and describes the linear type, and the present embodiment is not limited to the layout shown in this drawing. In addition, a layout corresponding to various other types of dishwashing parts 2 such as a corner type and a land rip is possible.

FIG. 3 shows an example of a block diagram of the tableware taking-out unit 3 according to the present embodiment. The tableware extraction unit 3 includes an image acquisition unit as a tableware information acquisition unit 14, a tableware extraction calculation unit 30, and an articulated robot 12. The tableware extraction calculation unit 30 can be configured by, for example, a personal computer (hereinafter referred to as “PC”), a tablet terminal, or the like, and includes a tableware position / posture detection unit 31, a tableware extraction order planning unit 32, and a tableware extraction operation command. It has a unit 33. The tableware position / posture detection unit 31 acquires information on the position, posture, type, etc. of the tableware from the tableware information acquisition unit 14.

The tableware take-out order planning unit 31 issues a command to the tableware transport unit 10 to operate or stop the conveyor, and also issues a command to the tableware input unit 1 to operate or stop the conveyor according to the operation of the conveyor.
If the tableware position / orientation detection unit 31 is connected to the tableware loading unit 1, the tableware transporting unit 10, the tableware washing unit 2, etc. in addition to the tableware information acquisition unit 14, and can communicate with each unit, for example, tableware insertion. The tableware position / posture detection unit 31 provides information on the tableware position or posture when the tableware is put in from unit 1, information on the tableware transport speed by the tableware transport unit, information on the strength of the washing water flow by the tableware washing unit, etc. It can also be used to acquire information on posture and type.

The tableware removal operation command unit 33 outputs a command for the tableware removal operation to the articulated robot 12, and based on this command, the articulated robot 12 picks up the tableware from the tableware transport unit 10 and puts the tableware together on the tableware storage unit. Place.

FIG. 4 is an example of an overview diagram of the tableware taking-out unit 3 according to the present embodiment. FIG. 4A is a plan view of the tableware taking-out section 3, and FIG. 4B is a perspective view of the tableware taking-out section 3. In the tableware taking out unit 3, based on the tableware information from the tableware information acquisition unit 14, the tableware transported by the tableware transporting unit 10 by the articulated robot 12 as the tableware collecting upper part is picked up, and the picked up tableware is picked up. They are collectively placed on the mounting unit 13. Here, in FIGS. 4A and 4B, the arrangement of the articulated robot 12, the tableware transporting unit 10 and the tableware placing unit 13 is an example of the present embodiment, and for example, the articulated robot 12 sandwiches the tableware transporting unit 10. A tableware carrier 10 may be provided on the opposite side. In this way, by setting the distance between the position where the articulated robot 12 picks up the tableware and the tableware placing portion 13 on the tableware transporting unit 10 is set as short as possible, the operation of the tableware collecting upper portion 12 can be speeded up.

At the end of the tableware transport unit 10 on the advancing side, a tableware receiving unit (not shown) for receiving the tableware that the articulated robot 12 could not pick up is provided. Since each joint of the articulated robot 12 has a movable limit, there is a limitation due to the movable limit when displacing from one posture to the next posture. Depending on the posture of the tableware, it may not be possible to pick up the tableware due to the movable limit of the articulated robot 12. Further, if the tableware information acquisition unit 14 cannot properly read the AR marker 42, for example, if the tableware overlaps with each other, the articulated robot 12 may not be able to pick up the tableware. Therefore, by providing a tableware receiving unit at the advancing end of the tableware transporting unit 10, even if the articulated robot 12 cannot pick up the tableware, the tableware can be safely received and the tableware is automatically installed. Redundancy can be added to the cleaning system.

FIG. 5 is an example of an explanatory diagram of a usage example of the marker according to the present embodiment. In this embodiment, a marker, for example, an AR marker 42 is provided on the back surface (thread bottom) of the tableware, and the tableware is recognized by acquiring an image of the tableware with a tableware detection camera as the tableware information acquisition unit 14 and recognizing the image. In addition, an example of acquiring information on the position, posture, and type of tableware will be described.

As shown in FIG. 4, the tableware detection camera as the tableware information acquisition unit 14 is, for example, above the tableware transport unit 10, upstream of the tableware collector upper portion 12, and on both sides of the tableware transport unit 10. Is provided so as to face the center of the tableware transport unit 10 in the width direction. On the back surface (thread bottom) of the tableware, a marker different for each type of tableware, for example, an AR marker 42 is provided. The marker of the present embodiment is not limited to the AR marker 42, and any marker can identify at least one of the type of tableware, the position of the tableware, and the posture of the tableware. Markers may be used.
Although FIG. 4 illustrates that two tableware information acquisition units 14 are provided, the number of tableware information acquisition units 14 is not limited to two and may be one or more. For example, four tableware information acquisition units 14 may be provided. When the four tableware information acquisition units 14 are provided, the tableware information acquisition units 14 are provided at positions that are the same height as each of the two tableware information acquisition units 14 shown in FIG. 4 and are separated from each other in the transport direction of the conveyor. For example, when the four tableware information acquisition units 14 are arranged at positions corresponding to each corner of the quadrangular line and the orientation is adjusted so that the optical axes are gathered in a predetermined area at the center in the width direction of the conveyor, the four tableware information The acquisition unit 14 can acquire images of tableware from both the forward and reverse directions of the transport direction.
That is, when the four tableware information acquisition units 14 are provided, the optical axis of the tableware 11 that deviates from the rule of arranging the tableware 11 so that the back surface (thread bottom) of the tableware 11 faces the traveling direction of the conveyor is arranged on the traveling direction side of the conveyor. An image of the back surface (thread bottom) of the tableware 11 can be acquired by the tilted tableware information acquisition unit 14.

Regarding the AR marker 42, "The AR marker is an image having a fixed pattern that serves as a marker for designating a position for displaying additional information in an image recognition type AR (Augmented Reality) system." ( Website IT Glossary e-Words). By using the AR marker 42, in the image detected by the tableware detection camera, the image of the tableware changes depending on the viewing angle and distance, but if the AR marker 42 is used, the digital information added to the AR marker 42 can be obtained. It can be read out, and the posture of the AR marker 42, that is, the posture of the tableware with the AR marker can be detected. Further, if the type of tableware is coded in the AR marker 42, the position, posture, type, etc. of the tableware can be detected as above the tableware by photographing the AR marker with the tableware detection camera. The position information of the tableware acquired by photographing the AR marker 42 includes the center position of the back surface (thread bottom) of the tableware.

A hand 40 is attached to the tip side of the arm of the articulated robot 12, and a suction portion 41 for sucking tableware is provided on the tip side of the hand 40.

FIG. 6 shows an example of a perspective view of the hand 40 according to the present embodiment. The hand 40 includes a extending portion 45, a hand mounting portion 46 on one end side of the extending portion 45, a suction portion 41 on the other end side of the extending portion 45, and a vacuum hose 47. The hand mounting portion 46 is mounted on the arm side of the articulated robot 12. The piping of the vacuum hose 47 extends to the suction portion 41 through the extension portion 45, and the suction portion 41 can suck the back surface (thread bottom) of the tableware by vacuum. The other end of the vacuum hose is attached to a robot-side vacuum hose attachment portion (not shown) provided on the articulated robot 12, and the suction portion 41 is evacuated by the robot-side vacuum hose attachment portion. ..

When the articulated robot 12 sucks and transports tableware, if large or heavy tableware is transported at high speed, vibration will increase and subsequent stacking may not be successful. The cause of this vibration is that the suction portion 41 of the hand 40 deforms after sucking the tableware. A ring-shaped member having a higher rigidity than the pad is installed so as to surround the suction portion 41, and the tableware is sucked by an internal force at the time of suction to suppress the deformation of the hand 40, so that the tableware is sucked by the articulated robot 12. The vibration of tableware during transportation can be significantly reduced.

FIG. 7 shows an example of a perspective view of another embodiment of the hand 40A according to the present embodiment. The hand 40A includes an extension portion 45, a hand attachment portion 46 on one end side of the extension portion 45, a coupling portion 49 on the other end side of the extension portion 45, and a first suction portion 41A and a second suction portion 41A provided on the coupling portion. A suction portion 41B and a hand-side vacuum hose attachment portion 48 are provided. A vacuum hose (not shown) is attached to the hand-side vacuum hose attachment portion 48, and the other end of the vacuum hose is attached to the robot-side vacuum hose attachment portion (not shown). The piping from the hand-side vacuum hose attachment portion extends to the suction portions 41A and 41B through the extension portion 45, and the suction portions 41A and 41B can suck the back surface (thread bottom) of the tableware by vacuum. Both suction portions 41A and 41B are provided in the coupling portion 49 at intervals of approximately 90 degrees. Since the suction portions 41A and 41B are provided at two locations, the articulated robot 12 can pick up the tableware without taking an unreasonable posture. In addition, this can reduce the rate of failure to take out tableware.

FIG. 8A is a perspective view of another modified example of the hand, and FIG. 8B is a side view of another modified example of the hand. In FIG. 8, the same reference numerals are used for the same configuration as in FIG. 7, and the description thereof will be omitted.

The articulated robot 12 may pick up the tableware 11 at a position other than the tableware 11 due to an error in recognition of the tableware 11 (position and posture of the tableware 11). For example, the articulated robot 12 may pick up the tableware 11 at a position behind the tableware 11.
At that time, a part of the hand 40 such as the suction unit 41 may come into contact with the conveyor and other members of the tableware transport unit 10 and press the hand 40 against the conveyor without being able to pick up the tableware 11. There was a risk of damaging the conveyor, tableware 11, or the hand 40.

Conventionally, a force sensor (6-axis sensor) is provided on the hand 40, the force applied to the hand 40 is detected by the force sensor, and when an excessive external force is applied to the hand 40, the articulated robot 12 operates. Was stopped.
However, there is a demand for the force sensor to be omitted because it is vulnerable to water wetting, is easily broken by impact force, and is expensive.

The hand 40C includes a plurality of spring mechanisms 60 and a ball joint 61. The hand 40C also includes a force sensor 62.
The plurality of spring mechanisms 60 and the ball joint 61 reduce the reaction force (external force) received by the hand 40C from the pressing destination of the conveyor or the like and the pressing force received by the conveyor and / or the tableware 11 from the hand 40C.
The plurality of spring mechanisms 60 are provided between the hand mounting portion 46, which is a circular plate-shaped portion, and the circular plate-shaped member arranged to face the hand mounting portion 46.

The spring mechanism 60 reduces the external force by elastically deforming the elastic portion between the hand mounting portion 46 and the circular plate-shaped member in response to the external force when the hand 40C is pressed against the conveyor or the like. is there.
By appropriately setting the spring constant, the spring mechanism 60 does not deform in an elastically neutral state where no external force is applied, and is elastically deformed only when an excessive external force is applied. There is.

The ball joint 61 is arranged at the center position of the hand mounting portion 46. As a result, the elastic force of the spring mechanism 60 is applied only in the direction of rotation around the ball joint 61.
Therefore, even if the suction portions 41A and 41B are located away from the wrist (hand mounting portion 46) of the hand 40C and a moment is applied to the wrist larger than the translational force, the hand 40C can be moved in the translation direction. The spring mechanism 60 elastically deforms and imitates the moment received from the suction portions 41A and 41B while preventing unnecessary deflection.

By using this hand 40C, even if a part of the hand 40C such as the suction unit 41 is pressed against the conveyor without being able to pick up the tableware 11, the reaction force received by the hand 40C from the pressing destination of the conveyor or the like. (External force) and the pressing force received by the conveyor and the tableware 11 from the hand 40C can be reduced, so that the conveyor, the tableware 11 and the hand 40C can be prevented from being damaged.

In the present embodiment, the elastic force is applied only in the direction of rotation around the ball joint 61, but the elastic force may also be applied in the axial direction of the ball joint 61. Absent.

FIG. 9 is an explanatory view of the tableware placing part 13 according to the present embodiment, FIG. 9A is an explanatory view of the stacking type tableware placing part, and FIG. 9B is an explanatory view of the stacking type tableware placing part. In the case of the stacking type tableware placing portion 13 of FIG. 9A, the upper surface of the tableware placing portion 13 is set to be flat. Based on the tableware type information acquired by the tableware information acquisition unit 14, the articulated robot 12 stacks the tableware at a predetermined place in the tableware storage unit 13 for each type of tableware. Since the tableware extraction calculation unit 30 counts how many dishes are placed in the tableware storage unit 13 for each type of tableware, the height at which the tableware is stacked according to the type and number of tableware is predicted. It is possible to do. Further, the tableware placing portion 13 may be provided with a means for detecting the presence or absence of tableware according to the position and a means for detecting the height at which the tableware is stacked. For example, the tableware for photographing the tableware placing portion 13. By analyzing the tableware image acquisition unit (not shown), the position, type, and number of tableware placed on the tableware storage unit 13 can be detected.

In the case of the stacking type tableware placing portion 13 of FIG. 9B, a position for stacking the tableware is determined for each type, and a means for raising and lowering according to the weight of the tableware is provided at each stacking position. Therefore, the tableware to be stacked next can be adjusted so as to be located on the upper surface of the tableware placing portion 13. As a result, since the articulated robot 12 knows in advance the place and height at which the tableware is to be stacked, it is possible to accurately stack each tableware. If no means for raising and lowering according to the weight of the tableware is provided at each stacking position, the tableware extraction calculation unit 30 indicates how many tableware are placed in the tableware storage unit 13 for each type of tableware. By providing a means for detecting the height at which the tableware is stacked, each tableware can be placed up to a predetermined stacking height. Further, when an appropriate means for detecting the height on which each tableware is placed is provided, it is also possible to instruct the articulated robot 12 on the stacking position of each tableware.

[About the tableware information acquisition department]
In order to stack tableware on the tableware storage unit 13 for each type of tableware, different AR markers are provided on the back surface (thread bottom) of the tableware for each type. By providing the AR marker on the tableware, the articulated robot 12 can pick up the tableware even if the position and posture of the tableware are not constant. By using the inputs of the plurality of tableware detection cameras as the tableware information acquisition unit 14 and tracking the individual AR markers by the Kalman filter, the accuracy of recognizing the position and orientation of the tableware can be improved. Further, even when the tableware transport unit 10 is in operation, the articulated robot 12 can pick up the moving tableware.

[About the upper part of the tableware]
There are innumerable postures of the robot for picking up tableware by the articulated robot 12, but the postures that the robot can actually take or the postures that are easy to take are limited. If the posture of the robot is not properly selected, the robot will not be able to move, or it will stop because it is too close to the movable limit. For example, if the posture of the robot takes a singular point, the robot may stop. Therefore, using the operability, which is an index of the mobility of the articulated robot 12, and the proximity to the movable limit, the robot selects a posture that is easy to move from among several posture candidates and picks up the tableware. ..

Normally, the movement of the articulated robot 12 tries to move the shortest distance from the start to the goal, but if the shortest distance is moved when picking up tableware, the joints may exceed the movable range. Therefore, the movement of the articulated robot 12 from the start to the goal is simulated in advance, and the angle of the joint that may exceed the movable range is adjusted in advance to an angle close to the shortest distance to the goal before starting. As a result, it is possible to prevent a situation in which a specific joint exceeds the range of motion before reaching the posture to the goal.

When stacking tableware on the tableware storage unit 13, if the height does not match, the tableware may drop and make noise, or the tableware may be damaged by pressing the tableware. Therefore, a force sensor is provided between the arm of the articulated robot 12 and the hand 40 to detect the force with which the tableware is pressed and prevent the tableware from falling. By using the compliance control, the tip of the arm or the hand 40 of the articulated robot 12 is given a virtual spring property so that the tableware will not be damaged even if it is pressed against it. The force sensor detects the direction in which the stacking method is misaligned, and follows that direction.

If the tableware cannot be sucked when the tableware is recognized incorrectly, the tableware or tableware transporting unit 10 may be touched and pressed by a device other than the suction unit 41, which may cause a failure of the articulated robot 12 or the tableware transporting unit 10. There is a possibility of breaking the sex and tableware. While using compliance control, the force sensor detects the direction in which the stacking method is misaligned, and at the same time, if a force that exceeds the set hand force threshold is applied, abnormal processing (tableware is overlooked or tableware is overlooked) is performed as a pickup failure. , Processing such as measuring with a sensor again and measuring the correct position and orientation) makes it possible to continue the processing.

When stacking tableware on the tableware storage unit 13, if the position of the tableware storage unit 1 is misaligned or the position and orientation of the tableware placed earlier is misaligned, a force to press the tableware is generated and the tableware is tableware. There is a possibility that the tableware may be damaged or the tableware may bounce and fail to be placed. Therefore, the probability of these failures can be reduced by using compliance control. In addition, since the direction in which the hand shifts can be known by compliance control, it is possible to estimate and eliminate the position / posture shift itself by moving the target value of the hand there, that is, by imitating it.

[Forecast]
When detecting an AR marker, as a general characteristic of the camera, the accuracy in the optical axis direction (depth direction) of the camera deteriorates. If the position of the AR marker is not accurate, the position where the robot picks up the tableware may shift and picking may fail. In addition, if the direction of the AR marker cannot be seen well from the camera or if it is hidden, the tableware may not be detected. Therefore, multiple cameras are used to track the AR marker. A Kalman filter is used to track the AR marker. By mixing and estimating the positions and orientations of markers measured from multiple cameras with a Kalman filter, even if there is a moment when the AR markers cannot be measured from the cameras, the positions and orientations can be tracked without losing sight of the AR markers. be able to.

The following two improvements were made to make more accurate estimates using the Kalman filter.
(A) The magnitude of the error (covariance matrix of the error) can be set as a parameter in the Kalman filter. Since it is known that the error in the optical axis direction of the camera is large in principle, setting a large value of the error in the optical axis direction of the camera improves the position and orientation estimation system.
In the Kalman filter, other parameters such as the state quantity, for example, (a) the size of the marker is different from the assumption, (b) the attitude of the camera is slightly different from the assumption, (c) the direction of the conveyor is slightly delicate. It is possible to estimate the amount of deviation. By using these state quantities, the position and orientation of the marker can be estimated more accurately.

Regarding the placement of the camera, it is known that the measurement accuracy is poor in the direction of the optical axis of the camera, so by setting the direction of the optical axis of the camera as different as possible, it is possible to prevent a bias in the deterioration of accuracy. Therefore, the two cameras as the image acquisition unit are arranged so that the optical axis faces the center of the conveyor from both sides of the conveyor.

There are certain rules for putting dishes into the dishwashing department, so you can predict the position of the dishes when you take them out according to these rules. Further, when the marker information is read when the tableware is put into the dishwasher, the arrangement of the tableware on the conveyor is known in advance, so that the position where the tableware is picked up can be predicted accurately.

In addition, the procedure for picking up tableware is as follows: 1) Measure the markers with multiple cameras, estimate their positions with the Kalman filter, and 2) When the markers are out of the field of view of the cameras, the amount of movement of the conveyor (with the encoder installed on the conveyor) The procedure is that the robot continues to estimate the position of the marker after movement from the directly measured movement amount), and 3) the robot picks up the tableware that should be there based on the estimated position. If the tableware hits something else on the conveyor, or if the tableware is unstable on the conveyor, the tableware may move while the camera is not observing the marker. It is most certain that you will continue to observe the markers with your camera until just before you pick up the dishes. If the tableware can always be kept within the field of view of the camera, the accuracy of measuring the position of the tableware will be further improved. However, the Kalman filter itself is necessary to improve the accuracy of observation using multiple cameras, even if the tableware is always in the field of view. If the camera is brought close to the robot in order to have the camera's field of view close to the robot's operating range, the robot may hit the camera, so it is necessary to devise the overall configuration and the arrangement of the cameras.

[About learning]
When the tableware is put in by a robot, the rules for putting the tableware can be corrected at any time in response to the pick-up failure case. By learning, the rules for throwing tableware can be optimized so that the robot does not take an unreasonable posture. For example, it is possible to optimize the tableware insertion rule by learning, for example, reinforcement learning, how to insert tableware suitable for pickup, how to insert regular tableware, or how to insert tableware in an easy-to-take orientation. ..

Reinforcement learning is a method of setting a reward for the result of a robot's action and acquiring an action that raises the reward as much as possible. As an example of the reward, the success or failure result of tableware pickup is judged by the pressure sensor of vacuum suction, and the reward is given when the tableware can be picked up. It is effective to input the observed tableware image and perform reinforcement learning to acquire the correction amount (how to determine) of the trajectory at the time of picking up the robot so as to maximize this reward.
In addition, since it is necessary to perform reinforcement learning while actually operating the tableware automatic washing system of the present embodiment, a predetermined number of trials are required. Therefore, until the reinforcement learning progresses and the pickup system is improved, the tableware and the arm It is effective to adopt the above-mentioned compliance control in order to prevent damage or failure due to collision with.

Because the posture of the robot is restricted, it may not be possible to move the tip of the arm in the shortest distance. Learning to optimize the correction amount of the trajectory of the arm tip of the articulated robot 12 and which tableware to pick up so that the next tableware can be picked up smoothly following the operation of picking up the current tableware. Is also possible. For example, using learned AI enables robots to quickly pick up dishes with smooth movements like a skilled person from the beginning of the introduction of the dishwasher. For example, it is effective to put a correction amount in the trajectory when the robot is picked up and optimize the correction amount by reinforcement learning in the direction in which the pickup works, or supervised learning by DNN (deep neural network). It also contributes to speeding up the robot.

[Embodiment 2]
The tableware automatic washing system, the tableware automatic washing method, the tableware automatic washing program, and the storage medium of the second embodiment of the present invention will be described with reference to FIGS. 10 to 11. In this embodiment, a plurality of modified examples of the markers used in the first embodiment will be described.

[Modification 1]
In the first modification, a method of attaching a sticker on which an AR marker is printed will be described. Although not particularly limited, for example, (a) a sheet member in which an AR marker is printed with black ink on a white sheet member,
(B) A sheet member in which an AR marker is printed with black ink on a transparent sheet member,
(C) A sheet member in which an AR marker is printed with white ink on a black sheet member, or
(D) A sheet member in which an AR marker is printed with white ink on a transparent sheet member,
Any one of the sheet members is attached to the back surface (thread bottom) of the tableware with an adhesive or an adhesive. Here, the colors of the sheet member and the ink are not limited to transparent, white or black, and are a combination of colors that can read the AR marker when the printed sheet member is attached to the back surface of the tableware. All you need is. By adjusting the image recognition software, it is possible to handle various color combinations, but since the color of tableware is generally white or black, black ink is used on a white background and white ink is used on black characters. The combination of is often used. In this case, for the black tableware, the black and white inverted pattern of the marker pattern for the white tableware can be used. The sheet member of (b) is used for tableware of a light color such as white that can identify black ink, and the sheet member of (d) is used for tableware of a dark color such as black that can identify white ink.

As the waterproof sheet, it is desirable to use, for example, a sheet member having a waterproof film or a waterproof coating and having high water resistance and heat resistance. It is desirable to use an adhesive or an adhesive for attaching the sheet member to the tableware, which has high water resistance and heat resistance so that it can be repeatedly washed and used. In addition, the shape and material of the tableware may be any as long as the sheet member can be attached. The AR marker needs to be provided on a flat surface such as a thread bottom on the back surface of tableware. Since the sheet member can be attached to the existing tableware later, there is no need to prepare special tableware.

It has been explained that a waterproof sheet or coating having high water resistance and heat resistance is used, but this embodiment is not limited to this, and the tableware is repeatedly washed and used by serving food. For example, a tarpaulin or coating should be chemically and physically durable. That is, it has resistance to various foods such as alkali resistance and acid resistance, and also to detergents, is hard to peel off even after repeated washing, and has durability against friction during use and washing. It is desirable to have.

[Modification 2]
The second modification will be described with reference to FIG. The second modification is a method of engraving the AR marker 42 on the back surface of the tableware by laser processing. The tableware used in the second modification may be made of a material capable of forming a groove by laser processing, and may be made of melamine, for example. The shape of the tableware may be any shape as long as it is a flat portion on which an AR marker is provided, for example, a tableware having a flat thread bottom on the back surface.

Laser processing a hatched shape, that is, a plurality of substantially parallel linear grooves, for example, at intervals of about 0.3 mm on the part of the back of the tableware where the ink corresponding to the AR marker is to be filled. Fill this groove with ink of a color different from the color of the tableware. The hatching interval is not particularly limited, and may be narrower than, for example, about 0.3 mm, or may be wider than this. It is desirable that the ink used is water resistant and heat resistant. Since the ink packed in the groove is held in the groove and has durability, it is not necessary to provide a coating or the like. However, in order to further improve the durability, a waterproof film or a preventive coating having water resistance and heat resistance may be provided on the AR marker portion. Since the existing tableware can be laser-processed later, there is no need to prepare special tableware.

FIG. 10A is an example of tableware on a white background filled with black ink. The size of the marker is not particularly limited, but can be about 20 mm square, and may be smaller, such as about 15 mm square to accommodate smaller tableware, and further about 30 mm square. It does not matter if the size is larger than the above. FIG. 10B is an example of black tableware filled with white ink. The color of the tableware and the color of the ink are not limited to black or white, and any combination of colors that can read the AR marker 42 may be used. By adjusting the image recognition software, it is possible to support various color combinations and markers of various sizes. It is desirable to unify the size of the AR marker in order to simplify the image recognition algorithm and improve the recognition accuracy of the position or posture.

[Modification 3]
A modified example 3 will be described with reference to FIG. The third modification is a method in which the AR marker 42 is directly printed on the back surface of the tableware and then coated. First, a method of providing an AR marker in the manufacturing process of melamine tableware will be described. In the process of manufacturing tableware made of melamine, an AR marker is printed with an appropriate ink, and a coating, for example, a melamine coating is applied to the surface thereof. The material of the coating is not limited to melamine, but when the material of the tableware is, for example, melamine, it is desirable that the coating on the surface of the AR marker is also a melamine coating. The ink is not particularly limited, but for example, an ultraviolet coloring ink, an ultraviolet discoloring ink, an ultraviolet curing ink, and the like can be used.

Although the melamine tableware has been described as an example, the AR marker printing method of the modified example 3 is not limited to the melamine tableware, and can be applied to various tableware such as pottery, glass, and resin. , Appropriate coating according to the material, for example, resin coating shall be applied. In addition, the AR marker can be printed later on the existing tableware. In this case, it is not necessary to prepare special tableware because the existing tableware can be laser-processed later. Moreover, since the AR marker is coated, it has high durability.

FIG. 11A is an example of printing black ink on white tableware. Depending on the material of the tableware, the specifications and condition of the ink, and the specifications and condition of the coating, the color of the black ink may become lighter, but even if the color of the black ink becomes lighter due to the adjustment of the image recognition software, the AR marker Can be recognized. FIG. 11B is an example of printing white ink on black tableware. Depending on the material of the tableware, the specifications and condition of the ink, and the specifications and condition of the coating, the color of the white ink may be slightly brownish, for example, but the AR marker can be recognized by adjusting the image recognition software. The combination of the tableware color and the ink color of the modification 3 is not limited to black or white, and various other color combinations are possible as long as the AR marker can be identified, and the image recognition software. By adjusting the wear, AR markers with various color combinations can be recognized.

[Modification example 4]
In the fourth modification, a method of adhering a tag having an AR marker to tableware will be described. For example, an AR marker is printed on a coin-shaped (thin circular shape) tag, and the tag is attached to the back surface of tableware, for example, with an adhesive or adhesive, with the printed surface of the tag as the front surface. It is desirable to provide a waterproof film or preventive coating having water resistance and heat resistance on the printed surface of the tag. Further, it is desirable to use an adhesive or an adhesive for attaching the tag to the tableware, which has water resistance and heat resistance. The specifications such as the shape and material of the tag of the modification 4 are not limited to the coin shape, and any specifications may be used as long as the tag has an AR marker. Since the tag with the AR marker is attached to the existing tableware later, there is no need to prepare special tableware.

[Modification 5]
In the fifth modification, the type of marker and the position of the marker will be described. In the modified examples 1 to 4, an example in which an AR marker is used as a marker has been described, but in the present embodiment, the marker is not limited to the AR marker, and either the type of tableware, the position of the tableware, or the posture of the tableware. Any marker may be used as long as it can identify at least one marker.

Further, the marker can be arranged on the back side of the tableware, for example, in the center of the thread bottom, but the present embodiment is not limited to this. For example, if the marker is at a predetermined position for each type of tableware, the position for picking up the tableware can be grasped by registering the position in advance in the image recognition software.

[Embodiment 3]
The automatic dishwashing system, the automatic dishwashing method, the automatic dishwashing program, and the storage medium of the third embodiment of the present invention will be described with reference to FIG. In FIG. 12, the same reference numerals are used as those described in FIGS. 1 to 11, and the description thereof will be omitted. In the first and second embodiments, the tableware information acquisition unit 14 recognizes the tableware using the AR marker, but in the second embodiment, the example of recognizing the tableware without the AR marker will be described.

By recognizing the tableware from a plurality of tableware detection cameras as the tableware information acquisition unit 14 and the tableware by a three-dimensional point cloud (for example, obtained from a plurality of camera images), the tableware can be displayed without the AR marker. The position and posture can be recognized. Although not particularly limited, for example, a stereo vision depth camera capable of depth measurement of the Intel RealSense Depth Camera D400 series can be used. Equipped with an image sensor set (2 depth sensors, maximum resolution 1280 x 720) that can detect the difference between two images and perform real-time depth image processing, can scale color data, and is equipped with an IR projector. , Works with USB power supply. For example, in order to obtain depth data more accurately when observing a plain white wall, a random dot pattern of infrared light is projected by an active IR projector that illuminates the object, and the pattern is displayed on the apparent object. Is added to make it easier to match the two camera images.

In addition to the stereo camera method described above, as a method of obtaining a three-dimensional point group, 1) a method of irradiating a pattern such as a laser beam or an LED, photographing the appearance of the pattern, and taking a three-dimensional point group (Structured Light). Method), 2) A method of irradiating light such as laser light or LED and measuring the time until it is reflected and returned (Time of Flight method), and 3) Searching for the position where the image is in focus. A method of knowing the depth information (focusing method) or the like can also be used.

Further, for example, as shown in FIG. 12, if DNN is used for image recognition of tableware, it is possible to cut out an image or a point cloud area including only the tableware, so that the detection accuracy of the position and orientation of the tableware is also improved. .. In FIG. 12, the recognized tableware is displayed by the detection frame 50, and the type is further displayed as the type display 51.

In the third embodiment, images from a plurality of tableware detection cameras and a method of recognizing tableware by a three-dimensional point group are illustrated, but the present embodiment is not limited to this, and for example, two-dimensional. A method of detecting tableware using an image, a method of recognizing tableware by learning control, a method of detecting a thread bottom from an image of tableware, a method combining these, and the like can also be applied. For example, since it is possible to detect the position and orientation of the tableware even from a two-dimensional image, the tableware is image-recognized without the AR marker, and the thread bottom of the tableware is attached to the suction portion of the articulated robot 12 based on this image recognition. It can be picked up by being sucked by the hand 40 having 41.

[Embodiment 4]
The automatic dishwashing system, the automatic dishwashing method, the automatic dishwashing program, and the storage medium of the fourth embodiment of the present invention will be described with reference to FIG. FIG. 13 is a plan view of the tableware taking-out portion of the fourth embodiment. In FIG. 13, the same reference numerals are used for the same configurations as those in FIGS. 1 to 12, and the description thereof will be omitted. In the fourth embodiment, an example in which the container housing robot 56 is used together with the articulated robot 12 will be described.

When the tableware 11 is picked up from the tableware transport unit 10 and stored in the container 57 by one articulated robot 12, for example, a 6-axis articulated robot, the movable range of the articulated robot 12 is limited. , It is difficult to horizontally stack and store tableware in the container 57. Therefore, if all the operations of picking up the tableware 11 from the tableware transport unit 10 and storing it in the container 57 are to be realized by one articulated robot 12, the device will become large.

In FIG. 13, the container storage robot 56 is used together with the articulated robot 12 to pick up the tableware 11 from the tableware transport unit 10 and store the tableware 11 in the container 57. The articulated robot 12 picks up the tableware 11 from the tableware transporting unit 10 and places the tableware 11 on the delivery table 55 in a state of lying down, that is, in a posture in which the thread bottom faces upward. The tableware 11 placed on the delivery table 55 is housed in the container 57 by the container storage robot 56. The container accommodating robot 56 is not particularly limited, but for example, a horizontal articulated robot can be used.

The articulated robot 12 and the container housing robot 56 have means for picking up the tableware 11. For example, the articulated robot 12 has a hand 40 including a suction unit 41, and the suction unit 41 can suck the thread bottom of the tableware 11 to pick up the tableware 11. Further, for example, the container accommodating robot 56 also has a suction portion 41 on the tip end side of the arm, and the suction portion 41 can suck the thread bottom of the tableware 11 to pick up the tableware 11.

The articulated robot 12 picks up the tableware 11 transported by the tableware transport unit 10 and places the tableware 11 on the delivery table 55. The container accommodating robot 56 picks up the tableware 11 placed on the delivery table 55 and stores it in the container 57. In FIG. 13, four containers are arranged within the range in which the container storage robot 56 can store tableware, but the number of containers is not limited to this, and the range in which the container storage robot 56 can store tableware. One or more are placed inside. The container accommodating robot 56 can store the tableware 11 in a predetermined container according to the type of the tableware 11. For example, one type of tableware 11 can be stored in one container.

The articulated robot 12 and the container accommodating robot 56 have places where their movable areas overlap each other, such as around the delivery table 55. Therefore, the articulated robot 12 and the container accommodating robot 56 are cooperatively controlled so as not to interfere with each other. In the present embodiment, the roles of the actions of the articulated robot 12 and the container accommodating robot 56 are clearly separated, and the scenes of interference in the actions are limited. Therefore, for example, a simple interlock control or the like can be applied to the cooperative control of the present embodiment, and does not necessarily require complicated control, and various control methods are applied to the cooperative control. It is possible.

When the tableware 11 is stored in the container 57, the outer shape of the tableware, that is, the width when stacked flat (for example, the radius in the case of circular tableware, the length of one side (particularly the long side) in the case of square tableware, etc. ) Is not surely acquired, the tableware 11 may come into contact with each part of the tableware automatic cleaning system and be damaged. If the outer shape of the tableware of the tableware 11 cannot be reliably acquired, for example, when estimating the size of the tableware by image recognition, an image acquisition unit is added or a computer for image processing is required, resulting in high cost. turn into.

Therefore, an abutment member having a predetermined shape is provided at a predetermined position on the delivery table 55. When the articulated robot 12 lowers the tableware 11 onto the delivery table 55 in a flat position, it horizontally moves until the side surface of the tableware is pressed against the counterbore member, and detects a position in contact with the counterbore member. For this detection, for example, a force sensor, a vibration sensor, an acceleration sensor, a load current sensor of the drive motor of the articulated robot 12, or an image acquisition unit provided on the arm tip of the articulated robot 12 or the hand 40 or the like. Etc., various sensors can be used. Since the position of the abutment member is known in advance and the locus of the articulated robot 12 can be calculated, the outer shape of the tableware can be calculated from the position where the tableware 11 taken up by the articulated robot 12 comes into contact with the abutment member. .. Further, if the outer shape of the tableware 11 can be calculated in this way, even if the dimensions of the tableware 11 are not registered, the storage of the tableware 11 in the container 57 can be automated.

[Embodiment 5]
The automatic dishwashing system, the automatic dishwashing method, the automatic dishwashing program, and the storage medium of the fifth embodiment of the present invention will be described with reference to FIG. 14A is a top view of the automatic dishwashing system, and FIG. 14B is a perspective view of the automatic dishwashing system. In FIG. 14, the same reference numerals are used for the same configurations as those in FIGS. 1 to 13, and the description thereof will be omitted. In the fifth embodiment, an example in which the container accommodating robot 56 is used together with the articulated robot 12 will be described as in the fourth embodiment.

When the articulated robot 12 is placed beside the tableware transport unit 10 (see FIG. 13), the hand 40 of the articulated robot 12 is located in the area around the end of the tableware transport unit 10 on the side opposite to the installation side of the articulated robot 12. It becomes difficult to reach, and the range in which the posture can be taken is narrowed. As a result, if the articulated robot 12 is arranged beside the tableware transport unit 10, it may be difficult to pick up the tableware 11 depending on the transport area of the tableware 11 of the tableware transport unit 10.

Also, if the articulated robot 12 is placed beside the tableware transport unit 10, the entire system becomes large and space efficiency deteriorates.

The tableware automatic washing system of the present embodiment includes a transport unit straddling stand 70 for installing the articulated robot 12 at a position straddling the tableware transport unit 10 in the width direction of the tableware transport unit 10.

The transport section straddling pedestal 70 is a gate type, and the articulated robot 12 is fixed to the wall bridging the upper part. In the present embodiment, the articulated robot 12 is fixed to the transport unit straddle stand 70 so that the movable range of the articulated robot 12 is a range symmetrical about the center in the lateral width direction of the tableware transport unit 10.

In the automatic tableware washing system of the fifth embodiment, the articulated robot 12 is fixed to the transport unit straddling pedestal 70 so that the movable range of the articulated robot 12 is a range in which the center in the width direction of the tableware transport unit 10 is symmetrical. Therefore, the articulated robot 12 can pick up the tableware 11 in a wide range in the width direction on the tableware transport unit 10, and can easily take various postures.

Further, in the automatic tableware washing system of the fifth embodiment, since the articulated robot 12 is arranged near the upper side of the exit of the tableware transport unit 10, which has not been effectively utilized so far, the entire system can be made compact and the whole system can be made compact. Space can be saved by not occupying the passage.

[Embodiment 6]
The automatic dishwashing system, the automatic dishwashing method, the automatic dishwashing program, and the storage medium of the sixth embodiment of the present invention will be described with reference to FIG. FIG. 15 is a diagram showing the vicinity of the delivery table 55 of the automatic dishwashing system. In FIG. 15, the same reference numerals are used for the same configurations as in FIGS. 1 to 14, and the description thereof will be omitted.

When the articulated robot 12 picks up the tableware 11 and puts it face down on the delivery table 55, if the articulated robot 12 recognizes the type of the tableware 11 and the height of the tableware 11, the tableware 11 is adjusted to the height of the delivery table 55. Can be placed. However, if the type of tableware 11 is erroneously recognized, the tableware 11 may be pressed against the delivery table 55 or the tableware 11 may be dropped from above the delivery table 55.
If the type of tableware 11 is erroneously recognized in this way, there is a problem that the container accommodating robot 56 cannot operate because the tableware 11 is damaged or the position of the tableware 11 on the delivery table 55 is greatly deviated.
Such a problem can be solved by reducing the speed of the articulated robot 12 and increasing the certainty of the recognition of the tableware 11, but the processing performance of the tableware 11 is not obtained.

The tableware automatic washing system of the present embodiment includes a detector 80 that detects that the tableware 11 has crossed a horizontal plane separated from the delivery table 55 by a predetermined interval.

The detector 80 is realized by, for example, a transmissive sensor. When a transmissive sensor is used as the detector 80, the detector 80 projects a linear laser onto, for example, a horizontal plane a few cm above the delivery table 55, indicating that the tableware 11 has crossed the laser. To detect.

In the tableware automatic washing system, when the articulated robot 12 picks up the tableware 11 and puts it face down on the delivery table 55, it does not recognize the type of the tableware 11 or the height of the tableware 11. Even so, the height of the tableware 11 is estimated at the same time as the articulated robot 12 is decelerated at the timing when the detector 80 detects that the tableware 11 has crossed the horizontal plane separated from the delivery table 55 by a predetermined interval. Place the tableware 11 according to the height of the delivery table 55.

In the automatic tableware washing system of the sixth embodiment, the detector 80 detects that the tableware 11 has crossed a horizontal plane separated from the delivery table 55 by a predetermined interval, so that the tableware 11 is delivered without relying on image recognition. Since the tableware 11 can be placed according to the height of the table 55, the tableware 11 can be processed at high speed and reliably.

Although some embodiments of the present invention have been described above, these embodiments exemplify an automatic dishwashing system for embodying the technical idea of the present invention, and specify the present invention to these. It is not a thing, and it can be applied equally to those of other embodiments, and it is possible to omit, add, or change a part of these embodiments, or to combine the embodiments of each embodiment. is there.

1 ... Tableware input unit 2 ... Tableware washing unit 3 ... Tableware extraction unit 10 ... Tableware transport unit 11 ... Tableware 12 ... Articulated robot 13 ... Tableware storage unit 14 ... Tableware information acquisition unit 30 ... Tableware extraction calculation unit 31 ... Tableware position・ Attitude detection unit 32… Tableware extraction order planning unit 33… Tableware extraction operation command unit 35… Image acquisition unit 36… Tableware loading unit 37… Tableware transport unit 38… Dishwashing unit 40, 40C… Hand 41… Suction unit 41A… No. 1 suction part 41B ... second suction part 42 ... AR marker 45 ... extension part 46 ... hand attachment part 47 ... vacuum hose 48 ... hand side vacuum hose attachment part 49 ... joint part 50 ... detection frame 51 ... type display 55 ... Delivery table 56 ... Container storage robot 56
57 ... Container 60 ... Spring mechanism 61 ... Ball joint 62 ... Force sensor 70 ... Transport section straddle mount 80 ... Detector

Claims (14)

1. A dishwashing section that cleans the dishes that are thrown in from the tableware throwing section
   A tableware transporting unit that transports dishes washed by the dishwashing unit,
   A tableware information acquisition unit that identifies the tableware to be transported by the tableware transport unit and acquires information on the tableware.
   A tableware storage unit on which the tableware transported by the tableware transport unit is placed,
   The tableware is picked up from the tableware transport section, and the tableware is placed on the tableware storage section.
   In an automatic dishwashing system equipped with
   The information acquired by the tableware information acquisition unit includes the type of tableware, the size of the tableware, the shape of the tableware, the color of the tableware, the position of the tableware, the posture of the tableware, the orientation of the tableware, the marker attached to the tableware, or Includes at least one of the codes attached to the tableware,
   When the tableware information acquisition unit identifies the tableware being transported by the tableware transport unit, the tableware collector picks up the tableware according to the state of the tableware based on the information acquired by the tableware information acquisition unit. An automatic dishwashing system featuring.
2. The automatic tableware cleaning system according to claim 1, wherein the tableware information acquisition unit picks up the tableware while the tableware is being transported by the tableware transport unit.
3. The tableware automatic washing system according to claim 1 or 2, wherein the tableware information acquisition unit includes an image acquisition unit.
4. When the information acquired by the tableware information acquisition unit is a marker,
   The marker is
   (1) A sticker attached,
   (2) Formed by laser processing,
   (3) What was given by printing,
   (4) Coated or coated
   (5) Another member to which the marker is printed is attached.
   The automatic dishwashing system according to any one of claims 1 to 3, wherein the dish is at least one of the above.
5. The automatic tableware washing system according to any one of claims 1 to 3, wherein the tableware is identified without a marker provided on the tableware.
6. The tableware automatic washing system according to any one of claims 1 to 5, wherein the tableware information acquisition unit or the tableware pick-up upper portion predicts the position of the tableware.
7. The automatic tableware washing system according to any one of claims 1 to 6, wherein the upper part of the tableware picker is provided with an elastic member.
8. The invention according to any one of claims 1 to 7, wherein at least one of the tableware input unit, the tableware information acquisition unit, and the tableware picking upper portion is controlled by using machine learning. Automatic dishwashing system.
9. The item according to any one of claims 1 to 8, wherein the tableware information acquisition unit includes one or more image pickup units, and can acquire tableware information from one or a plurality of images acquired by the image pickup unit. Automatic dishwashing system.
10. The upper part of the tableware collecting includes an articulated robot that picks up the tableware by the suction part.
    The automatic dishwashing system according to any one of claims 1 to 9, wherein the suction portion includes a plurality of suction portions having different suction directions.
11. Further, the automatic tableware washing system according to any one of claims 1 to 10, wherein the tableware is stored in a container using a container storage robot.
12. The dishwashing process to wash the thrown dishes and
    The tableware transporting process for transporting the dishes washed in the dishwashing step and
    A tableware information acquisition process for identifying the tableware to be transported in the tableware transportation process and acquiring information on the tableware,
    The tableware picking process of picking up the tableware transported in the tableware transporting process and placing the picked up tableware,
    In the automatic dishwashing method including
    The information acquired in the tableware information acquisition process includes the type of tableware, the size of the tableware, the shape of the tableware, the color of the tableware, the position of the tableware, the orientation of the tableware, the orientation of the tableware, the marker attached to the tableware, or Includes at least one of the codes attached to the tableware,
    When the tableware transported in the tableware transporting process is identified by the tableware information acquisition process, the tableware picking process picks up the tableware according to the state of the tableware based on the information acquired in the tableware information acquisition process. An automatic dishwashing method characterized by this.
13. An automatic dishwashing program characterized in that each step in the automatic dishwashing method according to claim 12 can be executed by a computer unit.
14. A storage medium that stores the automatic dishwashing program according to claim 13.

Images