JP6711555B1 - Transport system, area determination device, and area determination method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a transport system capable of realizing highly reliable automatic traveling of a transport vehicle. A transport vehicle 2 can automatically travel based on a current position specified by using a laser scanner and a reflector 10 (laser guidance). The carrier 2 can also automatically travel based on the current position specified by the SLAM using information from a measurement sensor that measures the surrounding environment (SLAM guidance). The control device of the carrier vehicle 2 controls the carrier vehicle 2 so that the carrier vehicle 2 switches between the laser guidance and the SLAM guidance under predetermined conditions and travels in the facility 1. Also provided is an apparatus and method for determining a region 40 in which the effectiveness of SLAM guidance within the facility 1 of the transport system is assured using a trained model. [Selection diagram] Figure 1

Description

The present invention relates to a transportation system that automatically drives a transportation vehicle in a facility. The present invention also relates to an apparatus and method for determining an area in the facility where SLAM guidance is effective in the transportation system.

As disclosed in Patent Document 1, laser guidance is known as a guidance system for automatically traveling a carrier vehicle. In the laser guidance, a laser scanner is installed in a transport vehicle, and a plurality of reflectors are installed on walls, pillars, etc. in the facility.

The laser scanner projects the laser light to the surroundings and detects the laser light reflected and returned from the plurality of reflectors. The current position of the carrier is specified based on the triangulation principle using information from the laser scanner.

As shown in Patent Documents 2 and 3, SLAM induction using SLAM (Simultaneous Localization And Mapping) is also known as an induction method. The current position of the transport vehicle is specified (estimated) by SLAM using information from a measurement sensor that measures the surrounding environment.

JP-A-2000-56828 JP, 2013-45298, A JP, 2014-219721, A

Laser guidance has higher positional accuracy than SLAM guidance. However, the current position cannot be specified unless the laser scanner can recognize at least three reflectors. If the laser light is blocked by luggage or an unexpected obstacle, the current position may not be specified. As a result, the transport vehicle cannot run automatically.

In contrast, SLAM guidance does not have the above-mentioned problems of laser guidance. However, SLAM guidance has lower positional accuracy than laser guidance. In many cases, SLAM guidance cannot be used for locations where high positional accuracy is required.

The laser guidance and the SLAM guidance each have advantages and disadvantages, and the certainty of automatic traveling is not necessarily high.

According to one aspect of the present invention, there is provided a transport system capable of realizing highly reliable automatic traveling. Further, according to another aspect of the present invention, there is provided an apparatus and a method for determining an area in the transportation system where the effectiveness of SLAM guidance in the facility is certain.

According to one aspect of the present invention, there is provided a transfer system, wherein the transfer system is provided in a facility, a plurality of reflectors installed in the facility, a vehicle traveling in the facility, and a vehicle. A laser scanner that projects laser light around the transport vehicle and detects the laser light reflected by a reflector, a measurement sensor that is provided on the transport vehicle, and measures the environment around the transport vehicle, and the transport vehicle. And a control device that controls the transport vehicle. The control device includes a reflector storage unit in which positional information of the reflectors in the facility is stored in advance, and a positional relationship between three reflectors among the plurality of reflectors recognized by using the information from the laser scanner and the reflectors. By collating the position information with the first position identifying unit that identifies the current position of the transport vehicle, and by collating the information from the measurement sensor with the environmental map of the facility, the current position of the transport vehicle is identified. A second position specifying unit. The transport vehicle has a first guidance that automatically travels based on the current position identified by the first position identifying unit, and a second guidance that automatically travels based on the current position identified by the second position identifying unit ; You can drive in.

Further, according to the present invention, the control device, at the time of the second guidance of the guided vehicle, specifies the current position of the guided vehicle by the second position specifying section and simultaneously creates an environmental map using the information from the measurement sensor. Further, the facility partially includes a certain area in which the effectiveness of the second guidance is certain in the facility, and when the transport vehicle cannot run on the first guidance in the certain area, When the vehicle is switched to the guidance and the vehicle travels within the certain area by the second guidance and cannot travel outside the certain area by the first guidance, the control device controls to stop traveling.

MAP creation unit may further create an environmental map even when the first induction transport vehicle.

Conveying system further includes an area determining device for determining the probability the real area, the control unit may control the transporting vehicle according sure real area determined by the area determination unit. Area determination apparatus, the shape of the facility, the breadth, and the luggage layout for the learned model obtained by the machine learning the correlation between the probability the real area of the facility, the shape features of the transport system, breadth, and inputs the information indicating the cargo layout may comprise a region determining unit that determines a probability real area features of the transport system.

According to another aspect of the present invention, the region determination apparatus for determining the probability real area in the facility in the transport system of the above aspects are provided. Realm determination device, the shape of the facility, the breadth, and the luggage layout for the learned model obtained by the machine learning the correlation between the probability the real area of the facility, the shape of the target facility, the breadth It is, and, by entering the information indicating the cargo layout comprises a region determination unit that determines a probability real area of the target site.

According to yet another aspect of the present invention, the area determination method for determining the probability real area in the facility in the transport system of the above aspects are provided. Realm determination method, the shape of the facility, the breadth, and the luggage layout for the learned model obtained by the machine learning the correlation between the probability the real area of the facility, the shape of the target facility, the breadth And the step of inputting information indicating the luggage layout,
And a step of determining a probability real area in the target facility using the information outputted from the learned model.

According to the transport system of the present invention, the transport vehicle automatically switches through the facility by switching between the first guidance and the second guidance as described above . Thus, by combining the two guidance systems, highly reliable automatic driving can be realized. According to the area determination device and the area determination method of the present invention, the area in which the effectiveness of the second guidance is certain is determined by using the learned model, which is useful for implementing the transport system of the present invention.

It is a schematic plan view which shows the conveyance system which concerns on one Embodiment. It is a side view which shows a conveyance vehicle. It is a block diagram which shows a conveyance vehicle. It is a figure for explaining laser guidance. It is a block diagram which shows the area|region determination apparatus which concerns on one Embodiment. It is a figure for explaining an example of teacher data. It is a figure for explaining an example of teacher data. 9 is a flowchart illustrating a region determination method according to an embodiment. It is a conceptual diagram which shows the area|region determination method which concerns on one Embodiment.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[Transport system]
FIG. 1 is a plan view schematically showing a transfer system according to an embodiment. The transport system includes a facility 1 such as a factory or a warehouse, and at least one transport vehicle 2 that automatically travels in the facility 1. The transport system includes a plurality of reflectors 10 installed in the facility 1. The reflector 10 is installed on a structure such as a wall or a pillar in the facility 1 as shown in FIG. 1, for example. Inside the facility 1, luggage 11 is placed on the floor while being stacked. It should be noted that the luggage 11 may be stored on a shelf arranged in the facility 1.

As shown in FIG. 2, in the embodiment, the transport vehicle 2 is an unmanned forklift that automatically travels and performs cargo handling work. The transport vehicle 2 includes a vehicle main body 20 having traveling wheels 21, a mast 22 provided on the vehicle main body 20, and a fork 23 that can be moved up and down with respect to the mast 22.

A laser scanner 24 is provided on the transport vehicle 2 and projects the laser light L (FIG. 4) around the transport vehicle 2 and detects the laser light reflected by the reflector 10.

The SLAM measurement sensor 25 is provided in the transport vehicle 2 and measures the environment around the transport vehicle 2. SLAM (Simultaneous Localization And Mapping) is a technique that simultaneously estimates the self-position and creates an environment map, and is a well-known technique, so its details are omitted.

The measurement sensor 25 may be, for example, a distance measurement sensor that measures the distance to surrounding objects. The distance measuring sensor is, for example, a laser range finder (LRF), a stereo camera, or a depth camera. In addition, when the Visual SLAM is used, the measurement sensor 25 may be a monocular camera that captures the surrounding environment. Not only one sensor 25 but two or more sensors may be provided in one transport vehicle 2. For SLAM, in addition to the external sensor as described above, an internal sensor may be provided in the transport vehicle 2 as needed.

As illustrated in FIG. 3, the transport vehicle 2 further includes a traveling device 26, a cargo handling device 27, and a communication device 28. A control device 3 is provided on the transport vehicle 2 to control the transport vehicle 2. The traveling device 26 includes traveling wheels 21 and a drive system for driving the traveling wheels 21. The cargo handling device 27 includes a mast 22, a fork 23, and a drive system for driving these. The control device 3 and the communication device 28 are provided inside the vehicle body 20. A part of the traveling device 26 and the cargo handling device 27 is provided inside the vehicle body 20.

The communication device 28 wirelessly communicates with a management device (not shown) installed in the facility 1, receives information (work instruction) regarding work to be performed, and sends the information to the control device 3. Further, the communication device 28 sequentially transmits the work progress and the like to the management measures.

The control device 3 includes a first position specifying unit 31, a second position specifying unit 32, a travel control unit 33, and a cargo handling control unit 34. The first and second position specifying units 31 and 32 specify the current position of the transport vehicle 2 by the method described later.

The traveling control unit 33 controls the traveling device 26 in accordance with the work command while referring to the current position specified by the first or second position specifying unit 31, 32, and thereby controls the traveling of the carrier vehicle 2. Specifically, the traveling control unit 33 rolls the traveling wheels 21 while controlling the steering angle of the traveling wheels 21 so that the transport vehicle 2 moves toward a predetermined cargo handling work position. That is, the transport vehicle 2 automatically travels along the travel route.

The cargo handling control unit 34 controls the cargo handling device 27 according to a work command while referring to the current position identified by the first or second position identifying unit 31, 32, and thereby controls the cargo handling operation of the carrier vehicle 2. Specifically, when the transport vehicle 2 arrives at a predetermined cargo handling work position, the cargo handling control unit 34 moves the mast 22 and the fork 23 to perform loading or unloading. That is, the carrier 2 automatically performs the cargo handling work.

The control device 3 further includes a reflector storage unit 35 in which positional information of the reflector 10 in the facility 1 is stored in advance. The first position specifying unit 31 and the reflector storage unit 35 are laser guidance (first guidance) in which the transport vehicle 2 automatically travels based on the current position specified by the principle of triangulation using information from the laser scanner 24. ) Is realized.

As shown in FIG. 4, the laser scanner 24 horizontally projects the laser light L by 360 degrees, projects the laser light L around the laser light L, and detects the reflected laser light L. The first position specifying unit 31 uses the information from the laser scanner 24 to calculate the distance to each reflector 10 and the azimuth of the reflector 10. The first position specifying unit 31 collates the positional relationship of the three reflectors 10 among the plurality of reflectors 10 thus recognized with the positional information of the reflectors 10 stored in advance, and thus Identify your current location.

The traveling control unit 33 controls the traveling device 26 while referring to the current position identified by the first position identifying unit 31, whereby the laser guide of the transport vehicle 2 is automatically performed.

The control device 3 further includes a SLAM unit 30 that simultaneously identifies (estimates) the current position of the transport vehicle 2 and creates an environment map (global map) inside the facility 1. The SLAM unit 30 is used to realize SLAM guidance (second guidance) in which the vehicle 2 automatically travels based on the current position specified (estimated) by SALM using the information from the measurement sensor 25.

The SLAM unit 30 includes the second position specifying unit 32, the map creating unit 36, and the environment map storage unit 37 described above. The map creation unit 36 creates an environment map (global map) of the facility 1 using the information from the measurement sensor 25. The environment map is stored in the environment map storage unit 37 and updated. The second position specifying unit 32 specifies the current position of the transport vehicle 2 by matching the information from the measurement sensor 25 with the environment map. The environmental map is created and the current position is specified at the same time.

In the present embodiment, a basic environment map including information indicating fixedly installed structures inside the facility 1 such as walls and shelves inside the facility 1 is created and prepared in advance, and the environment map storage unit 37 is provided. It is stored in. Based on the basic environment map, the map creation unit 36 creates an environment map that includes information that indicates movable structures such as luggage and obstacles in the facility 1, in addition to information that indicates structures.

The traveling control unit 33 controls the traveling device 26 while referring to the current position specified by the SLAM by the second position specifying unit 32, so that the automatic traveling by the SLAM guidance of the transport vehicle 2 is realized.

The traveling control unit 33 refers to the current position identified by the first position identifying unit 31 or the current position identified by the second position identifying unit 32 to control the traveling device 26 according to a predetermined condition. Switch. That is, the traveling control unit 33 controls the traveling of the transport vehicle 2 so that the transport vehicle 2 switches between the laser guidance and the SLAM guidance according to a predetermined condition and travels.

As shown in FIG. 1, the facility 1 partially includes a SLAM certain area 40 (referred to by a chain line) (hereinafter simply referred to as a certain area), which is an area where the effectiveness of SLAM guidance in the facility 1 is certain. The certainty region 40 is a region in which position accuracy is not required, such as traveling in a wide region, and there is a low possibility that an obstacle will occur even in SLAM guidance with low position accuracy. The certainty region 40 may be preliminarily defined by actual experiments, or may be preliminarily defined by experiments by simulation or the like. Further, the certainty region 40 may be determined by the region determining device 5 (FIG. 5) as described later. The information indicating the certainty area 40 is stored in the area storage unit 38 (FIG. 3) of the control device 3.

The remaining area in the facility 1 is a SLAM uncertain area 41 (see a chain double-dashed line) (hereinafter simply referred to as an uncertain area) in which the effectiveness of SLAM guidance in the facility 1 is uncertain. The uncertainty region 41 is a region where positional accuracy is required, such as traveling in a narrow passage and cargo handling work that requires highly accurate positioning, and there is a high possibility that obstacles will occur in SLAM guidance. Therefore, the uncertain region 41 is a region where laser guidance having higher positional accuracy than SLAM guidance should be used.

Each storage unit of the control device 3 is realized by a semiconductor memory device such as a RAM or a flash memory, or a storage device such as a hard disk or an optical disk. The control device 3 includes, for example, a CPU, and each functional unit of the control device 3 is realized by the CPU executing a program stored in the storage device.

As an example, the carrier 2 is controlled by the controller 3 as follows.

When the control device 3 receives the work command, the travel control unit 33 controls the travel device 26 according to the work command while referring to the current position specified by the first position specifying unit 31. Therefore, the carrier 2 travels by laser guidance. In addition, the cargo handling control unit 34 controls the cargo handling device 27 according to the work instruction with reference to the current position identified by the first position identifying unit 31. Therefore, the carrier 2 performs the cargo handling work by laser guidance. In this way, the carrier 2 normally travels inside the facility 1 and performs cargo handling work by laser guidance having higher positional accuracy than SLAM guidance.

As mentioned above, in laser guidance three reflectors 10 must be recognized. Therefore, when the laser light L is blocked for some reason such as luggage or an obstacle and the three reflectors 10 cannot be recognized, the first position specifying unit 31 cannot specify the current position, and as a result, the transport vehicle 2 It becomes impossible to run automatically by laser guidance.

When the first position specifying unit 31 cannot specify the current position, the traveling control unit 33 determines that the last position of the carrier vehicle 2 specified by the first position specifying unit 31 (the current position specified immediately before being unable to specify) is It is determined whether or not it is within the certainty region 40.

When the last position is within the certainty region 40, the traveling control unit 33 refers to the position specified by the second position specifying unit 32 instead of the first position specifying unit 31 and follows the work command according to the work instruction 26. To control. Therefore, when the guided vehicle 2 cannot travel in the certain area 40 by the laser guidance, it switches from the laser guidance to the SLAM guidance and travels in the certain area 40 by the SLAM guidance.

Here, the traveling control unit 33 gives the last position specified by the first position specifying unit 31 to the SLAM unit 30 as an initial position for SLAM processing, and the second position specifying unit 32 uses SLAM using the initial position. Start specifying the current position of the carrier 2. The carrier 2 may be traveling before switching from laser guidance to SLAM guidance. Therefore, the traveling control unit 33 calculates the current position of the transport vehicle 2 using information from a sensor such as an encoder that detects the rotation of the traveling wheels 21 and the final position, and the second position specifying unit 32 This may be used as an initial position to start identifying the current position of the carrier 2.

On the other hand, the travel control unit 33 controls the travel device 26 to stop when the last position is outside the certain area 40 (inside the uncertain area 41). That is, when the guided vehicle 2 cannot travel by laser guidance outside the certain area 40, it stops traveling without switching to SLAM guidance.

After switching to SLAM guidance, the traveling control unit 33 uses the current position identified by the first position identifying unit 31 if the first position identifying unit 31 can identify the current position. To control. Therefore, when the guided vehicle 2 is ready to travel with laser guidance, it switches from SLAM guidance to laser guidance and travels.

However, after switching to SLAM guidance, the traveling control unit 33, when the current position specified by the second position specifying unit 32 is outside the certainty region 40 while the first position specifying unit 31 cannot specify the current position, The traveling device 26 is stopped and controlled. Therefore, the transport vehicle 2 stops traveling as soon as it comes out of the certain area 40 by SLAM guidance.

As described above, in the transport system of the embodiment, the transport vehicle 2 is controlled by the control device 3 so that the vehicle 2 can travel only in the certain area 40 by the SLAM guidance and the outside of the certain area 40 can only travel by the laser guidance. ing.

In the embodiment, the transport vehicle 2 is basically driven by the laser guidance with high positional accuracy, and is controlled by the control device 3 so as to switch to the SLAM guidance and travel when the laser guidance becomes impossible. As a result, the transport vehicle 2 can continue traveling even if the laser guidance is disabled. The certainty of automatic driving is improved.

The switching from laser guidance to SLAM guidance takes place only within the certainty region 40. In the region 41 where the effectiveness of the SLAM guidance is uncertain, the transport vehicle 2 is stopped without forcibly switching to the SLAM guidance, so that an accident caused by the SLAM guidance with relatively low position accuracy is suppressed. The safety of the transportation system is improved.

The predetermined condition for switching between the laser guidance and the SLAM guidance is not limited to the above embodiment.

It is preferable that the map creation unit 36 for SLAM guidance creates an environmental map of the facility 1 by SLAM using the information from the measurement sensor 25 not only at the time of SLAM guidance of the transport vehicle 2 but also at the time of laser guidance. As a result, in the transport system in which the laser guidance is the main guidance and the SLAM guidance is the preliminary guidance in an emergency, even if there is no basic environment map like the embodiment, a work command is issued when switching to the SLAM guidance. Enough environmental maps are created during laser guidance to ensure compliance with the following.

The transport system includes an area determining device 5 (FIG. 5) for determining the certainty area 40, and the control device 3 follows the certainty area 40 determined by the determining device 5 as described below. The transport vehicle 2 may be controlled as described above. Hereinafter, an example of the area determination device 5 will be described.

[Area determination device]
FIG. 5 is a block diagram showing an example of a region determination device (hereinafter, simply determination device). The determination device 5 includes a control unit 50, a storage unit 51, an acquisition unit 52, and a communication unit 53.

The control unit 50 includes a learning model generation unit 500, a region determination unit 501, and an output control unit 502. The control unit 50 includes, for example, a CPU, and these functional units of the control unit 50 are realized by the CPU executing the programs stored in the storage device.

The learning model generation unit 500 generates a learned model which will be described later and is used to determine the certainty region 40. The area determination unit 501 determines the certain area 40 using the learned model. The output control unit 502 wirelessly transmits information indicating the determined certainty region 40 to the control device 3 of the guided vehicle 2 using the communication unit 53.

The storage unit 51 stores various data. The storage unit 51 includes a teacher data storage unit 510 and a learning model storage unit 511. The storage unit 51 is realized by, for example, a semiconductor memory device such as a RAM or a flash memory, or a storage device such as a hard disk or an optical disk.

The teacher data storage unit 510 stores teacher data used for generating a learned model. The learned model storage unit 511 stores the learned model generated by the learning model generation unit 500.

The acquisition unit 52 acquires information necessary for determining the certainty area 40. In the embodiment, the acquisition unit 52 is a camera that is provided in the facility 1 in which the certainty region 40 is to be determined, images the inside of the facility 1, and acquires image data of images in the facility 1. The camera is arranged, for example, on the ceiling in the facility 1, captures an image of the facility 1, and acquires image data of a planar image of the facility 1.

The communication unit 53 is used for the determination device 5 and the control device 3 of the transport vehicle 2 to perform wireless communication via a communication network. The communication unit 53 includes a communication device and the like. The control device 3 of the transport vehicle 2 can communicate with the determination device 5 via the communication device 28.

Next, the trained model used to determine the certainty region 40 is described. In the embodiment, a trained model of a neural network (hereinafter referred to as NN), more specifically, a trained model of deep learning (multilayer NN) is used. However, other algorithms can be applied as long as a trained model capable of determining the certainty region 40 can be constructed.

For the generation of the learned model, the data related to the shape, the size, the luggage layout, and the certainty region 40 in the facility 1 shown in FIGS. 6 and 7 are used as the teacher data. In this teacher data, a combination of the shape, size and baggage layout in the facility 1 is input data, and the certainty region 40 in the facility 1 corresponding to this combination is correct answer data.

The shape of the facility 1 may include at least a planar shape as shown in FIGS. The planar shape is a planar shape or a flat cross sectional shape. The shape inside the facility 1 is the shape of a structure that defines the work place where the transport vehicle 2 travels and works inside the facility 1, and not only the walls inside the facility 1 but also the pillars and partitions (not shown). Including. The size of the facility 1 is the planar size of the work place in the facility 1.

The luggage layout in the facility 1 may include at least a planar layout. The luggage layout is the layout of the luggage 11 when the luggage 11 is placed on the floor as shown in FIG. 6, and the layout of the shelf 12 when the luggage 11 is stored on the shelf 12 as shown in FIG. 7. Is.

The traveling space in which the carrier 2 can travel in the facility 1 is formed by a structure in the facility 1 such as a wall, or the structure and the luggage 11/shelf 12 placed in the facility 1. Between them, or between them by adjacent luggage 11/shelf 12 placed in the facility 1. SLAM guidance is a guidance method with low position accuracy. That is, there is a correlation between the shape, size, and baggage layout within the facility 1 and the certain area 40 (area where the effectiveness of SLAM guidance is certain) inside the facility 1.

The certainty region 40 in the facility 1 as the correct answer data is specified by actual experiment or experiment by simulation or the like.

The teacher data may be in the form of image data of an image in which the shape, size, baggage layout, and certainty region 40 in the facility 1 are shown in association with each other. For example, the certainty region 40 is specified on the image by hatching. However, the format of the teacher data does not matter.

In addition to the correlation shown in FIG. 6 and FIG. 7, a plurality of teacher data showing the above correlation in other facilities is stored in the teacher data storage unit 510 in advance.

The learning model generation unit 500 performs machine learning using a plurality of teacher data stored in the teacher data storage unit 510 to generate a learned model. This learned model uses the information indicating the shape, size, and baggage layout in the target facility Z (see FIG. 9), which is the target facility for determining the certainty region 40, as input data, and the certainty region in the target facility Z is obtained. 40 is a model that predicts 40 and outputs information indicating this as output data.

As described above, the learned model in which the shape, the size, and the luggage layout in the facility 1 and the correlation between the certainty region 40 in the facility 1 are machine-learned is generated. The learned model is stored in the learning model storage unit 511.

[Area determination method]
Hereinafter, an example of a region determination method (hereinafter, simply referred to as a determination method) for determining the certainty region 40 by the region determination device 5 will be described. FIG. 8 is a flowchart showing the flow of the determining method, and FIG. 9 conceptually shows the flow of the determining method. The learned model is generated in advance by the learning model generation unit 500 and stored in the learning model storage unit 511.

The determination method includes steps S1 to S4 as shown in FIG.

S1 is a step in which the determination device 5 acquires information necessary for determining the certainty region 40 in the target facility Z. That is, the acquisition unit 52 acquires information indicating the shape, the size, and the luggage layout in the target facility Z. In S1 of the embodiment, the camera as the acquisition unit 52 captures an image of the inside of the target facility Z (here, the facility 1 of the transport system in FIG. 1) to determine the shape, size, and luggage layout of the target facility Z. The image data of the image shown is acquired.

S2 is a step in which the determining device 5 inputs the information acquired in S1 to the learned model. In S2 of the embodiment, the area determination unit 501 inputs information indicating the shape, the size, and the baggage layout in the target facility Z into the learned model. Here, before the input to the learned model, the area determination unit 501 performs a pre-processing to convert the information (data) obtained in S1 into the same format as the input data of the teacher data described above. ..

S3 is a step in which the determination device 5 determines the certainty region 40 in the target facility Z using the information output from the learned model. In S3 of the embodiment, the area determination unit 501 causes the learned model to predict the certain area 40 in the target facility Z, and outputs information indicating this as output data. The area determination unit 501 determines the certainty area 40 predicted by the learning model as the certainty area 40 in the target facility Z.

S4 is a step in which the determination device 5 transmits information indicating the certain area 40 of the target facility Z determined in S3 to the transport vehicle 2 (the control device 3 thereof). In S5 of the embodiment, the output control unit 502 transmits information indicating the certain area 40 of the target facility Z to the transport vehicle 2 using the communication unit 53. The control device 3 of the transport vehicle 2 receives the information via the communication device 28 and stores it in the area storage unit 38.

After that, in the transport system, a step is performed in which the control device 3 controls the transport vehicle 2 as described above according to the certain area 40 determined by the area determination device 5.

According to the determination device 5 and the determination method according to the above-described embodiment, the certainty region 40 in the target facility Z is determined by the region determination unit 501 using the learned model. This eliminates the need for the operator to actually or experimentally determine the certainty region 40 when carrying out the transport system.

While the transport vehicle 2 is working in the facility 1, the acquisition unit 52 may acquire information indicating the shape, the size, and the luggage layout in the facility 1 at regular intervals. Then, each time the acquisition unit 52 acquires the information, the area determination unit 501 uses the information to determine the certain area 40, and the output control unit 502 provides information indicating that to the guided vehicle 2 (the control device 3 thereof). May be sent to.

For example, when the luggage 11 is placed on the floor as in the transport system shown in FIG. 1, the luggage layout changes with time as the work progresses. According to the above configuration, the certain area 40 suitable for the luggage layout at that time is determined, and the transport vehicle 2 is controlled by the control device 3 in accordance with it, which is advantageous.

Although the preferred embodiment of the present invention has been described above, the present invention is not limited to the above embodiment.

1 Facility 10 Reflector 11 Baggage 12 Shelf 2 Transport Vehicle 24 Laser Scanner 25 Measurement Sensor 26 Traveling Device 3 Control Device 30 SLAM Unit 31 First Position Identification Unit 32 Second Position Identification Unit 33 Travel Control Unit 36 Map Creation Unit 40 SLAM Certain Area 41 SLAM uncertainty area 5 Area determination device 50 Control unit 500 Learning model generation unit 501 Area determination unit 502 Output control unit 51 Storage unit 52 Acquisition unit 53 Communication unit L Laser light Z Target facility

Claims (5)

Facilities and
A plurality of reflectors installed in the facility,
A carrier that travels in the facility,
A laser scanner that is provided in the transport vehicle, detects the laser light reflected by the reflector while projecting the laser light around the transport vehicle,
A measurement sensor that is provided in the transport vehicle and measures the environment around the transport vehicle,
A control device that is provided on the transport vehicle and controls the transport vehicle;
The control device is
A reflector storage unit in which positional information of the reflector in the facility is stored in advance,
The current position of the transport vehicle is specified by collating the positional information of the three reflectors among the plurality of reflectors recognized using the information from the laser scanner with the positional information of the reflectors. A first position specifying unit,
A second position specifying unit that specifies the current position of the transport vehicle by collating the information from the measurement sensor with the environmental map of the facility,
The transport vehicle, a first inductive running automatically based on the current position specified by the first position specifying section, the second running automatically based on the current position identified by the second position specifying section With guidance , you can drive with
The control device creates the environment map by using the information from the measurement sensor at the same time when the second position of the carrier is specified by the second position specifying unit at the time of the second guidance of the carrier. More parts,
The facility partly includes a certain area, which is an area where the effectiveness of the second guidance is certain in the facility,
When the transport vehicle cannot travel in the certain area by the first guidance, it switches to the second guidance, travels in the certain area by the second guidance, and moves outside the certain area by the first guidance. When it becomes impossible to travel by induction, it is controlled by the control device to stop traveling,
A transport system characterized in that The mapping unit further also creates the environment map when the first induction of the transport vehicle,
The transport system according to claim 1 , wherein: Further comprising a region determination unit for determining a pre Ki確 real area,
Wherein the control device, the transport vehicle is controlled in accordance Ki確 real area before determined by the area determination unit,
The area determination device,
The shape of the facility, the breadth, and the luggage layout for the learned model the correlation between the pre Ki確 real area was machine learning of the facility, the shape of the facility of the transfer system, the breadth is, and, by entering the information indicating the cargo layout comprises a region determination unit that determines a pre Ki確 real area of the facility of the transport system,
The transport system according to claim 1 , wherein: A region determination apparatus for determining the reliable region in the facility in a transfer system according to claim 1,
The area determination device,
The shape of the facility, the breadth, and the luggage layout for the learned model the correlation between the pre Ki確 real area was machine learning of the facility, the target shape of the facility, the breadth, and, type information indicating the cargo layout comprises a region determination unit that determines a pre Ki確 real area of the target site,
An area determination device characterized by the above. A region determination method for determining the reliable region in the facility in a transfer system according to claim 1,
The area determination method is
The shape of the facility, the breadth, and the luggage layout for the learned model the correlation between the pre Ki確 real area was machine learning of the facility, the target shape of the facility, the breadth, and, A step of inputting information indicating a luggage layout,
And a step of determining a pre Ki確 real area of the target site using the information outputted from the learned model,
A region determination method characterized by the above.

Images