Automated guided vehicle and automated guided vehicle system

Abstract

To provide an unmanned carrier that can autonomously travel using an inexpensive image sensor even in an environment where disturbance light is incident in an indoor facility such as a storehouse, and an unmanned carrier system.SOLUTION: An unmanned carrier includes: an image sensor that captures an image of an object in the direction of travel; a drive unit that independently controls the drive of multiple wheels; a robot control unit that transmits a control command to the drive unit for autonomous driving along a travel path based on an environmental map; and a wireless communication unit. The robot control unit has: a function for constantly updating the environmental map and a route based on image information from the image sensor; a function for registering an image of a specific shape or pattern arranged along the travel path as a landmark image in association with coordinate information on the environmental map; a first current position estimation function for estimating a current position based on the image information from the image sensor and the environmental map; and a second current position estimation function for estimating a current position based on the coordinate information on the landmark image and the environmental map, when the image information from the image sensor includes the landmark image. A projection image is included as the landmark image.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an autonomous mobile vehicle having a map update function by, for example, 3D-SLAM (3 dimension Simultaneous Localization AND Mapping) and an autonomous mobile vehicle system including the autonomous mobile vehicle.

Background Art

[0002] Conventionally, as a guidance method for a rail-guided automatic guided vehicle (AGV), for example, the following methods are known. (1) A method in which an unmanned transport vehicle detects radio waves from a guide cable buried in a travel path (2) A method in which a guide tape is attached to a travel path and an unmanned transport vehicle detects magnetism or reflected light (3) A method in which a guide sign is installed on a travel path and an unmanned transport vehicle performs image recognition on the guide sign

[0003] However, in these methods, it is necessary to install guide facilities on the travel path in advance. Therefore, in recent years, a trackless unmanned transport method has been studied in which an unmanned transport vehicle recognizes obstacles with a sensor or the like and autonomously travels. As obstacle avoidance path generation methods, a rule-based method, a (virtual) potential method, and a search method are mainly known. Patent Document 1 discloses an autonomous mobile device that employs a virtual potential method and can create a plurality of partial maps that constitute an environmental map of a moving area based on distance information and angle information from surrounding objects read from a laser range finder. In this autonomous mobile device, SLAM (Simultaneous Localization and Mapping) is used to estimate the self-position and create an environmental map in real time.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the guidance method of (1) above, the cost of laying the guidance cable is high, and the travel route cannot be easily changed. In the guidance method of (2) above, if the guidance tape installed on the travel route is damaged, guidance becomes impossible, so it is necessary to periodically replace the guidance tape. In an environment where the travel route of an automated guided vehicle and the traffic line of a forklift overlap, such as in a warehouse, there is a problem that the replacement span of the guidance tape becomes shorter than usual.

[0006] In the guidance method of (3) above, it has been pointed out that inconsistencies occur in the environmental map created due to the accumulation of measurement errors. In particular, when creating an annular environmental map, the start part and the end part do not match (the annular path problem). Also, when the layout around the travel route is changed, it has been pointed out that it is necessary to remake the entire environmental map. Therefore, in Patent Document 1, it is proposed to manage the environmental map by dividing it into a plurality of partial maps, so as to limit the range of measurement errors to the partial maps and limit the range that needs to be remade when the layout is changed to the partial map unit. However, in an environment where a manned transport vehicle or a worker performs work in the same space, since the layout within the work space constantly changes, there is a problem that the accuracy of self-position estimation decreases due to unexpected environmental changes. Also, there is a need to minimize the system setting changes required when the layout in the warehouse is changed.

[0007] By the way, there is a need to use an inexpensive sensor as a means for grasping surrounding objects instead of an expensive laser rangefinder (for example, LiDAR (Light Detection and Ranging)). Although it is conceivable to use an inexpensive image sensor such as a stereo camera as an alternative means, when disturbing light enters the field of view of the automated guided vehicle, grasping of surrounding objects becomes temporarily impossible, and there arises a problem that the automated guided vehicle stops. In an environment where disturbing light constantly enters from the loading / unloading locations communicating with the outdoors, such as a warehouse, there exists a problem that it is difficult to configure an automated guided vehicle using a stereo camera.

[0008] Therefore, an object of the present invention is to provide an automated guided vehicle and an automated guided vehicle system capable of solving the above problems.

Means for Solving the Problems

[0009] The automated guided vehicle of the present invention System is An unmanned transport vehicle that autonomously travels along a travel route based on an environmental map, a landmark image projection device that projects a landmark image of a specific shape or pattern along the travel route, and a server that transmits a projection command to cause the landmark image projection device to project the landmark image. The unmanned transport vehicle system includes: an image sensor that captures an image of an object on the traveling direction side, a drive unit that independently controls the driving of a plurality of wheels, Color a robot control unit that transmits a control command for autonomous traveling to the drive unit, and a wireless communication unit, and the robot control unit the has a function of constantly updating an environmental map and a route based on image information from the image sensor, to a function of registering in association with coordinate information on the environmental map a landmark image, a first current position estimation function that performs feature point recognition on the image information from the image sensor and estimates the current position by comparing with the environmental map, Color and a second current position estimation function that estimates the current position based on the coordinate information of the landmark image when the landmark image is included in the image information from the image sensor, and is characterized in that it is attached to a wall surface or a ceiling. the projected landmark image to and a second current position estimation function that estimates the current position based on the coordinate information of the landmark image when the landmark image is included in the image information from the image sensor, and is characterized in that it is attached to a wall surface or a ceiling. Color When the landmark image is included in the image information from the image sensor, it has a second current position estimation function that estimates the current position based on the coordinate information of the landmark image, and is characterized in that it is attached to a wall surface or a ceiling. The landmark image projection device projects a floor line at the boundary between the travel route and the work area, projects a boundary line image of a color different from that of the floor line at a portion that serves as a passage at the boundary between the travel route and the work area, projects a floor guidance line having specific color information on the travel route, on a wall surface or a ceiling has a function of projecting a wall or ceiling guidance line having specific color information. The unmanned transport vehicle includes a first unmanned transport vehicle and a second unmanned transport vehicle. The robot control unit of the first unmanned transport vehicle performs first travel control based on the floor guidance line, and the robot control unit of the second unmanned transport vehicle performs second travel control based on the wall or ceiling guidance line is characterized by that. The above automated guided vehicle SystemIn this case, it is also preferable that the drive unit has a function of transmitting the rotation information of the plurality of wheels to the robot control unit as odometry information, and the robot control unit has a third current position estimation function of estimating the current position based on the odometry information and the current position information stored immediately before. The above unmanned carrier vehicle System In this case, the robot control unit In an environment where external light is incident, Color A function of generating a composite image with a virtual landmark image added to the image information from the image sensor, dynamically and a function of performing travel control based on the virtual landmark image included in the composite image may be provided. The above unmanned carrier vehicle System In this case, it is also preferable that the landmark image includes an image of an artificial code including a two-dimensional code.

[0010] the above In an unmanned carrier vehicle system, the landmark image is is a character string, a logo, or an image of a character, including which may be a feature. the above In an unmanned carrier vehicle system, the server for causing the first travel control to be performed sends a projection command to project the for causing the second travel control to be performed image in a first time zone and sends a projection command to project the image in a second time zone, which may be a feature.

[0011] According to the present invention, it is possible to provide an unmanned carrier vehicle and an unmanned carrier vehicle system that can autonomously travel using an inexpensive image sensor even in an environment where external light enters in an indoor facility such as a warehouse. In addition, it is possible to easily cope with layout changes in the warehouse.

Brief Description of the Drawings

[0012]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Mode for Carrying Out the Invention

[0013] Hereinafter, the system configuration of an example embodiment of the present invention will be described, and then the procedure of the inbound / outbound operation will be described as an example of the conveyance operation method of the present invention. <System Configuration> The automated guided vehicle system of the example embodiment includes an RM server 1, a WMS server 2, a management terminal 3, a handy terminal 4, a wireless LAN access point 5, a cargo vehicle 6, a landmark image projection device 7, and a plurality of automated guided vehicles 10. The handy terminal 4, the wireless LAN access point 5, the cargo vehicle 6, the landmark image projection device 7, and the plurality of automated guided vehicles 10 are installed at a work location in an indoor facility such as a warehouse.

[0014] The RM server 1 is a server that enables the management of the automated guided vehicle 10, and is constructed by installing RM software and database software on a server device equipped with a CPU and a storage device. This RM software includes, as software modules that realize the main functions, a carrier vehicle management unit 1a, an operation adjustment unit 1b, a work location management unit 1c, and a billing information management unit 1d. The carrier vehicle management unit 1a manages the work location information where the automated guided vehicle 10 is located and the operation information of the automated guided vehicle 10. In the carrier vehicle management unit 1a, the vehicle ID of the automated guided vehicle 10 and the work location ID of the work target are associated and managed. The operation adjustment unit 1b manages the operation time of a plurality of automated guided vehicles 10. The operation adjustment unit 1b manages the vehicle ID of the automated guided vehicle 10 and the work location ID of the operation target in an associated manner, which serves as input data for calculating billing information. Note that there may be cases where multiple automated guided vehicles 10 operate in the same space within the same building, or there may be cases where only one operates. The work location management unit 1c manages the basic map for each work location and transmits the basic map to the automated guided vehicle 10. In the basic map, a passable course and a prohibited area are set. Note that the management of the basic map of the work location may not be performed by the RM server 1, and it may be configured to be managed by the automated guided vehicle 10 and the management terminal installed at the work location. The billing information management unit 1d acquires the operation information of the automated guided vehicle 10 from the carrier vehicle management unit 1a and calculates the billing information in consideration of the rate table for each client. The projection device management unit 1e transmits a projection command to project a specific image onto a specific landmark image projection device 7. The projection device management unit 1e also has a function of transmitting a projection command to project different images in the first time zone and the second time zone.

[0015] The WMS server 2 is a server that enables the realization of a Warehouse Management System. It is constructed by installing WMS software and database software on a server device equipped with a CPU and a storage device. This WMS software includes an inventory management unit 2a and an order processing unit 2b as software modules that realize the main functions.

[0016] The inventory management unit 2a updates an inventory database that manages product IDs, storage addresses, statuses, etc. It also has functions such as allocating a storage address according to the received stock-in information from the management terminal 3, sending a stock-in instruction to the handy terminal 4 and the automated guided vehicle 10, and managing the quantity of products to be stocked and the status of the delivery situation. The order processing unit 2b updates an order database that manages the order information received from the management terminal 3. The order processing unit 2b sends a shipping instruction to the handy terminal 4 and the automated guided vehicle 10 based on the order information, and updates the status of the order to completed when receiving the shipping information from the handy terminal 4. It also has a reservation function for reserving inventory according to the order. When reserving with the order quantity of the inventory corresponding to a certain order, the quantity cannot be reserved from other orders, thus ensuring the inventory.

[0017] The landmark image projection device 7 is, for example, a laser beam irradiation device or a projector, and a plurality of them are installed on the ceiling, wall surface, or floor surface along the traveling path in the warehouse. The landmark image projection device 7 is configured to be communicable with the RM server 1, and projects an image of a specific shape or pattern that can be used as a landmark image onto the floor surface, wall surface, or ceiling based on a projection command from the RM server 1. Examples of the image of a specific shape or pattern projected by the landmark image projection device 7 include a line including a dotted line, a circle or an ellipse, a polygon including a triangle, a geometric pattern, an artificial code, a character string, a company logo, and a character. Different images may be projected onto the landmark image projection device 7 depending on the time zone. The landmark image projection device 7 in the embodiment example includes a turning device and can change the projection position based on a command from the RM server 1. As shown in FIG. 1, the automated guided vehicle 10 includes a robot control unit 11, a display unit 12, a wireless communication unit 13, a speaker 14, a power supply unit 15, a drive system (21 to 23), and a sensor group (31 to 33).

[0018] As shown in FIGS. 2 and 3, the drive system (21 to 23) of the automated guided vehicle 10 is arranged on the lower surface side of the base 16, and the other robot control unit 11 etc. are arranged on the upper surface side of the base 16. The wheel 19 is a single wheel without a driving device and is arranged on the lower surface of the base 16 via a turntable 18. On the advancing direction side of the base 16, a portal-shaped frame 17 is provided, and on the upper column of the frame 17, an arm for attaching the display unit 12, an emergency stop button 24, and a fixture for suspending and fixing the image sensor 33 are provided. On the side opposite to the advancing direction of the base 16, a connecting portion 41 provided with a locking column 42 is arranged. In the illustrated example, a locking plate 53 of a trailer 50 capable of towing a container is attached to the locking column 42. The trailer 50 is a non-self-propelled cart including a block-shaped body portion 51, a long plate-shaped support portion 52, and a pair of wheels 54a, 54b. The trailer 50 can be further connected to other trailers.

[0019] The robot control unit 11 is a computer on which the transport software configured using ROS (Robot Operating System), a standard platform for robot development, operates, and includes a map management unit 111, an operation management unit 112, a route generation unit 113, a sensor management unit 114, and a virtual image generation unit 115. The map management unit 111 has a map update function by 3D-SLAM (3-dimension Simultaneous Localization AND Mapping). The environmental map updated by the map update function is composed of a basic map for parts that do not change such as a floor plan and basic passages, and a plurality of partial maps. The map management unit 111 grasps the position and shape information of objects that are constantly changing, such as a cage car and a cargo box, based on signals from the sensor group (31 to 33), and real-time updates the partial map (peripheral map) around itself while traveling.

[0020] The operation management unit 112 controls the drive system (21 to 23) of the automated guided vehicle 10 based on the inbound and outbound instructions, enabling autonomous driving. The information on the rotation amounts of the wheels 23a and 23b obtained by the encoders provided on the axles is stored as odometry information. The operation management unit 112 has a function of estimating its own position without communicating with the RM server 1 using the image information, odometry information, environmental map, and observation data of the sensor group (31 to 33) (first current position estimation function). Here, a separate 6-axis sensor may be provided and the information from the 6-axis sensor may also be used in combination. A known estimation method (for example, the Monte Carlo method) can be used for the estimation of the own position. Also, it has a line following function described later. When a plurality of automated guided vehicles 10 are traveling on the same travel route, a function of exchanging their own positions with each other and ensuring a certain inter-vehicle distance may be provided. When the image information from the image sensor includes the landmark image described later, the operation management unit 112 has a second current position estimation function of estimating the current position based on the coordinate information of the landmark image. The route generation unit 113 generates an autonomous driving route without communicating with the RM server 1 by using the environmental map stored in the storage device of the robot control unit 11. The address of the storage location of the product to be the loading / unloading position is managed by the inventory management unit 2a in association with the product ID. The loading / unloading position for loading and unloading from a transport vehicle such as a truck is managed by the address on the basic map. When a loading / unloading instruction is issued, the route generation unit 113 generates a route (global route) from the loading / unloading position to the loading / unloading position based on the basic map and the partial map, and corrects the surrounding route (local route) of the automated guided vehicle 10 as necessary when the partial map is updated while driving.

[0021] The sensor management unit 114 detects observation signals from the front sensors (31a to 31c), the side sensors (32a, 32b), and the image sensor 33, and transmits them to the map management unit 111, the operation management unit 112, and the like. The three front sensors (31a to 31c) provided in the front are distance sensors (for example, ultrasonic sensors), and detect an object existing in the front in a range of, for example, 180 degrees or more. The two side sensors (32a, 32b) provided on the side are distance sensors (for example, ultrasonic sensors), and detect an object existing on the side in a range of, for example, 120 degrees or more. When the five sensors (31a to 31c, 32a, 32b) provided in the front and on the side detect an obstacle (including a person), the operation management unit 112 that has received the detection information from the sensor management unit 114 performs control to avoid the obstacle or temporarily stop the driving when it is difficult to avoid the obstacle.

[0022] The virtual image generation unit 115 has a composite image generation function for dynamically generating a composite image with a virtual image added to the captured image of the image sensor 33. For example, by dynamically generating a composite image with an image of a virtual line added, a line following function described later can be executed based on the added virtual line, or by dynamically generating a composite image with an image of a mark indicating a turning position added, turning control can be performed based on the added virtual mark. Also, in an environment where external light is incident, driving control can also be performed by dynamically generating a composite image with a virtual landmark image added. In these cases, since it is assumed that the current position information held by the operation management unit 112 is correct, it is preferable to adopt a configuration in which the current position is acquired by using complementary information based on artificial code information such as an AR marker, a QR code (registered trademark), or a barcode with known coordinates.

[0023] FIG. 12(a) is a captured image of the image sensor 33, (b) is a composite image with a virtual image added to the captured image of the image sensor 33, and (c) is a planar schematic diagram of (b). In the composite image shown in FIG. 12(b), a virtual guidance line 81 and a virtual control mark 82 are added to the captured image of the image sensor 33. Based on the added virtual guidance line 81 and virtual control mark 82, the operation management unit 112 performs driving control. For example, it discloses performing line following so that the virtual guidance line 81 is located at the center of the base 16 and performing driving control to turn right 90 degrees at the virtual control mark 82. When the landmark image in the captured image changes due to the driving of the automated guided vehicle 10, the current position of the automated guided vehicle 10 is corrected in real time. Therefore, even if the drawing positions of the virtual guidance line 81 and the virtual control mark 82 are inaccurate, the drawing positions are corrected as the automated guided vehicle 10 moves.

[0024] The image sensor 33 is composed of a stereo camera that captures a two-dimensional image placed at a distance. The stereo camera may be configured by fixing two independent cameras, or may be configured by one that integrally includes two lens mechanisms and shutter mechanisms. The image sensor 33 is preferably composed of a color camera so as to be able to recognize color information. The image sensor 33 is attached to the front side of the base 16, captures a two-dimensional image in front of the automated guided vehicle 10, and transmits it to the map management unit 111. The map management unit 111 generates a three-dimensional image having distance information based on the captured image of the image sensor 33. In addition, the map management unit 111 also has a function of registering, as landmark images, images of objects having specific shapes or patterns or images of specific shapes or patterns projected on the floor surface, wall surface, etc., in association with coordinates. As landmarks composed of physical objects, it is possible to register protective signs such as triangular cones (safety cones) and pillars, it is also possible to register color tapes and hanging signs, etc., and it is also possible to register two-dimensional codes such as AR markers and QR codes (registered trademarks) and information signs such as emergency exit plates. In addition, it is also possible to use, as landmarks, images of specific shapes or patterns projected from the landmark image projection device 7 onto the floor surface, wall surface or ceiling. The projection of the image from the landmark image projection device 7 can also be performed at specified coordinates. A server for managing landmark images may be installed at the workplace or in the cloud, and the landmark information (image and coordinate information) may be shared by a plurality of automated guided vehicles 10 arranged within the workplace. The protective signs and information signs are preferably colored with a color that provides high contrast with the background of the workplace.

[0025] The display unit 12 is composed of a touch screen and also serves as an information input unit. In the example of the embodiment, the display unit 12 is composed of a tablet-type terminal on which a program capable of communicating with the robot control unit 11 operates. The recommended route created by the route generation unit 113 is displayed on the display unit 12. When the operator approves the route, the route generation unit 113 converts the approved route into array data. If the recommended route is not preferable, a route change instruction can be issued, and another route created by the route generation unit 113 is displayed. The wireless communication unit 13 enables bidirectional communication with the handy terminal 4 and the RM server 1 via a wireless LAN (e.g., Wi-Fi). The communication modules of the handy terminal 4 and the RM server 1 may be configured by separate modules, and the communication with the RM server 1 may be configured to be performed via an LPWA (Low Power Wide Area Network). The speaker 14 is composed of a general-purpose speaker, and a melody is emitted to notify that the unmanned carrier vehicle 10 is in motion during its travel. When the unmanned carrier vehicle 10 stops due to an abnormality, an alarm notifying the abnormality is emitted from the speaker 14. Note that instead of emitting an alarm during an abnormal stop, an image notifying the abnormality may be displayed on the display unit 12, or a configuration may be adopted in which an abnormality is notified by a separately provided indicator light. The power supply unit 15 includes a rechargeable secondary battery and enables the unmanned carrier vehicle 10 to travel without a cable. In addition to a contact-type charging device, the power supply unit 15 is provided with a power receiving device at the bottom surface portion or the side surface portion that enables non-contact charging by positioning the unmanned carrier vehicle 10 in the charging area. Note that non-contact charging is not an essential configuration, and a configuration including only a contact-type charging device may be adopted.

[0026] The drive system (21 - 23) is composed of a drive control device 21, drive devices 22a, 22b, and wheels 23a, 23b. The drive control device 21 is a computer that independently controls the driving of the driving devices 22a and 22b based on instructions from the robot control unit 11. The direction change of the automated guided vehicle 10 is performed by changing the rotational angular velocity of the wheel 23a by the right driving device 22a and the rotational speed of the wheel 23b by the left driving device 22b. The driving devices 22a and 22b are configured to include general-purpose motors such as DC motors, AC motors, motors with encoders, and geared motors, and can rotate the wheels 23a and 23b at arbitrary rotational angular velocities. The driving devices 22a and 22b can carry a load of 500 kg at a speed of, for example, 50 m per minute or less. Different from the embodiment example, a steering device for steering the wheels 23a and 23b may be separately provided in the driving device. In this case, the above-mentioned odometry information also includes steering angle information. In the embodiment example, tires suitable for traveling on a floor surface coated with epoxy resin or the like are adopted for the wheels 23a and 23b. The diameter, width, and presence or absence of a tread pattern of the wheels 23a and 23b can be optimally selected according to the weight of the article to be transported and the condition of the floor surface.

[0027] The management terminal 3 is a terminal such as a notebook PC, a desktop PC, or a tablet, and is used for the management of the RM server 1, the WMS server 2, etc. It may be possible to directly send a projection command to the landmark image projection device 7 from the management terminal 3. The handy terminal 4 is a portable terminal used for reading barcodes during picking operations. The handy terminal 4 includes a display unit composed of a liquid crystal screen or the like, a storage unit composed of a storage device or the like, a control unit composed of a processor, a built-in clock, etc., an operation unit composed of buttons or the like, a barcode reader unit, a wireless communication unit, and a secondary battery. In this embodiment example, an operation unit is provided separately from the display unit, but the display unit may be configured as a liquid crystal touch screen capable of information input. The handy terminal 4 in a preferred embodiment can wirelessly communicate with the robot control unit 11 of the automated guided vehicle 10. The wireless LAN master unit 5 is arranged within the floor of the workplace, providing an environment where access to a wireless LAN (e.g., Wi-Fi) is possible anywhere within the floor.

[0028] Figure 4 is a plan view for explaining the travel path of a warehouse according to an embodiment example. In the figure, the shaded areas are the travel paths. At the boundary between the floor travel path and the work area, a floor line 71 is partially projected by the landmark image projection device 7. For locations where it is difficult to project the floor line 71, the floor line 71 may be formed by color tape (e.g., tiger pattern tape) or painting. In the figure, the boundary line 72 shown by a dotted line is the boundary between the travel path and the work area, indicating the part that serves as a passage for people and forklifts. It is preferable to project an image of a different color, different shape, or different pattern from the floor line 71 onto the boundary line 72 by the landmark image projection device 7. In the embodiment example, the floor line 71 and the boundary line 72 are made distinguishable by forming them with lines of different colors. The projected boundary line 72 does not deteriorate even when people or forklifts pass through, and also helps people to visually recognize the work space. Also, by using the boundary line 72 as a landmark image, the landmark information in the travel path increases, making it possible to improve the accuracy of autonomous driving. Further, the floor guidance line 76 described later may be projected by the landmark image projection device 7 onto the main part of the travel path.

[0029] As shown in Figure 5, color tape (e.g., tiger pattern tape) 75 for prompting the operator's attention is partially pasted at a height (e.g., lower than a person's back) that is reflected in the image sensor 33 of the wall along the travel path. Instead of the color tape 75, a color line may be formed by painting. At locations necessary for travel control, a wall guidance line 77 and a ceiling guidance line 78 described later are projected by the landmark image projection device 7.

[0030] In the illustrated example, any one of areas A, B, and C can be set as the starting point and the goal point. What is indicated by reference numeral 73 is a support column about waist height, and the support column is provided with coloring (for example, a tiger pattern tape). What is indicated by reference numeral 74 is a color cone (registered trademark. Omitted hereinafter). As shown in FIG. 5, a two-dimensional code 79 such as an AR marker is attached to a part of the color cone 74. A color tape 75 is pasted on the corner of the column in contact with the traveling path in the vertical direction. The color tape 75 can also be replaced with painting, or an image of a specific shape or pattern may be projected by the landmark image projection device 7. The support column 73, the color cone 74, and the color tape 75 also function as signs for workers. Thus, since signs for workers are installed in the space where the unmanned carrier vehicle 10 travels, the unmanned carrier vehicle 10 uses this as feature point information and uses it to grasp the current position.

[0031] The automated guided vehicle 10 according to this embodiment also has a function of detecting the guiding lines 76 and 77 provided on the floor or wall and autonomously traveling. Specifically, the automated guided vehicle 10 first recognizes, in the captured image captured by the image sensor 33, the floor guiding line 76 formed on the traveling path of the floor surface and the wall guiding line 77 formed on the wall surface. Here, FIG. 6 is a diagram for explaining a method of recognizing whether the guiding line detected in the area in the captured image is a line formed on the floor surface or a line formed on the wall surface in this embodiment. As shown in FIG. 6, in this embodiment, the operation management unit 112 of the automated guided vehicle 10 sets the area below the center of the screen of the captured image as the floor guiding line recognition area, and sets the area excluding the lower side of the center and the upper part of the screen of the captured image as the wall guiding line recognition area. Then, when the operation management unit 112 detects a line image having specific color information in the floor guiding line recognition area, it recognizes that the line is the floor guiding line 76 formed on the floor surface, and performs travel control based on the floor guiding line 76. Further, when the operation management unit 112 detects a line image having specific color information in the wall guiding line recognition area, it recognizes that the line is the wall guiding line 77 formed on the wall surface, and performs travel control based on the wall guiding line 77. As shown in FIG. 6, when both the floor guiding line 76 formed on the traveling path of the floor surface and the wall guiding line 77 formed on the wall surface are detected, it is possible to perform travel control according to both lines 76 and 77, or to cause a specific vehicle to perform first travel control based on the floor guiding line 76, and to cause another vehicle to perform second travel control based on the wall guiding line 77.

[0032] FIG. 7 is a diagram for explaining a travel control method using a floor guidance line 76 formed on the floor surface. For example, when the automated guided vehicle 10 detects the floor guidance line 76 on the travel path of the floor surface, as shown in FIG. 7(A) for example, it can be equipped with a line following function that performs travel control so as to travel along the center line in the longitudinal direction of the detected floor guidance line 76. Alternatively, as shown in FIG. 7(B) for example, the automated guided vehicle 10 can perform travel control so as to regard the detected floor guidance line 76 as the edge of the passage and not exceed the floor guidance line 76, or travel at a certain distance away from the floor guidance line 76. Here, the floor line 71 or the boundary line 72 may be used as the floor guidance line 76. FIG. 8 is a diagram for explaining a travel control method using a wall guidance line 77 formed on the wall surface. Further, when the automated guided vehicle 10 detects the wall guidance line 77 on the wall surface along the travel path, as shown in FIGS. 8(A) and 8(B), it can perform travel control so as to travel along the detected wall guidance line 77 at a position a certain distance away from the wall guidance line 77. Here, the color tape 75 pasted on the wall may be used as the wall guidance line 77.

[0033] Furthermore, in the present embodiment, in addition to the function of registering the images of the color tape attached to the ceiling, the hanging signboard suspended from the ceiling, the beam, the lighting fixture, etc. as landmark images in association with the coordinates, the automated guided vehicle 10 is also equipped with a function of detecting the guidance line provided on the ceiling and performing autonomous travel. FIG. 9 is a diagram for explaining a travel control method using a ceiling guide line 78 provided on the ceiling. As shown in FIG. 9, the operation management unit 112 of the automated guided vehicle 10 sets an area above the center of the upper side of the captured image screen as the ceiling recognition area. When the operation management unit 112 detects a line image in the ceiling recognition area, it recognizes that the line is the ceiling guide line 78 formed on the ceiling, and performs travel control based on the ceiling guide line 78 formed on the ceiling. For example, the automated guided vehicle 10 can be equipped with a line following function for performing travel control so as to travel along the ceiling guide line 78 on the center line in the longitudinal direction of the detected ceiling guide line 78. It is also possible to perform travel control in combination with travel based on the ceiling guide line 78 and the landmark information of the ceiling (for example, a hanging signboard suspended from the ceiling) so as to travel to a position directly above the hanging signboard. In particular, in facilities such as hospitals with low ceilings, it is useful to perform travel control based on the ceiling guide line 78 formed on the ceiling.

[0034] Note that the floor line recognition area and the wall line recognition area shown in FIG. 6, and the ceiling recognition area shown in FIG. 9 are examples, and the range can be changed as appropriate. Also, when the automated guided vehicle 10 estimates the current position from each line (71, 72, 76 to 78) as feature point information, as described above, the line recognized in the floor line recognition area of the captured image is recognized as the floor line 71, the boundary line 72, or the floor guide line 76 formed on the floor surface, the line recognized in the wall line recognition area of the captured image is recognized as the wall guide line 77 formed on the wall surface, or the line recognized in the ceiling recognition area of the captured image is recognized as the ceiling guide line 78, so that the current position can be estimated.

[0035] In the above-described example, in order to perform highly accurate and stable driving control on a driving route such as a narrow passage or a curve, struts 73, color cones 74, color tapes 75, and a signboard (not shown) attached to the ceiling are used as feature point information to estimate the current position. However, the present invention is not limited to this. For example, in order to be able to charge in the vicinity of a non-contact charging device, a configuration for estimating the current position in the vicinity of the non-contact charging device can also be adopted. For example, in FIG. 4, reference numeral D denotes a charging area, reference numeral E denotes an entrance / exit, and reference numeral F denotes a loading / unloading location. In the charging area D, a power transmission device (not shown) for performing non-contact charging is arranged. However, in the charging area, charging cannot be performed unless the AGV 10 is guided with an accuracy on the order of cm. Therefore, a floor guide line 76 for guiding the AGV 10 to the power transmission device is provided. The floor guide line 76 may be formed of a color tape, may be formed by painting, or may be projected by a landmark image projection device 7. Here, the color tape only needs to be a visible tape, and includes a translucent tape or a patterned tape. Further, the floor guide line 76 may be a dotted line. When the image sensor 33 detects a line image, the operation management unit 112 has a line following function for guiding the AGV 10 along the line. As an implementation form of the line following function, for example, it is disclosed that driving control is performed by extracting preset color information and recognizing the shape of the line. By this line following function, it is possible to guide with an accuracy on the order of cm so that the power receiving device of the AGV 10 is positioned above the power transmission device. The loading / unloading location F is also a location where a transport vehicle such as a truck departs and arrives, and there is an incidence of external light.

[0036] The AGV 10 travels while performing route correction at regular intervals. FIG. 10 is a flowchart of route correction during travel. STEP601: The robot control unit 11 acquires image information in the traveling direction and its periphery by means of an image sensor, and performs feature point recognition on the image information. Details of the feature point recognition will be described later. STEP602: If the robot control unit 11 can obtain the current position, it proceeds to STEP603; if it cannot obtain the current position, it proceeds to STEP606.

[0037] STEP603: The robot control unit 11 updates the partial map corresponding to the obtained current position based on the image information. STEP604: The robot control unit 11 determines whether correction of the set route is necessary. If necessary, it proceeds to STEP605; if not, it returns to STEP601. STEP605: The robot control unit 11 corrects the set route and stores it in the storage device.

[0038] STEP606: The robot control unit 11 obtains the odometry information from the storage device and calculates the current position based on the current position information and the odometry information stored immediately before in the storage device. STEP607: If the robot control unit 11 can obtain the current position, it proceeds to STEP608; if it cannot obtain the current position, it proceeds to STEP609. STEP608: The robot control unit 11 updates the partial map corresponding to the obtained current position based on the image information. STEP609: Stop the automated guided vehicle 10, display a screen notifying an abnormality on the display unit 12, or emit an alarm notifying the stop from the speaker 14. STEP610: The robot control unit 11 determines whether correction of the set route is necessary. If necessary, it proceeds to STEP605; if not, it returns to STEP601.

[0039] A supplementary explanation is given regarding the feature point recognition in STEP601. In a warehouse where the objects installed around are constantly changing, feature points are extracted in a form including objects that are constantly moved such as a cage cart and a commodity box, so misdetection is likely to occur. Therefore, in the embodiment, when the current position cannot be obtained by extracting feature points based on the image information, complementary information such as landmark information and artificial code information is used in combination to obtain the current position. FIG. 11 is a flowchart of obtaining the current position using complementary information in combination.

[0040] STEP701: The robot control unit 11 performs feature point recognition on the traveling direction and the surrounding image information acquired by the image sensor, compares with the basic map and the partial map, and attempts to obtain the current position. STEP702: If the current position can be obtained, proceed to STEP706; if the current position cannot be obtained, proceed to STEP703. STEP703: Based on the result of feature point recognition for the acquired image information, detect the complementary information registered in association with coordinates in advance. For example, read the coordinate information of a specific AR marker extracted from the image information, and calculate the current position based on the position and size of the AR marker in the image. STEP704: If the current position can be obtained, proceed to STEP706; if the current position cannot be obtained, proceed to STEP705. STEP705: Stop the automated guided vehicle 10, display a screen notifying an abnormality on the display unit 12, or emit an alarm notifying the stop from the speaker 14. STEP706: The robot control unit 11 stores the acquired current position in the storage device.

[0041] The automated guided vehicle 10 of the embodiment described above can perform autonomous driving based on the projected image, the surrounding image information, and an inexpensive obstacle sensor without installing a guiding cable or a guiding tape. Therefore, even in an existing warehouse, it is possible to construct an unmanned transportation system with a minimum facility investment.

[0042] Also, by changing the projection position etc. of the projected image of the landmark image projection device 7, it is possible to easily respond to a layout change in the warehouse. Also, by using the projected virtual landmark image in combination, it is possible to improve the acquisition accuracy of the current position even in an environment where external light enters.

[0043] The preferred embodiments of the present invention have been described above. However, the technical scope of the present invention is not limited to the descriptions of the above embodiments. Various changes and improvements can be made to the above embodiments, and forms with such changes or improvements are also included in the technical scope of the present invention. In the embodiment, a configuration in which the automated guided vehicle 10 receives warehousing information and order information from the WMS server 2 has been described. However, an automated guided vehicle system that does not cooperate with the WMS server 2 may be constructed. For example, an inbound / outbound instruction may be transmitted to the automated guided vehicle 10 from a handy terminal 4 or a management terminal installed at the workplace. Also, it is not necessary to perform current position acquisition in the priority order described with reference to FIG. 10, and it is possible to set an arbitrary priority by the robot control unit 11 as to what information to prioritize for current position acquisition. For example, the priority of the distance sensor or the line following function may be prioritized over current position acquisition based on information from the image sensor 33.

[0044] Furthermore, in the above-described embodiment, a warehouse has been exemplified and described as an indoor facility where GPS cannot be used. However, the present invention is not limited to use in a warehouse, and can also be used in indoor facilities other than warehouses where articles are transported (for example, factories or hospitals where GPS cannot be used). In addition, the present invention is an invention for preventing an automated guided vehicle from stopping when it cannot appropriately recognize surrounding objects with a camera. As such a scenario, an environment in which disturbing light is incident has been exemplified and described. However, the scenario where surrounding objects cannot be appropriately recognized is not limited to an environment where disturbing light is incident. For example, when the quantity and arrangement of transported goods such as a cage car or luggage such as merchandise change depending on the season or time of day, or when people and forklifts coexist in the same environment, there is also an environment where the arrangement of surrounding articles changes with time. The present invention is also useful in such an environment.

[0045] Furthermore, the operation using the automated guided vehicle according to the present invention is not limited to the inbound / outbound operation, and can be applied to various operations that require the automated guided vehicle 10 to transport luggage.

[0046] In addition, in the above-described embodiment, a configuration having the image sensor 33 is exemplified. However, the number of image sensors 33 is not particularly limited, and it may be a configuration having a single image sensor 33 or a configuration having a plurality of image sensors 33. In the case of a configuration having a plurality of image sensors 33, based on the captured images in a plurality of directions captured by the plurality of image sensors 33, feature point information such as the support column 73, the color cone 74, the color tape 75, and a signboard (not shown) attached to the ceiling is recognized, and the current position is recognized and driving control is performed, thereby improving the recognition accuracy of the current position and the accuracy of driving control.

Explanation of Signs

[0047] 1 RM server 2 WMS server 3 Management terminal 4 Handheld terminal 5 Wireless LAN master unit 6 Cage car 7 Landmark image projection device 10 Automated guided vehicle 11 Robot control unit 12 Display unit 16 Base 21 Drive control device 22 Drive device 23 Wheels 31 Front sensor 32 Side sensor 33 Image sensor 71 Floor line 72 Passage line 73 Support column 74 Color cone 75 Color tape 76 Floor guidance line 77 Wall guidance line 78 Ceiling guidance line 79 Two-dimensional code 81 Virtual guidance line 82 Virtual control marker

Claims

1. An unmanned transport vehicle that autonomously travels along a travel route based on an environmental map, a landmark image projection device that projects a landmark image of a specific shape or pattern along the travel route, a server that transmits a projection command for projecting the landmark image to the landmark image projection device, an unmanned transport vehicle system comprising: the unmanned transport vehicle, a color image sensor that captures an image of an object on the forward direction side, a drive unit that independently controls the drive of a plurality of wheels, a robot control unit that transmits a control command for the autonomous travel to the drive unit, a wireless communication unit, and comprising: a function in which the robot control unit constantly updates an environmental map and a route based on image information from the color image sensor; a function of registering the projected landmark image in association with coordinate information on the environmental map; a first current position estimation function that performs feature point recognition on the image information from the color image sensor and estimates the current position by comparing with the environmental map; a second current position estimation function that estimates the current position based on the coordinate information of the landmark image when the landmark image is included in the image information from the color image sensor; and comprising: the landmark image projection device projects a floor line at the boundary between the travel route and the work area, projects a boundary line image of a color different from the floor line at a portion that is a passage at the boundary between the travel route and the work area, projects a floor guidance line having specific color information on the travel route, and projects a wall or ceiling guidance line having specific color information on a wall or ceiling; the unmanned transport vehicle includes a first unmanned transport vehicle and a second unmanned transport vehicle; the robot control unit of the first unmanned transport vehicle performs first travel control based on the floor guidance line, and the robot control unit of the second unmanned transport vehicle performs second travel control based on the wall or ceiling guidance line. An unmanned transport vehicle system characterized by that.

2. the drive unit has a function of transmitting the rotation information of the plurality of wheels to the robot control unit as odometry information; The unmanned transport vehicle system according to claim 1, wherein the robot control unit has a third current position estimation function of estimating the current position based on the odometry information and the currently stored current position information.

3. in an environment where external light is incident, the robot control unit has a function of dynamically generating a composite image with a virtual landmark image added to the image information from the color image sensor. The unmanned carrier vehicle system according to claim 1 or 2, further comprising a function of performing travel control based on a virtual landmark image included in the composite image.

4. The unmanned carrier vehicle system according to any one of claims 1 to 3, wherein the landmark image includes an image of an artificial code including a two-dimensional code.

5. The unmanned carrier vehicle system according to any one of claims 1 to 4, wherein the landmark image includes an image of a character string, a logo, or a character.

6. The unmanned carrier vehicle system according to any one of claims 1 to 5, wherein the server transmits a projection command for projecting an image for causing the first travel control to be performed in a first time zone, and transmits a projection command for projecting an image for causing the second travel control to be performed in a second time zone.

Images