Mobile body control method, mobile body, program, mobile body control system, mold change system

Abstract

To provide technology that enables overall control of mobile objects, from wide-area control to close-range control. [Solution] In a wide-area image including the area in which a mobile body moves, a wide-area map is generated in which a movement area indicating the area or path in which the mobile body can move and a target area indicating the vicinity of the object to be worked on by the mobile body are defined. Based on the wide-area image including the mobile body, the mobile body area in which the mobile body exists in the wide-area map is recognized. Within the coordinate system of the wide-area map, a path for the mobile body to the target area is defined based on the target area and the mobile body area. A first control signal is generated to guide the mobile body along the path. The system switches to a proximity control step to guide the mobile body, which has been guided to the target area, to the object to be worked on, and transmits the generated control signal to the mobile body.

Description

Technical Field

[0001] The present invention relates to a mobile control method for controlling the operation of a mobile body, a mobile body, a program, a mobile control system, and a mold replacement system.

Background Art

[0002] Conventionally, control technologies for moving mobile bodies such as automated guided vehicles and robots along a predetermined route to a destination have been widely used. These control technologies have been utilized for the purpose of automation and labor saving in industrial fields such as the transportation of objects to be transported, automatic driving in warehouses, and component supply in manufacturing lines.

[0003] In an autonomous mobile robot (AMR), a method of moving while recognizing the surrounding environment using sensors mounted on the mobile body is adopted. However, AMRs have problems such as being able to only grasp their own surrounding environment and having difficulty avoiding humans approaching from blind spots, being unable to respond to changes in the surrounding environment if the accuracy of self-position estimation is low, and the cost of teaching being high. In particular, in AMRs, even when a two-dimensional code or the like is used assistively for positioning, the positioning accuracy is generally about ±5 mm. When attempting to achieve higher positioning accuracy in AMRs, long hours of work by specialized personnel are required, and the period can range from several days to several months for one location.

[0004] In addition, for the position detection of a mobile body, there is also a method of installing physical markers such as magnetic tapes, induction lines, and two-dimensional codes on the floor or surroundings, and by reading these, self-position estimation and path following of the mobile body become possible. However, the method using physical markers has problems in terms of operation cost and flexibility because installation of the markers and repair work due to aging deterioration are required.

[0005] In response to this, a method has been proposed that uses image recognition technology to determine the current position of a moving object and controls it based on that position. Such technology allows for non-contact position detection using cameras and can respond relatively flexibly to changes in the environment, making it a promising means of achieving more advanced autonomous movement.

[0006] The automated guided vehicle (AGV) control system described in Patent Document 1 includes a camera that captures images of the area in which the AGV moves, and a control means that recognizes and controls the position of the AGV using the captured images. The control means has a map setting unit that sets a map of the travel area, a map storage unit that stores the map, a position recognition unit that recognizes the position of the AGV on the map, a destination setting unit that sets a destination, and a communication unit that communicates with the AGV. This system makes it possible to move the AGV along a predetermined transport route to a destination without requiring work such as setting up or repairing the guided vehicle. [Prior art documents] [Patent Documents]

[0007] [Patent Document 1] Japanese Patent Publication No. 2018-092393 [Overview of the project] [Problems that the invention aims to solve]

[0008] While the aforementioned prior art documents are advantageous for achieving wide-area control of moving objects within a given area, a challenge they face is their inability to handle precise control (e.g., on a millimeter-by-millimeter basis).

[0009] In view of the circumstances described herein, the first problem to be addressed is to provide a technology that enables overall control of a moving object, from wide-area control to proximity control. The second problem to be addressed is to provide a technology that enables control of a moving object that performs mold changes for a molding machine. [Means for solving the problem]

[0010] To solve the above problems, the mobile body control method according to this disclosure is: A method for controlling a mobile body having a drive device, Computers A map generation step that generates a wide-area map in which a moving object moves, in a wide-area image that includes an area where the moving object moves, a moving area that indicates the area or path the moving object can move, and a target area that indicates the area around the object to be worked on by the moving object. A recognition step of recognizing a mobile body region that represents the location and orientation of the mobile body in the wide-area map based on the wide-area image including the mobile body, A wide-area control step includes defining a path for the moving object to the target area based on the target area and the moving object area within the coordinate system of the wide-area map, and generating a first control signal for guiding the moving object along the path, The process involves transmitting the generated control signal to a mobile object, and then executing the transmission process. The wide-area control process switches to a proximity control process for guiding the moving body, which has been guided to the target area, to the work target.

[0011] Furthermore, the mold exchange system relating to this disclosure is A mold change system that autonomously performs mold change operations, It comprises a mobile body equipped with a mold changing device and a control device for controlling the mobile body, The control device is A map generation unit generates a wide-area map in which a moving object moves within a wide-area image that includes the region in which the moving object moves, and a target region that indicates the periphery of the molding machine. A recognition unit recognizes a mobile body region that represents the location and orientation of the mobile body in the wide-area map based on the wide-area image including the mobile body, A wide-area control unit that defines a path to the target area based on the target area and the moving body area, and generates a first control signal for guiding the moving body along the path, A transmission unit that transmits the generated control signal to the moving body. The moving body is guided to the target area based on the first control signal, and based on a second control signal for controlling the moving body in the target area, conveys a mold to the molding machine and performs the replacement work of the mold.

Advantages of the Invention

[0012] According to the present invention, it is possible to provide a technology for realizing overall control from wide - area control to proximity control of a moving body. Further, according to the present invention, it is possible to provide a technology for realizing control of a moving body that performs mold replacement for a molding machine.

Brief Description of the Drawings

[0013] [Figure 1] An outline diagram of the mobile body control system of the present embodiment. [Figure 2] A hardware configuration diagram of the control device of the present embodiment. [Figure 3] A hardware configuration diagram of the moving body of the present embodiment. [Figure 4] A functional block diagram of the control device of the present embodiment. [Figure 5] A processing flowchart of the mobile body control method of the present embodiment. [Figure 6] An outline diagram of the generation process of the wide - area image of the present embodiment. [Figure 7] An outline diagram of the wide - area map of the present embodiment. [Figure 8] A processing flowchart of the proximity control of the present embodiment. [Figure 9] An outline diagram of the docking process of the present embodiment.

Modes for Carrying Out the Invention

[0014] The following description will use drawings to explain a mobile body control system and a mobile body control method according to embodiments of the present invention. Note that the embodiments shown below are examples of the present invention, and the present invention is not limited to these embodiments; various configurations can be adopted.

[0015] In this embodiment, the configuration and operation of the mobile control system and control device will be described, but control methods, computer programs, and program recording media on which such programs are recorded will also have similar effects. Using a program recording medium, for example, the program can be installed on a computer. The series of processes according to this embodiment, described below, are provided as a program executable by a computer and can be provided via non-transient computer-readable recording media such as CD-ROMs and flexible disks, as well as via communication lines.

[0016] The control system consists of a computer unit and a mobile unit. The computer unit includes an arithmetic unit such as a CPU (Central Processing Unit) and a memory device. The computer unit can function as a control device by executing a program stored in the memory device using its arithmetic unit. The mobile unit operates based on signals transmitted from the control device. The control method is realized through the processing of each device, including the control device and the mobile unit.

[0017] A mobile object refers to an object that can move spatially along a predetermined path or any trajectory, and specifically includes transport vehicles and robotic cars that travel on the ground or the floor of a factory, etc. Furthermore, the mobile unit is controlled autonomously, by signals transmitted from a control device, or a combination thereof, and may be equipped with multiple sensors, communication means, computing devices, etc.

[0018] 1. System Configuration Figure 1 shows an overview diagram of the mobile object control system 1. The mobile object control system 1 controls a mobile object 3 operating in a factory, warehouse, etc., and automates tasks such as the transport of goods. The mobile object control system 1 comprises a control device 2, a mobile object 3, a work target 4, multiple imaging devices 5A, 5B, and a communication device 6. The control device 2 is connected to the imaging devices 5A, 5B and the communication device 6 by wireless or wired communication. The mobile object 3 is connected to the communication device 6 by wireless communication. In the illustrated example, only one mobile object 3 and one work target 4 are shown, but there may be multiple of each.

[0019] The work object 4 is the object on which the mobile unit 3 performs work. The work object includes production equipment that produces products, machine tools that process workpieces, assembly lines that assemble parts, molding machines that mold plastic products, and storage shelves for goods. The mobile unit 3 can perform operations to load or unload various goods to and from the work object 4. Note that the work object 4 does not have to be an object on which the mobile unit 3 directly performs work. For example, if the work object 4 is a machine tool, the mobile unit 3 can transport the workpiece to the vicinity of the machine tool, and the workpiece can be installed on the machine tool by a worker or robot.

[0020] In this embodiment, the work object 4 is described as being configured as a molding machine. The molding machine is equipped with interchangeable molds according to the shape of the product to be produced. These molds are transported by a mobile body 3 carrying a mold changing device, and are loaded into or unloaded from the molding machine by operating the mold changing device. The molding machine is equipped with an unloading device that operates when the mold is changed.

[0021] The control device 2 recognizes the position and direction of the moving object 3 based on images captured by the imaging device 5, which is installed on the ceiling or the like. In this embodiment, the moving object 3 is equipped with a marker portion on its upper surface, such as a two-dimensional code or an LED (Light Emission Diode). The control device 2 can recognize the position and direction of the moving object 3 based on the marker portion in the captured image. The marker portion may also have identification information unique to the moving object 3. The marker portion may be formed by multiple LEDs, and information can be transmitted through the lighting / exit patterns of each LED, as well as combinations of colors.

[0022] The control device 2 generates control signals to operate the recognized mobile object 3. These control signals include, for example, control signals to guide the mobile object 3 to the area surrounding the work target 4, control signals to guide the mobile object 3 to the area adjacent to the work target 4, and control signals to decelerate or stop the mobile object 3. However, the control signals are not limited to these examples and may include various signals to operate the mobile object 3. The control device 2 transmits the generated control signals to the mobile object 3 via the communication device 6.

[0023] The mobile unit 3 receives a control signal generated by the control device 2 via the communication device 6. The mobile unit 3 moves by operating the drive device according to the received control signal. This allows the mobile unit 3 to move to, for example, the work target 4 and perform the task of loading or unloading goods.

[0024] The mobile body 3 and the work object 4 are each provided with a pair of connecting parts and a connected part, and are mechanically docked by connecting them. The mobile body 3 has connecting parts on one or more of its side surfaces that can be connected to the connected part of the work object 4. Conversely, the work object 4 has connected parts that can be connected to the connecting parts of the mobile body 3. The connecting parts may be provided on either the mobile body 3 or the work object 4, and the connected part may be provided on the other.

[0025] In this embodiment, the control device 2 is not limited to being installed outside the mobile body 3, but may be mounted on the mobile body 3. In that case, the control device 2 is connected to the communication device 6 by wireless communication and can receive images captured by the imaging device 5 via the communication device 6. Furthermore, the control device 2 is not limited to being implemented as an on-premise server in a factory, but may be implemented as an external cloud server or the like. In addition, the control device 2 may be configured as multiple computer devices, and for example, if a local server in a factory malfunctions, it may be configured to continue operation by switching to an external server.

[0026] 2. Hardware Configuration Figure 2 shows the hardware configuration diagram of the control device 2. The control device 2 comprises a processor 201, memory 202, and communication interface 203, with each component connected by a bus interface.

[0027] The processor 201 consists of one or more processors, such as a CPU, and controls the overall processing in the control unit 2 by executing programs, an OS (Operating System), and other applications. The memory 202 is a RAM (Random Access Memory), HDD (Hard Disk Drive), SSD (Solid State Drive), flash memory, etc., and stores programs and various data. The communication interface 203 is an interface for wired communication, wireless communication, etc., and controls data communication with external devices.

[0028] Figure 3 shows the hardware configuration diagram of the mobile unit 3. The mobile unit 3 comprises a processor 301, memory 302, communication interface 303, drive unit 304, positioning sensor 305, environmental sensor 306, and height adjustment mechanism 307, with each component connected by a bus interface.

[0029] The processor 301 consists of one or more processors, such as a CPU, and controls the overall processing in the mobile device 3 by executing programs, the OS, and other applications. The memory 202 is RAM, HDD, SSD, flash memory, etc., and stores programs and various data. The communication interface 203 is an interface such as wired communication or wireless communication, and controls data communication with external devices.

[0030] The drive unit 304 is a moving unit that performs at least movement-related operations on the mobile body 3. The moving unit consists of a motor, battery, casters, etc. In this embodiment, the moving unit consists of three or more casters that support the mobile body 3 and drive each caster with a motor to perform operations such as movement and rotation on the mobile body 3. The casters are rotatable in all directions, allowing the mobile body 3 to move with a high degree of freedom. The drive unit 304 may further include a work unit that performs work-related operations on the mobile body 3. The work unit consists of, for example, a chain, rollers, or arm for loading or unloading articles placed on a mounting surface, and a motor that drives these.

[0031] The positioning sensor 305 is a sensor that measures sensor values ​​used for positioning the mobile body 3 relative to the work object 4. The positioning sensor 305 can employ an imaging device, a distance measuring sensor, or the like. However, the positioning sensor 305 is not limited to these examples, and may employ, for example, LiDAR (Light Detection And Ranging), ToF sensor, stereo camera, radar, ultrasonic sensor, etc. The positioning sensor 305 is particularly used for positioning the joint of the mobile body 3 and the jointed part of the work object 4.

[0032] The environmental sensor 306 is a sensor that measures the surrounding environment of the mobile body 3. The environmental sensor 306 can employ an imaging device, LiDAR, ToF sensor, stereo camera, radar, ultrasonic sensor, etc. The environmental sensor 306 can measure objects and people present in the surrounding environment of the mobile body 3 as environmental information.

[0033] The height adjustment mechanism 307 is an electrically operated mechanism capable of adjusting the height of the entire mobile body 3 or a predetermined part of the mobile body 3 (for example, the mounting surface). In this embodiment, the height adjustment mechanism 307 allows for individual adjustment of the support height of multiple casters that constitute the mobile unit. The height adjustment mechanism 307 can correct the tilt of the mobile body 3. For example, the height adjustment mechanism 307 can make the mobile body 3 approximately horizontal to the floor surface, or make the mobile body 3 horizontal relative to the work object 4.

[0034] 3. Functional Components Figure 4 shows a functional block diagram of the control device 2. The processor 201 of the control device 2 functions as the image generation unit 21, map generation unit 22, recognition unit 23, wide-area control unit 24, proximity control unit 25, and transmission unit 26 by executing a program stored in the memory 202.

[0035] The transmitting unit 26 transmits control signals generated by the wide-area control unit 24 and the proximity control unit 25 to the mobile unit 3 via the communication device 6. The mobile unit 3 can operate by driving the drive unit 304 according to these control signals. The transmitting unit 26 generates and transmits the control signals to the mobile unit 3 in real time. In the following description, the involvement of the transmitting unit 26 will be omitted to the extent that it does not become particularly unclear.

[0036] The overall processing flow of the mobile object control method of this disclosure will be explained below with reference to Figure 5. Figure 5 shows a processing flowchart of the mobile object control method using the mobile object control system. More detailed functions of each functional component will be described later.

[0037] In the image generation process (S101), the image generation unit 21 generates a wide-area image based on the image obtained from the imaging device 5. The wide-area image is an image that includes the entire area of ​​the target, such as a factory, where the moving object 3 or the work object 4 is operated. The wide-area image is used for generating a wide-area map by the map generation unit 22 and for recognizing objects by the recognition unit 23.

[0038] In the map generation process (S102), the map generation unit 22 generates a wide-area map based on the wide-area image. The wide-area map is map information of a target area, such as a factory, where the mobile body 3 or work object 4 is operated. The wide-area map is used for object recognition by the recognition unit 23 and wide-area control of the mobile body 3 by the wide-area control unit 24.

[0039] In the recognition process (S103), the recognition unit 23 recognizes objects based on the wide-area image. The recognition unit 23 detects objects by image recognition of the wide-area image and recognizes them as areas indicating the current location of the objects on the wide-area map. Here, objects include moving objects 3, dynamic objects including people, and static objects including articles.

[0040] In the wide-area control process (S104), the wide-area control unit 24 generates control signals for wide-area control of the moving object 3 recognized by the recognition unit 23. Wide-area control includes controls for guiding the moving object 3 along a path from its current location to the target area, and controls for avoiding dangers such as collisions with objects in the vicinity of the moving object 3.

[0041] In the proximity control process (S105), the proximity control unit 25 generates a control signal for proximity control of the mobile body 3 when the mobile body 3 reaches its destination. Proximity control includes at least control to switch the mobile body 3, which has reached the target area by wide-area control, to proximity control. Proximity control may also include control to bring the mobile body 3 close to the work object 4 installed in the target area, or control to connect the mobile body 3 to the work object 4.

[0042] In the transmission process (not shown), the transmission unit 26 transmits the generated control signal to the mobile body 3. The transmission process is performed as needed by the transmission unit 26 during the wide-area control process and the proximity control process.

[0043] 4. Image generation process The image generation unit 21 synthesizes the images obtained from multiple imaging devices 5 to generate a single wide-area image. Figure 6 shows an overview of the process of generating a wide-area image using imaging devices 5A and 5B installed on the ceiling. According to Figure 6(a), imaging device 5A captures an image I1 of a portion of the target area. In Figure 6(a), the shaded area indicates a region not included in image I1. According to Figure 6(b), imaging device 5B captures an image I2 of a portion of the target area. In Figure 6(b), the shaded area indicates a region not included in image I2. Figure 6(c) shows the wide-area image I3 obtained by synthesizing images I1 and I2 by the image generation unit 21.

[0044] In this embodiment, the imaging device 5 employs a camera capable of capturing a wide area, such as a fisheye camera. The image generation unit 21 converts images obtained from multiple fisheye cameras into planar images and generates a wide-area image by combining these planar images. Note that the imaging device 5 is not limited to a fisheye camera, but may also be a wide-angle camera or a camera that captures planar images.

[0045] In this embodiment, the multiple imaging devices 5A and 5B are arranged in complementary positions so that an image of the entire target area can be obtained. Furthermore, it is preferable that the imaging devices 5A and 5B are arranged so that a portion of the area of ​​image I1 and image I2 overlap. For example, it is more preferable that the entire target area is imaged by at least two or more imaging devices 5. Note that the area corresponding to the edge of the target area is not limited to this. As a result, even if some of the multiple imaging devices 5 are unable to capture an image due to a malfunction or other reason, or if a blind spot occurs in some of the imaging devices 5 due to the operation of a crane on the floor, a wide-area image can be continuously generated by using images of the overlapping area captured by the other imaging devices 5.

[0046] 5. Map generation process The map generation unit 22 generates a wide-area map based on a wide-area image. The wide-area map is map information of a target area, such as a factory, where the mobile object 3 or work object 4 is operated. Preferably, the wide-area image used to generate the wide-area map is an image taken in a state where there are no objects that are not permanently placed within the target area, including the mobile object 3 and people. On the other hand, fixed objects within the target area, such as the work object 4, walls, and equipment, are included in the wide-area image as they constitute basic map information in the wide-area map. In this embodiment, the position information of the wide-area map is defined by a coordinate system, such as a Cartesian coordinate system (XY). However, the coordinate system applied to the wide-area map is not limited to a Cartesian coordinate system.

[0047] The map generation unit 22 defines attributes for each region included in the wide-area image. The map generation unit 22 can define the addition, updating, and deletion of attributes for each region by receiving user input via an operation interface of a terminal device (not shown) connected to the control device 2. Alternatively, the map generation unit 22 may recognize each object in the image through image recognition of the wide-area image and automatically define the addition, updating, and deletion of attributes for each region according to the type of object recognized.

[0048] The attributes defined for each region include, but are not limited to, attributes that restrict the movement of the mobile object 3, attributes that restrict the work of the mobile object 3, and attributes that indicate the type of device or object to be installed. By defining these attributes for regions, attributes can be assigned to desired regions of the wide-area image, and a wide-area map can be generated.

[0049] Figure 7 shows an overview of the wide-area map M1 in which each attribute is defined. In this embodiment, the wide-area map is defined to include at least a movement area that shows the area or path that the mobile body 3 can move, and a destination area that is the destination of the mobile body 3.

[0050] In Figure 7, the movement area is a passageway, etc., and is shown as the white area A10. In contrast, the movement restriction area, which indicates an area or path where the moving object 3 is prohibited from moving, is shown as the shaded areas A20 and A21 in Figure 7. Note that the movement restriction area can also be defined for passageways, etc., as in area A21.

[0051] In Figure 7, the target area is shown as a first target area A31-A35 defined as the area around the work object 4 requiring proximity control, and a second target area A36-A37 defined as destinations that do not require proximity control. The target area may be defined as a circular shape set by the coordinates of the destination and the radius centered at those coordinates, or as a rectangular shape set by the lengths of each side, and is not limited to the shape shown in the example. The first target area A31-A35 is defined as a destination for proximity control of the mobile body 3 relative to the work object 4. The second target area A36-A37 is defined as a destination for use, for example, as a relay point or waiting area for the mobile body 3. In the first target area, the position and orientation of the mobile body 3 at the destination are defined, and in the second target area, at least the position of the mobile body 3 at the destination is defined.

[0052] The first objective area is, in particular, the area where the mobile body 3 and the work object 4 are joined in close proximity, and includes the area of ​​the work object 4 having the part to be joined. For example, if the work object 4 is a molding machine, the mobile body 3 can join its joint to the part to be joined on the work object 4, and then load or unload a mold into or out of the molding machine.

[0053] In Figure 7, the work area is shown as the shaded area A41 to A46 where the work object 4 is installed. The extraction machine area is the area where the extraction machine for the output product (such as a finished product) of the work performed by the work object 4 (e.g., a molding machine) is installed, and is shown as the black rectangular area A51 to A55.

[0054] In the wide-area map, the first target area in particular defines proximity guidance information to enable smooth initiation of proximity control by the mobile body 3 toward the work object 4 and to enable high-precision docking. In this embodiment, the proximity guidance information includes the opposing position and the height adjustment value.

[0055] The opposing position is defined as a specific location within the target area, along with the orientation of the mobile body 3, such that the joint of the mobile body 3 and the jointed part of the work object 4 are precisely facing each other. This opposing position is the switching point from wide-area control to proximity control, and is the final destination point that the mobile body 3 should reach by horizontal movement in the coordinate system of the wide-area map.

[0056] The height adjustment value defines the height required to correct the posture (especially the tilt) of the mobile body 3 in the target area. In particular, factory floors have uneven surfaces and slopes, and the tilt of the mobile body 3 changes depending on its position on the plane. When the mobile body 3 transports heavy objects such as molds, even a slight tilt can cause the object to collide with the work unit or the replacement mechanism of the work object 4, potentially having a serious impact on durability, so tilt correction is extremely important. Furthermore, if the mounting accuracy of the mold to the molding machine, which is the work object 4, is low, it will affect the accuracy (quality) of the molded product and the durability of the mold, so tilt correction is also important. Posture adjustment information is set based on the local tilt of the floor surface measured for each target area, and the relative height measurements of the joint and joined parts. In addition, the system may automatically update the posture adjustment parameters based on the docking history (success or failure of docking, etc.) to improve the correction accuracy.

[0057] This document shows an example of defining a virtual map for mold changes in a molding machine. This virtual map defines areas such as the automated mold storage area, the mold temperature control area, and the mold maintenance area. The automated mold storage area is defined as a warehouse for storing multiple molds. The mold temperature control area is defined as an area for pre-conditioning molds that have been removed from the automated mold storage area. The mold maintenance area is defined as an area for maintaining molds that have been removed from the molding machine. Note that the areas defined in the virtual map are not limited to these examples, and areas with desired attributes can be defined.

[0058] The image generation process and map generation process described above generate a virtual map for controlling the mobile object 3. The following explanation describes each step for controlling the mobile object 3 using this virtual map.

[0059] The memory 202 of the control device 2 stores the action plan for the mobile unit 3. The control device 2 can acquire production tasks from a production management system (not shown) and generate an action plan. The action plan may also be set via the operation interface of the user's terminal device.

[0060] A production task includes a series of instructional information regarding the production that the mobile unit 3 should perform. Specifically, a production task includes data such as identification information of the work object 4 (production equipment such as a molding machine), identification information of the items to be supplied to the work object 4 (molds, workpieces, parts, etc.), the scheduled start time of production, and the target production quantity. The control device 2 obtains these production tasks from the production management system and automatically generates an action plan for the mobile unit based on the information contained in the task. This action plan includes, for example, a series of task contents such as retrieving a predetermined mold from the automated warehouse, transporting the mold to a predetermined molding machine, and performing a mold change operation, as well as target values ​​for the execution time of each step. The task contents have a target area corresponding to the molding machine, and a path from the current position of the mobile unit 3 to the target area can be defined within the coordinate system of the wide-area map. This path is defined taking into account the position and orientation of the mobile unit 3 when it reaches the target area.

[0061] Furthermore, production tasks or action plans may be updated based on production performance data. Production performance data is data collected based on the operating history of production equipment such as molding machines, and includes the actual time (actual time) relative to the scheduled production start time / scheduled production completion time, production quantity, number of good / defective products, error codes and minor stoppage information of the production machine, and history of changes in production conditions. The control device 2 acquires this performance data from the molding machine or a higher-level system (such as a production management system) and re-evaluates and optimizes the action plan based on that performance. For example, if the control device 2 finds performance data showing a high number of error codes for a particular molding machine, it can update the subsequent action plan, such as lowering the priority of mold change work for that molding machine. As a result, the mobile body control system 1 assigns the mobile body 3 the role of replacing humans in the production process, realizing autonomous decision-making and operation within the smart factory.

[0062] 6.Recognition process The recognition unit 23 recognizes objects in the wide-area map based on the wide-area image generated in real time by the image generation unit 21. Here, the objects to be recognized can include any objects that were not defined as areas in the generated wide-area map and are predefined as attribute information. Objects include not only the moving body 3, but also static and dynamic objects. Static objects are objects that are stationary on their own and include things such as articles, tools, and devices. Dynamic objects typically include people such as workers, and also include objects such as autonomous vehicles, robots, and animals that can move on their own.

[0063] The recognition unit 23 can recognize regions corresponding to the attributes of objects by referring to attribute information stored in the memory 202 and performing image recognition processing (for example, processing using artificial intelligence such as pattern recognition, machine learning, and deep learning) on ​​the wide-area image. In this embodiment, the recognition unit 23 recognizes moving object regions, static object regions, and dynamic object regions. The static object region may be distinguished according to the attributes of the static object (e.g., luggage, trolley, garbage). The dynamic object region may be distinguished in particular from humans and other dynamic objects (e.g., autonomous vehicles and robots). When the moving object 3 is transporting a heavy mold, emphasis is placed on preventing the danger of collision with humans.

[0064] The recognition unit 23 detects the current position and orientation (direction) of the moving object 3 by detecting a marker formed on the upper surface of the moving object 3 in the wide-area image. Based on the detection result of the moving object 3, the recognition unit 23 recognizes a moving object region that represents the position and orientation of the moving object 3 in the wide-area map. The recognition unit 23 can reflect the moving object region on the wide-area map in real time and monitor the position and orientation of the moving object 3.

[0065] It is preferable that the mobile body 3 is configured to hold the marker portion at a predefined marker height. The recognition unit 23 can detect the marker portion and calculate the planar coordinates of the mobile body 3 in the coordinate system of the wide-area map based on the predetermined marker height, thereby deriving the accurate current position of the mobile body 3.

[0066] In addition to recognizing the moving object region, the recognition unit 23 recognizes at least one of the static object region and the dynamic object region in the wide-area map based on a wide-area image that includes at least one of the static object region and the dynamic object region. The static object region corresponds to static objects such as stationary items and obstacles that are not predefined as impassable on the wide-area map but exist temporarily. The dynamic object region corresponds to dynamic objects whose positions change on the wide-area map, mainly people such as workers or other autonomous vehicles (including the moving object 3). The recognition unit 23 reflects the static object region and the dynamic object region on the wide-area map in real time and monitors at least their current positions. The recognition unit 23 may also recognize the direction of the dynamic object. The direction of the dynamic object is determined by image recognition or by the change in the dynamic object's position over time.

[0067] The image generation unit 21 can use AI (Artificial Intelligence) functions to generate wide-area images used in the recognition process. When processing images (video) from multiple imaging devices 5 in real time, the image generation unit 21 can adopt a configuration that dynamically changes the processing frequency (frame rate) for each region within the video in order to reduce the computational load on the control device 2. Specifically, the image generation unit 21 uses AI learning to distinguish between "static regions" such as molding machines and walls, and "dynamic regions" where moving objects and workers exist. It keeps the video update frequency of static regions low (for example, once every few seconds or several times per second), while processing dynamic regions at a high frequency (for example, 30 times or more per second). By combining this optimization of the update frequency for each region with narrowing down the regions of interest through AI learning, it becomes possible to reduce the reliance on advanced GPU (Graphics Processing Unit) servers and dramatically improve the overall processing efficiency and cost-effectiveness of the system.

[0068] 7. Wide-area control process 7.1. Wide-area control to the target region The wide-area control unit 24 controls the mobile object 3 over a wide area on the wide-area map. The wide-area control unit 24 defines a path from the mobile object area to the target area within the coordinate system of the wide-area map. The wide-area control unit 24 generates a first control signal to guide the mobile object 3 to the target area along the defined path. This path is defined considering the position and orientation of the mobile object 3 once it reaches the target area. The path is determined, for example, by identifying the target area based on identification information of the work object 4 to be used, which is included in the production task or action plan, and calculating it using a pathfinding algorithm. The pathfinding algorithm can be, for example, the A-star search algorithm or Dijkstra's algorithm, but is not limited to these. The mobile object 3 can receive this first control signal and drive the drive unit 304 to autonomously move to the vicinity of the target area.

[0069] The first control signal is a signal for driving one or more drive units 304 of the mobile body 3. The first control signal is defined as multiple motion commands such as forward / backward straight movement, left / right straight movement, left / right rotation, forward / backward rotation, thrust (lateral correction), horizontal movement (forward / backward / left / right), and rotation. Each motion command is defined as a multi-stage parameter, such as the speed of each drive unit 304 and the angle of each caster. The wide-area control unit 24 generates the first control signal by combining these motion commands and parameters. The first control signal may also include a signal for switching from the wide-area control process to the proximity control process when the mobile body 3 arrives in the target area. The second and third control signals, which will be described later, are also generated by combining these motion commands and parameters.

[0070] In this embodiment, when the wide-area control unit 24 guides the mobile body 3 from the mobile body area to the first target area, it refers to the opposing position of the proximity guidance information defined in the target area and generates a first control signal to guide the mobile body 3 to the opposing position in the target area so that the joining portion and the joined portion are in a predetermined position and orientation facing each other. The wide-area control unit 24 generates a first control signal to move the mobile body 3 horizontally within the coordinate system of the wide-area map to this opposing position. The opposing position becomes the starting point of proximity control, and when the wide-area control unit 24 guides the mobile body 3 to the opposing position, it outputs an instruction to switch to proximity control. When the mobile body 3 is guided to the opposing position, it receives an instruction to switch to proximity control and switches to proximity control mode. In the proximity control process described later, the joining work with the joined portion of the work object 4 is completed while moving the mobile body 3 from this opposing position toward the opposing position.

[0071] The wide-area control unit 24 may generate a control signal by referring to the height adjustment value of the proximity guidance information defined in the target area of ​​the wide-area map. The height adjustment value is used in proximity control to correct the joint and joined parts to be substantially horizontal, based on the inclination of the floor surface in the target area. The wide-area control unit 24 can generate the height adjustment value as part of the first control signal. The height adjustment mechanism 307 of the mobile body 3 corrects the inclination of the mobile body 3 according to the height adjustment value. This inclination correction may be performed in the wide-area control process or in the proximity control process described later.

[0072] 7.2. Wide-area control for ensuring safety The wide-area control unit 24 performs control to ensure safety around the moving object 3 based on the static object area or dynamic object area including people recognized by the recognition unit 23. The wide-area control unit 24 treats these object areas as areas that the moving object 3 should avoid and dynamically defines (replans) the path to the target area on the wide-area map.

[0073] Static object regions correspond to temporary object placements and unexpected obstacles that are not predefined on the wide-area map. The wide-area control unit 24 uses a pathfinding algorithm to define new routes that bypass (avoid) these object regions, treating them as temporary constraint regions that cannot be traversed. For dynamic object regions, particularly human regions, the wide-area control unit 24 defines routes that avoid them by considering their current location or predicted future movement locations. This allows the moving object 3 to move smoothly to its destination while reducing the risk of collisions with unexpected obstacles and workers, even during wide-area guidance.

[0074] The wide-area control unit 24 generates a third control signal to decelerate or stop the moving body 3 based on the distance between the area of ​​dynamic objects (especially people) and the area of ​​the moving body 3 in which the moving body 3 is located. For example, if a person enters a certain range in the direction of travel of the moving body 3, the wide-area control unit 24 can generate the third control signal to decelerate or stop the moving body 3.

[0075] The mobile body 3 is equipped with environmental sensors 306 for monitoring its surroundings. The mobile body 3 may be equipped with multiple environmental sensors 306 to measure the distance to objects present in all directions around the mobile body 3. In this case, it is preferable that the multiple environmental sensors 306 are installed, for example, in front of and behind the mobile body 3 in the direction of travel. Furthermore, the multiple environmental sensors 306 may be installed at different heights of the mobile body 3 to reduce blind spots in the vicinity of the mobile body 3. That is, it is more preferable that the environmental sensors 306 are installed in four locations each on the front and rear and upper and lower parts of the mobile body 3. These environmental sensors 306 measure the distance to objects in the vicinity of the mobile body 3 independently of the control signals from the wide-area control unit 24.

[0076] Based on the distance between the mobile body 3 and the object measured by the environmental sensor 306, if the mobile body 3 determines that there is a high risk of collision, it generates a third control signal to decelerate or stop the mobile body 3. The memory 302 of the mobile body 3 stores a distance threshold between the mobile body 3 and the object, and generates a third control signal when the object approaches a distance exceeding this distance threshold. The distance threshold may be defined as a variable value corresponding to the speed of the mobile body 3. The third control signal shall take precedence over the first control signal transmitted from the wide-area control unit 24 in the control of the mobile body 3. Furthermore, the third control signal may take precedence over the second control signal generated by the proximity control unit 25 in the control of the mobile body 3.

[0077] In this embodiment, the wide-area control unit 24 comprehensively manages the paths of multiple mobile bodies 3 (for example, a first mobile body 3A and a second mobile body 3B) and detects interference between these paths. If interference between the path of the first mobile body 3A and the path of the second mobile body 3B is detected by the wide-area control unit 24, it changes at least one of the paths of the first mobile body and the second mobile body to a path that does not interfere with each other. This change is determined based on the task priority and current driving status of each mobile body 3. The wide-area control unit 24 avoids path interference by, for example, defining the path of one of the mobile bodies 3 as a different path, or by temporarily stopping one of the mobile bodies 3.

[0078] 7.3. Autonomous Driving Mode If the wide-area control unit 24 cannot recognize the mobile body 3, which is included in the action plan as a target for control, based on the wide-area image, it generates an instruction for the mobile body 3 to switch to an autonomous driving mode using self-position estimation based on the measurement results of the environmental sensor. The mobile body 3 may also autonomously switch to an autonomous driving mode if the control signal from the wide-area control unit 24 is interrupted.

[0079] The environmental sensor 306 mounted on the mobile unit 3 is equipped with SLAM (Simultaneous Localization and Mapping) functionality. This SLAM function allows the mobile unit 3 to autonomously navigate while estimating its own current position based on the sensor's measurement results, and is used in autonomous driving mode. By switching to autonomous driving mode, the mobile unit 3 continues to move towards the target area based on its own local map and self-position estimation, without relying on the coordinate system of the wide-area map.

[0080] The autonomous driving mode is particularly effective in areas where it is difficult to install the imaging device 5 used for generating wide-area maps, or in environments where the accuracy of self-position estimation is stable. For example, this could include areas where human access is infrequent and there are few structures, or outdoor areas without high structures such as ceilings. By combining the autonomous driving mode, the mobile unit 3 can expand its operational area and continue moving using self-contained navigation.

[0081] 8. Proximity control process The proximity control process guides the mobile body 3, which has been guided by the wide-area control unit 24 to the opposing position in the target area, to the final work object 4 (such as a molding machine), and performs high-precision positioning for executing work (e.g., mold change or loading / unloading of goods). The proximity control process also includes control to join the mobile body 3 with respect to the work object 4 after it has been positioned. The second control signal is generated by the proximity control unit 25 of the control device 2 or by the processor 301 mounted on the mobile body 3.

[0082] The following description explains the operation of the proximity control unit 25 of the control device 2. It is also possible to have an embodiment in which the processor 301 of the mobile unit 3 is equipped with the proximity control unit 25 instead of the proximity control unit 25 of the control device 2.

[0083] The proximity control unit 25 generates a second control signal based on the measurement results of the positioning sensor 305 mounted on the mobile body 3. The positioning sensor 305 can be an imaging device or a distance measuring sensor, or any other sensor capable of measuring the relative positional relationship between the mobile body 3 and the work object 4 in real time through high-resolution measurement. The proximity control unit 25 feeds back the measurement results of this sensor and uses the second control signal to precisely control the drive unit 304 of the mobile body 3 to the millimeter or less (sub-millimeter units, etc.), thereby accurately positioning the mobile body relative to the work object 4.

[0084] In this embodiment, the wide-area control unit 24 may be triggered by the measurement result of the jointed portion of the work object 4 by the positioning sensor 305. Based on the first control signal from the wide-area control unit 24, the mobile body 3 moves horizontally to the opposing position of the joint and the jointed portion. The positioning sensor 305 is configured to switch the mobile body 3 to the proximity control process when it measures the position where the joint and the jointed portion are directly facing each other as the mobile body 3 moves horizontally. This switching can be performed by the wide-area control unit 24 or the proximity control unit 25 of the control device 2 or the processor 301 of the mobile body 3.

[0085] Figure 8 shows a flowchart of the process for joining the joint of the moving body 3 to the joint of the workpiece during the proximity control process. In this explanation, the horizontal direction (left and right) of the joint and the joint is described as the X-axis, the opposing direction (front and back) of the joint and the joint is described as the Y-axis, and the height direction (up and down) of the joint and the joint is described as the Z-axis. These coordinate systems may be local coordinate systems independent of the coordinate system of the wide-area map. In this local coordinate system, for example, the reference point of either the joint or the joint can be used as the center coordinate. By aligning the coordinates of the other joint or joint with the center coordinate, the relative positional misalignment between the two can be corrected and they can be joined.

[0086] In step S201, the positioning sensor 305 detects the part to be joined. The proximity control unit 25, having detected the part to be joined, starts joining the joining part to the part to be joined. In step S202, the proximity control unit 25 refers to the height adjustment value defined in the target area of ​​the wide-area map and generates a second control signal to correct the tilt of the mobile body 3 so that the joint and the joined parts are substantially horizontal. This second control signal for tilt correction is used to control the height adjustment mechanism 307 mounted on the mobile body 3, adjusting the tilt of the mobile body 3 so that it is approximately horizontal with respect to the floor surface. This corrects local external factors such as unevenness and slope of the floor surface, and enables joining while ensuring that the joint and the joined parts are horizontal in the height direction. In step S203, the proximity control unit 25 generates a second control signal to adjust the positional tilt misalignment between the joint and the joined part according to the detection result of the positioning sensor 305. Here, positional tilt misalignment includes positional misalignment along the XYZ axes and tilt misalignment with respect to the XYZ axes (roll, pitch, yaw). If there is a positional tilt misalignment, the proximity control unit 25 generates a second control signal that includes adjustment values ​​for the drive device 304 or the height adjustment mechanism 307 to correct these misalignments. In step S204, the proximity control unit 25 moves in opposition (Y-axis) to join the joint and the part to be joined. In step S205, the positioning sensor 305 detects positional or tilt deviations in real time. If a positional or tilt deviation is detected (NO in step S205), the process returns to step S203, the positional or tilt deviation is adjusted, and then the opposing movement is performed again. If no positional tilt misalignment is detected (YES in step S205), in step S206, it is determined whether the joining of the joint and the joined parts has been completed by opposing movement. Joining completion is determined, for example, by the amount of movement in the opposing direction or by sensors or switches on the joint or joined parts. If joining is not completed (NO in step S206), the process returns to step S204 and the opposing movement continues. If joining is completed (YES in step S206), the joining process is completed.

[0087] In this embodiment, the tilt correction by the proximity control unit 25 in step S202 may be performed after the completion of joining in step S206.

[0088] When the mobile unit 3 is joined to the work object 4, it performs tasks such as changing molds and loading / unloading items for the work object 4. The proximity control unit 25 may generate control signals for the work unit to perform these tasks.

[0089] Figure 9 is an schematic diagram illustrating the docking process by switching from wide-area control to proximity control. In Figure 9(a), the mobile body 3 is equipped with a joint 30, and the work object 4 is equipped with a jointed part 40. The mobile body 3 is guided to the first target area A30 according to the first control signal of the wide-area control unit 24. The arrows in the illustrated example indicate the direction of travel of the mobile body 3. The wide-area control unit 24 is A first control signal is generated to move the moving body horizontally (in the direction of travel in the illustrated example) to the opposing position of the joint 30 and the jointed portion 40 in the first target region A30.

[0090] In Figure 9(b), the mobile body 3 is guided to an opposing position in the first target area A30 according to the first control signal. The proximity control unit 25 generates a second control signal to switch the mobile body 3 to proximity control. The mobile body 3 is switched to proximity control according to the second control signal. Based on the second control signal, which causes the mobile body 3 to move from the opposing position to face the joint 30 and the jointed part 40 so as to join them, the mobile body 3 moves in the direction of travel (opposing direction) indicated by the arrow in the illustrated example.

[0091] In Figure 9(c), the joint portion 30 of the mobile body 3 and the part to be joined 40 of the work object 4 are joined. The mobile body 3 joins the joint portion 30 of the mobile body 3 to the part to be joined 40 of the work object 4 based on a second control signal for positioning the joint portion 30 of the mobile body 3 relative to the part to be joined 40. This second control signal is generated based on the measurement results of a positioning sensor 305 mounted on the mobile body 3.

[0092] 9. Application Examples In the mobile control system 1 described herein, the mobile body 3 is configured as an automated guided vehicle (AGV) that autonomously transports goods in a wide area such as a factory, warehouse, or logistics facility. The mobile body 3 is primarily used for transporting goods within a factory and assisting with the operation of production equipment.

[0093] The mobile unit 3 moves within a movable area or path in the factory based on control signals from the control device 2, and is used to transport items such as raw materials, parts, or products to the work object 4. The mobile unit 3 can also supply the transported items to the work object 4 and collect the results of the work performed by the work object 4. Furthermore, the mobile unit 3 can load or unload parts (such as molds) used by the work object 4.

[0094] As an example of its application, the mobile control system 1 is configured as a mold exchange system that controls a mobile body 3 equipped with a mold exchange device and autonomously performs mold exchange operations. The mobile body 3 moves to a target area set around the molding machine to be replaced, according to a control signal from the control device 2. The mobile body 3 connects to the molding machine according to a control signal from the control device 2 and drives the mold exchange device (work unit) to load and unload molds into and out of the molding machine, thereby performing mold exchange operations.

[0095] In a mold exchange system, the virtual map defines areas such as an automated mold storage area, a mold temperature control area, and a mold maintenance area. The mobile unit 3 can autonomously perform a series of production tasks by moving between these areas and the target area containing the molding machine, and performing predetermined tasks in each area. Furthermore, some of these areas, or the areas between them, do not need to be defined as a virtual map, and the mobile unit 3 may be controlled by an autonomous driving mode when moving through sections not defined as a virtual map.

[0096] This section will explain specific examples of how mold exchange systems are operated. First, in the automated mold warehouse area, molds are automatically transferred between the automated warehouse and the mobile unit 3. In the mold temperature control area, the mobile unit 3 pre-temperature controls the molds that have been removed from the automated warehouse. Once the pre-temperature control is complete, the mobile unit 3 loads the molds into the molding machine. After the molding operation by the molding machine is completed, the mobile unit 3 loads the molds out of the molding machine. The mobile unit 3 transports the loaded molds to the mold maintenance area and performs mold maintenance. In the mold maintenance area, once the mold maintenance is complete, the mobile unit 3 loads the molds into the automated warehouse and stores them there.

[0097] As an example of an application, the mobile unit 3 moves to a target area set around a machine tool (such as a lathe or machining center) on a production line, according to a control signal from the control device 2. The mobile unit 3 approaches the machine tool according to the control signal from the control device 2, and either supplies a workpiece before processing or retrieves a processed part from the machine tool. The mobile unit 3 may be equipped with a loading / unloading mechanism or an arm, for example, as a work unit for transferring workpieces or parts. In one application example, the mobile unit 3 moves to a target area set around each station of the assembly line according to a control signal from the control device 2. The mobile unit 3 approaches the assembly line according to the control signal from the control device 2, supplies parts used for assembly, or unloads assembled items after the assembly work is completed. The mobile unit 3 is equipped with, for example, an loading / unloading mechanism and an arm as a work unit for loading and unloading parts and items. These parts and items may be loaded into containers or similar vessels, and the containers may be loaded and unloaded in that manner.

[0098] 10. Effects This disclosure realizes a paradigm shift in navigation control by decoupleing the navigation basis of a moving object from its dependence on a real-space coordinate system to a coordinate system within a wide-area map (virtual map) generated by a camera. Specifically, control related to wide-area navigation, such as the recognition of wide-area images including the moving object, route setting, and generation of the first control signal, can be completed within the coordinate system of the wide-area map. This eliminates the need for complex and highly accurate coordinate transformation processing between real-space coordinate systems, which was unavoidable in conventional navigation technologies, significantly reducing the computational load on the control device and enabling faster control. Furthermore, because the wide-area map is managed in a virtual coordinate system linked to the wide-area image, robustness to environmental changes is improved compared to systems that depend on real-space coordinates. In addition, initial calibration and teaching are greatly simplified, resulting in easier system implementation and operation. Specifically, even with dynamic environmental changes that frequently occur in real space, such as temporarily leaving items in passageways or slight positional shifts of movable equipment, the navigation rules (static layer) on the wide-area map are not affected, eliminating the need for remapping or reteaching. This fundamentally solves the operational cost problem that was the biggest challenge with conventional SLAM methods, which required correction work by experts every time the environment changed, and realizes truly teaching-less autonomous driving operation. In addition, the system enhances the reliability of the mobile unit's operation by smoothly switching between wide-area control and proximity control, which requires precise movements, depending on the control accuracy required for each task. Furthermore, from the perspective of safety, which is a problem in the transportation of heavy objects, the system significantly reduces the risk of accidents through a multi-stage safety function that combines a safety function that quickly detects the approach of people based on wide-area control and restricts the movement of the moving object, and a safety function that restricts the movement of the moving object at close range using sensors mounted on the moving object. Furthermore, this system goes beyond mere automation of transportation; through close integration with production management systems, it contributes to the smartification of factories. Mobile units act autonomously based on production tasks, and real-time feedback of production performance data such as completion times and error information enables visualization of production progress and automation of budget-versus-actual management. This contributes to the construction of an advanced production system that manages and optimizes the entire production process without human intervention. Furthermore, in conventional AMRs, it was necessary to complete advanced computational processing such as self-localization estimation (SLAM), local path planning, and dynamic obstacle avoidance decisions within each mobile unit (onboard). This made it unavoidable to equip each mobile unit with expensive LiDAR sensors and high-performance computing processors, which drove up the implementation cost. In contrast, the present invention centralizes these advanced environmental recognition and control decisions in a wide-area map coordinate system on the infrastructure side (control device). This allows the mobile unit to be configured as a dependent unit that follows control signals from the server, significantly reducing its dependence on expensive sensors and computing power. As a result, it is possible to dramatically reduce the manufacturing cost per mobile unit while dramatically improving the overall cost performance and scalability of the system when operating a large number of mobile units throughout a factory.

[0099] The disclosures herein include the following: a method for controlling a mobile body, a mobile body, a program, a mobile body control system, and a mold change system.

[0100] Item [1] A mobile body control method for controlling a mobile body having a drive device, wherein a computer performs a map generation step of generating a wide-area map in which a mobile body moves, in which a mobile area indicating an area or path that the mobile body can move, and a target area indicating the vicinity of a work target by the mobile body are defined; a recognition step of recognizing a mobile body area representing the position and orientation of the mobile body in the wide-area map based on the wide-area image including the mobile body; a wide-area control step of defining a path for the mobile body to the target area based on the target area and the mobile body area within the coordinate system of the wide-area map, and generating a first control signal for guiding the mobile body along the path; and a transmission step of transmitting the generated control signal to the mobile body, wherein the wide-area control step switches to a proximity control step for guiding the mobile body, which has been guided to the target area, to the work target. Item [2] The mobile body control method according to [1], wherein in the proximity control step, a second control signal is generated for positioning the mobile body relative to the work object based on the measurement result of a positioning sensor mounted on the mobile body. Item [3] The mobile body control method according to [1] or [2], wherein the mobile body comprises a joint portion, the work object comprises a portion to be joined to the joint portion, and the wide-area control step generates a first control signal for guiding the mobile body to a facing position in the target area where the joint portion and the portion to be joined face to face. Item [4] The mobile body control method according to [3], wherein the wide-area control step generates a first control signal to move the mobile body horizontally to a facing position in the target area where the joint and the joined portion face each other, and the proximity control step generates a second control signal to move the mobile body from the facing position to join the joint and the joined portion. Item [5] A method for controlling a moving body according to any one of [1] to [4], wherein a plurality of target areas of the wide-area map have height adjustment values ​​defined for the joint and the joined portion, and in the proximity control step, a second control signal is generated to correct the tilt of the moving body based on the height adjustment values ​​so that the joint and the joined portion become substantially horizontal. Item [6] A method for controlling a moving object according to any one of [1] to [5], wherein the recognition step recognizes a static object region or a dynamic object region in the wide area map based on the wide area image which includes a static object or a dynamic object which includes at least a person, and the wide area control step defines the path to the target area to avoid the object region based on the static object region or the dynamic object region, or generates a third control signal which decelerates or stops the moving object based on the distance between the dynamic object region and the object region. Item [7] The mobile body control method according to any one of [1] to [6], wherein the mobile body is equipped with an environmental sensor that monitors the surroundings of the mobile body, and generates a third control signal that decelerates or stops the mobile body based on the distance between the mobile body and an object measured by the environmental sensor, and the third control signal takes precedence over at least the first control signal. Item [8] The mobile body control method according to any one of [1] to [7], wherein the wide-area control step involves changing at least one of the paths of the first mobile body and the second mobile body to a path that does not interfere if the path of the first mobile body interferes with the path of the second mobile body. Item [9] The mobile body control method according to any one of [1] to [9], wherein the computer further performs an image generation step of synthesizing images obtained from multiple imaging devices to generate a single wide-area image. Item

[10] The mobile body is equipped with an environmental sensor, and if the mobile body area cannot be recognized in the recognition step, the mobile body switches to an autonomous driving mode using self-position estimation based on the measurement results of the environmental sensor to control the mobile body, according to any one of [1] to [9]. Item

[11] The aforementioned work object is a production device, and the mobile body control method according to any one of [1] to

[10] generates an action plan for the mobile body based on production tasks obtained from a production management system. Item

[12] A mobile body used in any of the mobile body control methods described in [1] to

[11] , which is an automated guided vehicle that transports goods within a factory. Item

[13] A mobile body used in any of the mobile body control methods described in [1] to

[11] , which is an automated guided vehicle in a production line that supplies a workpiece before processing to a machine tool that is the work target, or retrieves a processed part from the machine tool. Item

[14] A mobile body used in any of the mobile body control methods described in [1] to

[11] , which is an automated guided vehicle in a production line that supplies parts used for assembly to an assembly line or unloads assembled articles from the assembly line. Item

[15] A mobile body used in any of the mobile body control methods described in [1] to

[11] , wherein the mobile body is equipped with a mold changing device that transports molds and performs mold replacement work to a molding machine. Item

[16] A program that causes a computer to execute one of the mobile object control methods described in [1] to

[11] . Item

[17] A mobile body control system for controlling a mobile body having a drive device, comprising: a map generation unit that generates a wide-area map in which a mobile body can move and a target area indicating the vicinity of a work target by the mobile body are defined in a wide-area image including the area in which the mobile body moves; a recognition unit that recognizes a mobile body area representing the position and orientation of the mobile body in the wide-area map based on the wide-area image including the mobile body; a wide-area control unit that defines a path to the target area based on the target area and the mobile body area and generates a first control signal for guiding the mobile body along the path; and a transmission unit that transmits the generated control signal to the mobile body, wherein the wide-area control unit switches to proximity control for guiding the mobile body, which has been guided to the target area, to the work target. Item

[18] A mold exchange system that autonomously performs mold exchange work, comprising: a mobile body equipped with a mold exchange device; and a control device for controlling the mobile body, wherein the control device comprises: a map generation unit that generates a wide-area map in which a mobile body can move and a target area indicating the vicinity of a molding machine are defined in a wide-area image including the area in which the mobile body moves; a recognition unit that recognizes a mobile body area representing the position and orientation of the mobile body in the wide-area map based on the wide-area image including the mobile body; a wide-area control unit that defines a path to the target area based on the target area and the mobile body area and generates a first control signal for guiding the mobile body along the path; and a transmission unit that transmits the generated control signal to the mobile body, wherein the mobile body is guided to the target area based on the first control signal, and transports a mold to the molding machine and performs mold exchange work based on a second control signal for controlling the mobile body in the target area. [Explanation of Symbols]

[0101] 1. Mobile Control System 2 Control device 21 Image generation unit 22 Map generation unit 23 Recognition part 24 Wide-area control unit 25 Proximity Control Unit 26 Transmitter 201 Processor 202 memory 203 Communication Interface 3 Mobile Units 301 Processor 302 memory 303 Communication Interface 304 Drive unit 305 Positioning Sensor 306 Environmental Sensors 307 Height adjustment mechanism 4. Work Target 5. Imaging device 6. Communication equipment

Claims

1. A method for controlling a mobile body having a drive device, Computers A map generation step that generates a wide-area map in which a moving object moves, in a wide-area image that includes an area where the moving object moves, a moving area that indicates the area or path the moving object can move, and a target area that indicates the area around the object to be worked on by the moving object. A recognition step of recognizing a mobile body region that represents the location and orientation of the mobile body in the wide-area map based on the wide-area image including the mobile body, A wide-area control step includes defining a path for the moving object to the target area based on the target area and the moving object area within the coordinate system of the wide-area map, and generating a first control signal for guiding the moving object along the path, The process involves transmitting the generated control signal to a mobile object, and then executing the transmission process. The movable body includes a joint, The work object comprises the joint and the part to be joined, The target area of ​​the wide-area map is defined with proximity guidance information including opposing positions where the joint and the joined portion face each other. The wide-area control step refers to the opposing position of the proximity guidance information defined in the target area, generates a first control signal to guide the moving body to the opposing position in the target area, and switches to a proximity control step to guide the moving body guided to the opposing position to the joined portion of the work object. A method for controlling a moving body, comprising generating a second control signal in the proximity control step to move the moving body from the opposing position toward the joining portion and the joined portion toward the joining portion.

2. The mobile body control method according to claim 1, wherein in the proximity control step, a second control signal is generated for positioning the mobile body relative to the work object based on the measurement result of a positioning sensor mounted on the mobile body.

3. The multiple target areas of the wide-area map are defined by height adjustment values ​​for the joint and the joined portion, respectively. The method for controlling a moving body according to claim 1, wherein in the proximity control step, a second control signal is generated that corrects the inclination of the moving body based on the height adjustment value so that the joint and the joined portion become substantially horizontal.

4. The recognition step recognizes, based on the wide-area image which includes static objects or dynamic objects which include at least a person, a static object region or a dynamic object region which includes a dynamic object in the wide-area map. The method for controlling a moving object according to claim 1 or 2, wherein the wide-area control step defines the path to the target area so as to avoid the object area based on the static object area or the dynamic object area, or generates a third control signal to decelerate or stop the moving object based on the distance between the dynamic object area and the object area.

5. The mobile body is equipped with environmental sensors that monitor the surroundings of the mobile body. A third control signal is generated to decelerate or stop the moving object based on the distance between the moving object and the object measured by the environmental sensor. The mobile body control method according to claim 4, wherein the third control signal takes precedence over at least the first control signal.

6. The mobile body control method according to claim 1 or 2, wherein the wide-area control step involves changing at least one of the paths of the first mobile body and the second mobile body to a path that does not interfere with the path of the second mobile body when the path of the first mobile body interferes with the path of the second mobile body.

7. The mobile body control method according to claim 1 or 2, wherein the computer further performs an image generation step of synthesizing images obtained from multiple imaging devices to generate a single wide-area image.

8. The mobile body is equipped with an environmental sensor, If the mobile body area cannot be recognized in the recognition step, the mobile body switches to an autonomous driving mode using self-position estimation based on the measurement results of the environmental sensor and controls the mobile body, as described in claim 1 or claim 2.

9. The object of the work is a production device, A method for controlling a mobile body according to claim 1, wherein the mobile body generates an action plan based on production tasks obtained from a production management system.

10. A mobile body used in the mobile body control method according to claim 1 or claim 9, An automated guided vehicle (AGV) used to transport goods within a factory.

11. A mobile body used in the mobile body control method according to claim 1 or claim 9, A mobile automated guided vehicle (AGV) in a production line that supplies workpieces before processing to the machine tool that is the target of the work, or retrieves processed parts from the machine tool.

12. A mobile body used in the mobile body control method according to claim 1 or claim 9, A mobile automated guided vehicle (AGV) in a production line that supplies parts used for assembly to an assembly line, or unloads assembled goods from the assembly line.

13. A mobile body used in the mobile body control method according to claim 1 or claim 9, A mobile unit equipped with a mold changer that transports molds to a molding machine and performs mold replacement work.

14. A program that causes a computer to execute the mobile object control method described in claim 1.

15. A mobile body control system for controlling a mobile body having a drive device, A map generation unit generates a wide-area map in which, in a wide-area image including the region in which a moving object moves, the moving region in which the moving object can move and the target region indicating the periphery of the object to be worked on by the moving object are defined. A recognition unit recognizes a mobile body region that represents the location and orientation of the mobile body in the wide-area map based on the wide-area image including the mobile body, A wide-area control unit that defines a path to the target area based on the target area and the moving body area, and generates a first control signal for guiding the moving body along the path, It comprises a transmitting unit that transmits the generated control signal to a mobile object, The movable body includes a joint, The work object comprises the joint and the part to be joined, The target area of ​​the wide-area map is defined with proximity guidance information including opposing positions where the joint and the joined portion face each other. The wide-area control unit refers to the opposing position of the proximity guidance information defined in the target area, generates the first control signal for guiding the moving body to the opposing position in the target area, and switches to proximity control for guiding the moving body, which has been guided to the opposing position, to the joined portion of the work target. A mobile body control system that generates a second control signal in the proximity control described above, which causes the mobile body to move from the opposing position toward the joining portion and the joined portion so as to join them.

16. A mold change system that autonomously performs mold change operations, It comprises a mobile body equipped with a mold changing device and a control device for controlling the mobile body, The control device is A map generation unit generates a wide-area map in which a moving object moves within a wide-area image that includes the region in which the moving object moves, and a target region that indicates the periphery of the molding machine. A recognition unit recognizes a mobile body region that represents the location and orientation of the mobile body in the wide-area map based on the wide-area image including the mobile body, A wide-area control unit that defines a path to the target area based on the target area and the moving body area, and generates a first control signal for guiding the moving body along the path, It comprises a transmitting unit that transmits the generated control signal to the mobile body, The movable body includes a joint, The molding machine comprises the joining portion and the portion to be joined, The target area of ​​the wide-area map is defined with proximity guidance information including opposing positions where the joint and the joined portion face each other. The wide-area control unit refers to the opposing position of the proximity guidance information defined in the target area, generates the first control signal for guiding the moving body to the opposing position in the target area, and switches to proximity control for guiding the moving body, which has been guided to the opposing position, to the part to be joined in the molding machine. In the proximity control described above, a second control signal is generated to move the moving body from the opposing position toward the joining portion and the joined portion so as to join them together. The aforementioned moving body is Based on the first control signal, it is guided to the opposing position in the target region, A mold exchange system that, based on a second control signal for controlling the moving body at the opposing position in the target area, moves the joint and the joined portion toward each other to join them, transports the mold to the molding machine, and performs the mold exchange operation.

Images