Autonomous system and control method thereof

The autonomous body system addresses the challenge of interface and synchronization requirements in robot systems by allowing autonomous bodies to operate independently, ensuring robustness and adaptability even with changing specifications.

JP7595401B2Active Publication Date: 2024-12-06HITACHI LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
JP2018223859
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2018-11-29
Publication Date
2024-12-06
Estimated Expiration
2038-11-29

AI Technical Summary

Technical Problem

Existing robot systems require explicit interfaces and synchronized communication between robots and controllers, necessitating changes when robot specifications change or new robots are added, and simulation devices need prior knowledge of robot specifications for operation program simulation.

Method used

An autonomous body system with sensor units, situation grasping, operation determining, and execution units allows autonomous bodies to operate without explicit interfaces to higher bodies, enabling adaptive operation even with changing specifications and facilitating connection between bodies of different models.

Benefits of technology

Facilitates robust operation by eliminating the need for explicit interfaces and synchronization, allowing seamless integration of new or modified autonomous bodies, and enabling continuous system updates without prior information integration.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007595401000001
    Figure 0007595401000001
  • Figure 0007595401000002
    Figure 0007595401000002
  • Figure 0007595401000003
    Figure 0007595401000003
Patent Text Reader

Abstract

To determine, in an autonomous body system including a plurality of autonomous bodies, operation for each autonomous body, which is desirable for the entirety and also desirable for each autonomous body.SOLUTION: An autonomous body system has a plurality of autonomous bodies. Each of the plurality of autonomous bodies has: a status grasping unit that grasps a status; an operation determination unit that determines operation on the basis of the grasped status; and an operation execution unit that executes the determined operation. The plurality of autonomous bodies of the autonomous body system include one or more first autonomous bodies and two or more second autonomous bodies. A status grasped by the status grasping unit of the first autonomous body includes a status of the second autonomous body. A status grasped by the status grasping unit of the second autonomous body includes a result of operation executed by the operation execution unit of the first autonomous body. The operation determination unit of the second autonomous body determines operation on the basis of the result of the operation executed by the first autonomous execution unit of the first autonomous body.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to an autonomous body system including a plurality of autonomous bodies. [Background technology]

[0002] Techniques for controlling a system including a plurality of robots include those disclosed in, for example, Japanese Patent Application Laid-Open No. 2006-000954 (Patent Document 1) or Japanese Patent Application Laid-Open No. 2015-208811 (Patent Document 2).

[0003] Patent document 1 states regarding a multi-robot system in which multiple robots work cooperatively to achieve a specific purpose as a whole, "The multi-robot system is composed of multiple robots that operate on a stage, a monitor system that monitors the movements of the robots on stage, and an overall controller that operates the multi-robots. Based on the robot's status and the monitored position and direction, the overall controller issues operation instructions to the robots, instructions to correct the robot's position and direction, and operation instructions to the robots in the event of an abnormality occurring (or predicted) in real time via wireless LAN."

[0004] Patent Document 2 states, with regard to optimizing the operation program of a robot system having multiple robots, that "the simulation device comprises a simulation unit that executes, by simulation, an operation program including a command speed and command acceleration of the drive axis and an operation wait command, a memory unit that associates line numbers of the operation program with the command speed and command acceleration at the line numbers and stores them in chronological order, an execution time calculation unit that calculates an execution time of the operation program for each line number stored in the memory unit based on the results of the simulation executed by the simulation unit, and a wait time calculation unit that calculates an operation wait time corresponding to the operation wait command based on the execution time calculated by the execution time calculation unit." [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2006-000954 A [Patent Document 2] JP 2015-208811 A Summary of the Invention [Problem to be solved by the invention]

[0006] In the robot system described in the above Patent Document 1, each robot needs to have an interface suitable for communication with the controller in order to receive instructions from the controller. Also, in order for the controller to control each robot, it is necessary to integrate information as necessary, guarantee consistency and guarantee synchronization. Also, when the specifications of a robot are changed or a robot with new specifications is added, it may be necessary to change the information or interface on the controller accordingly.

[0007] Furthermore, in the simulation device described in Patent Document 2 above, in order to simulate the operation program of the robot system, it is necessary to know the specifications of each robot in advance, and if the specifications are changed or a robot with different specifications is added, the information in the simulation device needs to be updated accordingly. [Means for solving the problem]

[0008] In order to solve at least one of the above problems, a representative example of the invention disclosed in the present application is an autonomous body system having a plurality of autonomous bodies, each of which is a non-human autonomous body having a sensor unit, a situation grasping unit that grasps a situation based on a sensing result by the sensor unit, an action decision unit that decides an action based on the grasped situation, and an action execution unit that executes the decided action, the plurality of autonomous bodies of the autonomous body system include one or more first autonomous bodies and two or more second autonomous bodies, the situation grasped by the situation grasping unit of the first autonomous body includes a situation of the second autonomous body that is grasped based on a sensing result by the sensor unit of the first autonomous body, the situation grasped by the situation grasping unit of the second autonomous body includes a result of an action executed by the action execution unit of the first autonomous body, and the action decision unit of the second autonomous body determines a result of the action executed by the action execution unit of the first autonomous body based on the result of the action executed by the action execution unit of the first autonomous body. follow Determine the action and determining an action that is contrary to the result of the action performed by the action execution unit of the first autonomous body. The operation decision unit of the first autonomous body is characterized in that it determines the operation of the first autonomous body so that an evaluation parameter calculated based on the situation grasped by the situation grasping unit of the first autonomous body becomes a value indicating a high evaluation. Effect of the Invention

[0009] According to one aspect of the present invention, each lower autonomous body (second autonomous body) does not need to explicitly have an interface with the upper autonomous body (first autonomous body). This makes it easy to connect the lower autonomous body and the upper autonomous body regardless of the model and manufacturer of the autonomous body. In addition, since it is no longer necessary to integrate information about each autonomous body in advance or to guarantee their consistency and synchronization, the robustness of the operation is increased. Furthermore, even if the specifications of the lower autonomous body change or a lower autonomous body with new specifications is added, the upper autonomous body can adapt without performing any special work, so that the autonomous body system can be continuously updated.

[0010] Other objects, configurations and advantages than those mentioned above will become apparent from the following description of the embodiments. [Brief description of the drawings]

[0011] [Figure 1] FIG. 1 is an explanatory diagram illustrating an overall configuration of an autonomous body system according to a first embodiment of the present invention. [Diagram 2] FIG. 1 is a block diagram showing a configuration of a control system in an autonomous body system according to a first embodiment of the present invention. [Diagram 3] FIG. 2 is a block diagram showing the configuration of a wing vehicle in an autonomous body system according to a first embodiment of the present invention. [Figure 4] FIG. 2 is a block diagram showing a configuration of a forklift in the autonomous body system according to the first embodiment of the present invention. [Diagram 5] 4 is a flowchart showing a process executed by a control system in the autonomous body system according to the first embodiment of the present invention. [Figure 6] 1 is a flowchart showing the processing performed by a wing vehicle in the autonomous body system according to the first embodiment of the present invention. [Figure 7] 1 is a flowchart showing a process executed by a forklift in the autonomous body system according to the first embodiment of the present invention. [Figure 8] FIG. 11 is an explanatory diagram illustrating an overall configuration of an autonomous body system according to a second embodiment of the present invention. [Figure 9] FIG. 11 is an explanatory diagram illustrating an overall configuration of an autonomous body system according to a third embodiment of the present invention. [Figure 10] FIG. 11 is an explanatory diagram showing the overall configuration of an autonomous body system according to a fourth embodiment of the present invention. [Figure 11] FIG. 11 is an explanatory diagram illustrating the overall configuration of an autonomous body system according to a fifth embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS EXAMPLES

[0012] FIG. 1 is an explanatory diagram showing the overall configuration of an autonomous body system 100 according to a first embodiment of the present invention.

[0013] The autonomous body system 100 of the first embodiment is an automatic pallet unloading system from a wing vehicle by an autonomous forklift in a logistics warehouse. The autonomous body system 100 includes one or more wing vehicles 102, one or more forklifts 103, and a control system 101 as autonomous bodies.

[0014] Here, the term "autonomous body" refers to a general term for an object that grasps the situation, makes a decision, and operates by itself. For example, the wing car 102 and the forklift 103 that operate autonomously based on sensor information, etc., and the control system 101 that makes an autonomous decision based on sensor information, etc. and outputs the result are autonomous bodies. In the first embodiment, such an autonomous body will be mainly described. However, even the wing car 102 and the forklift 103 operated by a human can be treated as an autonomous body that grasps the situation by the human and operates according to the result of the decision. Furthermore, for example, when there is a human performing work such as transporting and sorting goods in a warehouse, the human himself can also be treated as an autonomous body that grasps the situation by himself based on the results of sensing by vision, hearing, etc., and acts autonomously by making a decision using the human intelligence.

[0015] The warehouse includes a receiving area 110, an inventory area 111, and a shipping area 112. When a wing vehicle 102 arrives at the receiving area 110, a forklift 103 unloads a pallet 104 loaded with goods and carries it to the inventory area 111. The goods transported to the inventory area 111 may be stored while still loaded on the pallet 104, or may be removed from the pallet 104 and stored on a shelf. The shelf on which the goods are stored may be fixed to the inventory area 111, or may be movable by an automatic guided vehicle (not shown) or the like.

[0016] The forklift 103 carries the necessary pallets 104 according to the shipping order from the inventory area 111 to the shipping area 112, and loads them onto the wing vehicle 102 that has arrived at the shipping area 112. After loading the pallets 104, the wing vehicle 102 leaves the shipping area 112 and transports the goods to the next delivery destination.

[0017] Although omitted in Figure 1 for simplicity's sake, an actual warehouse generally includes an area for temporarily storing pallets 104 unloaded from wing vehicles 102 (or pallets that will soon be loaded onto wing vehicles 102), an area for unpacking items loaded onto pallets 104, an area for picking, sorting, inspecting, etc. items according to shipping orders, and an area for loading items to be shipped onto pallets.

[0018] The control system 101 is a higher-level (or first type) autonomous entity that grasps the status of each wing vehicle 102 and each forklift 103 in the warehouse and outputs information for controlling them. This acquires information about the entire autonomous system and performs processing to optimize the entire system, which may be called holistic intelligence. In contrast, each wing vehicle 102 and each forklift 103 is a lower-level (or second type) autonomous entity that autonomously determines its own operation while utilizing information output from the control system 101 as necessary. Each acquires information about its surroundings and operates autonomously, which may be called an autonomous individual.

[0019] 1, one or more cameras 105 capable of communicating with a control system 101 and one or more traffic lights 106 capable of communicating with the control system 101 are provided in a warehouse. Each camera 105 takes pictures of the inside of the warehouse and transmits data of the captured images to the control system 101. Each traffic light 106 is provided, for example, in a receiving area 110 or a shipping area, and outputs some kind of signal in accordance with information transmitted from the control system 101. Furthermore, the control system 101 may communicate with each forklift 103.

[0020] The control system 101 may be installed either inside the warehouse or outside the warehouse (including a remote location) as long as it is capable of communicating with the camera 105, the traffic light 106, the forklift 103, and the like.

[0021] FIG. 2 is a block diagram showing a configuration of the control system 101 in the autonomous body system 100 according to the first embodiment of the present invention.

[0022] The control system 101 is a computer having a processor 201 , a communication interface (I / F) 202 , and a storage device 203 .

[0023] The processor 201 executes various processes according to programs stored in the storage device 203. The communication interface 202 is connected to a network 210, and communicates with the camera 105, the traffic light 106, the forklift 103, etc. (and also with the wing vehicle 102 if the wing vehicle 102 has a communication function with the control system 101) via the network 210.

[0024] The storage device 203 may be, for example, a combination of a main storage device such as a dynamic random access memory (DRAM) and an auxiliary storage device such as a hard disk drive or a flash memory. The storage device 203 holds a situation grasping unit 204, an action determining unit 205, an action executing unit 206, and task information 208.

[0025] The situation grasping unit 204, the action determining unit 205, and the action executing unit 206 are programs executed by the processor 201. In the following description, the processes executed by the above-mentioned units are actually executed by the processor 201 by controlling the communication interface 202, the camera 105, the traffic light 106, etc. as necessary in accordance with the programs stored in the storage device 203.

[0026] For example, the photographing function of the camera 105 and the function of the processor 201 to grasp the situation from the image photographed by the camera 105 in accordance with the program of the situation grasping unit 204 may be treated as a whole as a function of the situation grasping unit 204. In other words, the camera 105 may be considered as a part of the situation grasping unit 204. Similarly, the traffic light 106 may be considered as a part of the action execution unit 206. The processing of each of the above units will be described later.

[0027] The task information 208 is information about a task to be executed by the control system 101. For example, when the control system 101 of the first embodiment controls the wing vehicle 102 and the forklift 103 in a warehouse, the task information 208 may include a shipping order including the type, quantity, and shipping destination of the goods to be shipped from the warehouse.

[0028] The camera 105 is an optical camera that takes pictures inside the warehouse. If necessary to grasp the situation of the warehouse, a camera 105 that takes pictures outside the warehouse may be installed. The camera 105 takes pictures inside the warehouse and transmits the image data to the control system 101 via the network 210. The situation grasping unit 204 grasps the situation of the warehouse (for example, the arrangement of the pallets 104 in each area in the warehouse, the position and state of each wing car 102 and each forklift 103, etc.) based on the image data.

[0029] The camera 105 is an example of a sensor that acquires information for the control system 101 to grasp the situation of the warehouse. Instead of (or in addition to) the camera 105, a radar or a LIDAR (Laser Imaging Detection and Ranging) may be used as a sensor. Also, for example, a beacon may be used so that the control system 101 grasps the positions of each wing vehicle 102 and each forklift 103. In that case, a receiver for a beacon signal is installed at each location in the warehouse and connected to the network 210. A plurality of cameras 105, or a plurality of types of sensors may be installed in the warehouse.

[0030] The traffic light 106 may output a color signal, for example, green to allow proceeding, or yellow or red to instruct stopping, similar to a traffic light. The traffic light 106 is installed, for example, in the receiving area 110 and the shipping area 112, and is used to present information such as permission to proceed or stop to the wing vehicle 102. The traffic light 106 may output any color signal according to a signal transmitted from the control system 101 via the network 210. Alternatively, the traffic light 106 may display information such as proceeding or stopping by a signal including a figure, symbol, or character instead of a color signal. Furthermore, the traffic light 106 may display information instructing opening and closing of the wings of the wing vehicle 102. Alternatively, instead of (or in addition to) the traffic light 106, a device that outputs an audio signal including the above-mentioned information may be used.

[0031] Network 210 may be of any type so long as it enables communication between control system 101 and wing vehicle 102, forklift 103, camera 105, and traffic light 106. For example, network 210 may include a wide area network if control system 101 is in a remote location, or may include a wireless network for communication with moving forklift 103, etc.

[0032] FIG. 3 is a block diagram showing the configuration of the wing vehicle 102 in the autonomous body system 100 according to the first embodiment of the present invention.

[0033] The wing vehicle 102 is a vehicle that transports a pallet 104 loaded with goods. Here, a wing vehicle (i.e., a truck vehicle with a lift-up side panel and roof of the loading platform) that allows easy loading and unloading of pallets by a forklift 103 is shown as an example, but other types of vehicles may also be used.

[0034] The wing vehicle 102 has a processor 301 , a communication interface 302 , a sensor unit 303 , a drive unit 304 and a memory unit 305 .

[0035] The processor 301 executes various processes according to programs stored in the storage device 305. The communication interface 302 is connected to the network 210 and communicates with the control system 101. However, the wing vehicle 102 does not need to have a communication function with the control system 101, in which case the wing vehicle 102 does not need to have the communication interface 302.

[0036] The sensor unit 303 acquires information for grasping the surrounding situation of the wing vehicle 102. The sensor unit 303 may include, for example, an optical camera, a radar, or a LIDAR, or may include an audio sensor such as a microphone. The wing vehicle 102 may have multiple sensor units 303, or may have multiple types of sensor units 303.

[0037] The driving unit 304 has functions such as driving the wing vehicle 102 and opening and closing the wings. The driving unit 304 includes, for example, a prime mover such as an internal combustion engine or an electric motor, a transmission device that transmits the power generated by the prime mover to the wheels or wings, and a steering device that drives the wing vehicle 102 to drive in a desired direction.

[0038] The storage device 305 may be, for example, a combination of a main storage device such as a DRAM and an auxiliary storage device such as a hard disk drive or a flash memory. The storage device 305 holds a situation grasping unit 306, an action determining unit 307, an action executing unit 308, and task information 310.

[0039] The situation grasping unit 306, the action determining unit 307, and the action executing unit 308 are programs executed by the processor 301. In the following description, the processes executed by the above-mentioned units are actually executed by the processor 301 by controlling the communication interface 302, the sensor unit 303, the driving unit 304, etc. as necessary in accordance with the programs stored in the storage device 305.

[0040] For example, the sensing function of the sensor unit 303 and the function of the processor 301 grasping the situation according to the program of the situation grasping unit 306 based on the information sensed by the sensor unit 303 may be treated as a whole as a function of the situation grasping unit 306. In other words, the sensor unit 303 may be considered as a part of the situation grasping unit 306. When information received by the communication interface 302 is used for grasping the situation, the communication interface 302 may also be considered as a part of the situation grasping unit 306. Similarly, the drive unit 304 may be considered as a part of the operation execution unit 308.

[0041] The wing vehicle 102 of this embodiment has an automatic driving function for performing automatic driving without a driver, or a driving assistance function for assisting the driver in driving. At least a part of the above-mentioned situation grasping unit 306, action decision unit 307, and action execution unit 308 may be implemented as at least a part of the automatic driving function or the driving assistance function.

[0042] Although not shown in FIG. 3, the storage device 305 may further store map data and the like referenced by the automatic driving function or the driving assistance function.

[0043] The task information 310 is information about a task to be performed by the wing vehicle 102. For example, the task information 310 may include information about a warehouse or shipping destination to which the wing vehicle 102 should next head.

[0044] FIG. 4 is a block diagram showing a configuration of the forklift 103 in the autonomous body system 100 according to the first embodiment of the present invention.

[0045] The forklift 103 is a vehicle that performs tasks such as unloading the pallets 104 from the wing vehicle 102, transporting the pallets 104 between the receiving area 110, the inventory area 111, and the shipping area 112, and loading the pallets 104 onto the wing vehicle 102.

[0046] The forklift 103 includes a processor 401 , a communication interface 402 , a sensor unit 403 , a drive unit 404 , and a memory device 405 .

[0047] The processor 401 executes various processes in accordance with programs stored in the storage device 405. The communication interface 402 is connected to the network 210 and communicates with the control system 101.

[0048] The sensor unit 403 acquires information for grasping the situation around the forklift 103. The sensor unit 403 may include, for example, an optical camera, a radar, or a LIDAR, or may include an audio sensor such as a microphone. The forklift 103 may have a plurality of sensor units 403, or may have a plurality of types of sensor units 403. Specifically, the sensor unit 403 may include a sensor for detecting surrounding objects for the travel of the forklift 103, and a sensor for measuring the positional relationship between the prongs of the fork and the pallet 104.

[0049] The driving unit 404 has a function of driving the forklift 103 and lifting and lowering the pallet 104 with the forks. The driving unit 404 includes, for example, a prime mover such as an internal combustion engine or an electric motor, a transmission device that transmits the power generated by the prime mover to the wheels or the forks, a steering device that drives the forklift 103 to travel in a desired direction, and the like.

[0050] The storage device 405 may be, for example, a combination of a main storage device such as a DRAM and an auxiliary storage device such as a hard disk drive or a flash memory. The storage device 405 holds a situation grasping unit 406, an action determining unit 407, an action executing unit 408, and task information 410.

[0051] The situation grasping unit 406, the operation determining unit 407, and the operation executing unit 408 are programs executed by the processor 401. In the following description, the processes executed by the above units are actually executed by the processor 401 by controlling the communication interface 402, the sensor unit 403, the driving unit 404, etc. as necessary in accordance with the programs stored in the storage device 405.

[0052] For example, the sensing function of the sensor unit 403 and the function of the processor 401 grasping the situation according to the program of the situation grasping unit 406 based on the information sensed by the sensor unit 403 may be treated as a whole as a function of the situation grasping unit 406. In other words, the sensor unit 403 may be considered as a part of the situation grasping unit 406. When information received by the communication interface 402 is used for grasping the situation, the communication interface 402 may also be considered as a part of the situation grasping unit 406. Similarly, the drive unit 404 may be considered as a part of the operation execution unit 408. The processing of each of the above units will be described later.

[0053] The forklift 103 of this embodiment has an automatic driving function for automatic driving without a driver, or a driving assistance function for assisting the driver in driving. At least a part of the situation grasping unit 406, the operation decision unit 407, and the operation execution unit 408 may be implemented as at least a part of the automatic driving function or the driving assistance function.

[0054] Although not shown in FIG. 4, the storage device 405 may further store map data of the warehouse to be referenced by the automatic driving function or the driving assistance function.

[0055] The task information 410 is information related to a task to be performed by the forklift 103. For example, the task information 410 may include information indicating the pallet 104 to be transported by the forklift 103 and its destination.

[0056] FIG. 5 is a flowchart showing a process executed by the control system 101 in the autonomous body system 100 according to the first embodiment of the present invention.

[0057] First, the situation grasping unit 204 of the control system 101 grasps the situation of the warehouse (step 501). For example, the situation grasping unit 204 may perform image recognition or the like on image data transmitted from each camera 105 to grasp the arrangement of the pallets 104 in each area of ​​the warehouse, the position and state of each wing vehicle 102 (e.g., whether or not it is traveling, where it is parked if parked, whether the wings are open, etc.), and the position and state of each forklift 103 (e.g., whether or not it is traveling, whether or not it is loaded with a pallet 104, etc.).

[0058] Next, the operation determination unit 205 of the control system 101 determines the operation of the control system 101 (step 502). Specifically, when the situation grasped by the situation grasping unit 204 is input, the operation determination unit 205 determines the operation of the control system 101 based on the input. Here, the operation of the control system 101 is, for example, the transmission of a signal to each traffic light 106 and the transmission of an instruction to each forklift 103. For example, the operation determination unit 205 may grasp or predict the arrival of the wing vehicle 102 at the receiving area or the shipping area and the opening and closing of the wings based on the situation grasped by the situation grasping unit 204, and based on that, determine the timing when the forklift 103 goes to pick up (or loads) the pallet 104. Alternatively, the operation determination unit 205 may grasp the situation when the unloading of goods from the wing vehicle 102 or the loading of goods into the wing vehicle 102 is completed and the forklift 103 leaves the loading platform, and determine the timing when the wing vehicle 102 closes the wings and departs.

[0059] Specifically, for example, the operation determination unit 205 may hold a function that receives the situation grasped by the situation grasping unit 204 as an input and outputs an operation to be executed, and may use the function to execute step 502. This function may be any type, and may be implemented, for example, by a neural network.

[0060] Next, the operation execution unit 206 of the control system 101 executes the determined operation (step 503). For example, when a signal to be displayed by each traffic light 106 (e.g., a signal indicating the progress or stop of the wing vehicle 102, opening or closing of the wing, etc.) is determined in step 502, the operation execution unit 206 transmits an instruction to each traffic light 106 to display the determined signal via the communication interface 202 and the network 210. Also, when information to be transmitted to each forklift 103 (e.g., the destination of each forklift 103, identification information of a pallet to be transported, etc.) is determined in step 502, the operation execution unit 206 transmits the determined information to each forklift 103 via the communication interface 202 and the network 210.

[0061] Thereafter, the process returns to step 501, and the processes from step 501 to 503 are repeatedly executed.

[0062] The operation determination unit 205 may perform a learning process to optimize the operation determination at any timing (for example, when it is determined that a sufficient amount of data that can be used for learning has been accumulated). For example, the operation determination unit 205 may hold a KPI (Key Performance Indicator) set in the control system 101, and change the parameters of a function so that the value is improved (i.e., the value indicates a high evaluation).

[0063] Here, an example of the learning process executed by the operation determination unit 205 will be described. The KPI (i.e., the parameter for evaluating the performance of the control system 101) set in the control system 101 is, for example, the throughput or sales amount of the entire warehouse. For example, the operation determination unit 205 may change the parameter of the function used in step 502 so as to improve the KPI (for example, improve the throughput or sales amount) based on the KPI after the operation determined in step 502 is executed. By using the function learned in this way, an operation that improves the KPI is determined in step 502.

[0064] FIG. 6 is a flowchart showing the processing executed by the wing vehicle 102 in the autonomous body system 100 according to the first embodiment of the present invention.

[0065] First, the situation grasping unit 306 of the wing vehicle 102 grasps the situation around the wing vehicle 102 (step 601). For example, based on information acquired by the sensor unit 303, the situation grasping unit 306 may identify objects around the wing vehicle 102, the situation of the receiving area 110 or the shipping area 112 where the wing vehicle 102 unloads or loads, and the display content of the traffic light 106 visible from the wing vehicle 102, etc.

[0066] Next, the operation determination unit 307 of the wing vehicle 102 determines the operation of the wing vehicle 102 (step 602). Specifically, when the situation grasped by the situation grasping unit 306 is input, the operation determination unit 307 determines the operation of the wing vehicle 102 based on the input. Here, the operation of the wing vehicle 102 includes, for example, driving, stopping, steering for parking in the receiving area 110 or the shipping area 112 and leaving these areas, opening and closing of wings, etc.

[0067] Specifically, for example, the action determination unit 307 may hold a function that takes the situation grasped by the situation grasping unit 306 as an input and outputs an action to be executed, and may use the function to execute step 602. This function may be any type, and may be implemented, for example, by a neural network.

[0068] Next, the action execution unit 308 of the wing vehicle 102 executes the determined action (step 603). For example, if it is determined in step 602 that the wing vehicle 102 will run, stop, steer, or open / close the wings, the action execution unit 308 transmits an instruction to the drive unit 304 to execute the determined action.

[0069] Thereafter, the process returns to step 601, and the above steps 601 to 603 are repeatedly executed.

[0070] The motion determination unit 307 may perform a learning process to optimize the motion determination at any timing. For example, the motion determination unit 307 may hold a KPI set for the wing vehicle 102 and change the parameters of a function so that the value is improved (i.e., the value indicates a high evaluation).

[0071] Here, an example of the learning process executed by the operation determination unit 307 will be described. The KPI (i.e., the parameter for evaluating the performance of the wing vehicle 102) set for the wing vehicle 102 is set so that the shorter the stay time of the wing vehicle 102 in the warehouse, the higher the evaluation. For example, the operation determination unit 307 may change the parameter of the function used in step 602 based on the KPI calculated after the operation determined in step 602 is executed, so as to improve the KPI (e.g., shorten the stay time).

[0072] More specifically, for example, the operation determination unit 307 inputs the surrounding situation grasped by the situation grasping unit 306 based on the information acquired by the sensor unit 303 into a function, and determines an operation based on the output. At this time, the grasped surrounding situation includes the recognition result of the display content of the signal of the traffic light 106 outputted toward the wing vehicle 102, in addition to the recognition result of the objects around the wing vehicle 102. The display content of the traffic light 106 is the operation result of the control system 101, and instructs the operation that the control system 101 judges that the wing vehicle 102 should perform in order to improve the KPI of the control system 101.

[0073] However, simply executing such instructions from the control system 101 does not necessarily improve the KPI of the wing vehicle 102. This is because, for example, due to differences in the type of sensors used and their installation positions, the warehouse situation that the control system 101 can acquire does not necessarily match the surrounding situation that the wing vehicle 102 can acquire, and because the control system 101 does not know the KPI of the wing vehicle 102. The KPI set for the wing vehicle 102 is generally different from that set for the control system 101. Also, there are cases where different KPIs are set for each of the multiple wing vehicles 102.

[0074] For this reason, the operation determination unit 307 may change the parameters used in step 602, for example, by changing the weighting given to the result of recognizing the operation result of the control system 101 among the surrounding conditions and other conditions so as to improve the KPI set in the wing vehicle 102. By using the function learned in this way, an operation that improves the KPI is determined in step 602.

[0075] FIG. 7 is a flowchart showing a process executed by the forklift 103 in the autonomous body system 100 according to the first embodiment of the present invention.

[0076] First, the situation grasping unit 406 of the forklift 103 grasps the situation around the forklift 103 (step 701). For example, based on the information acquired by the sensor unit 403 and the information received by the communication interface 402 from the control system 101, the situation grasping unit 406 may identify objects around the forklift 103, the situation of the loading platform of the wing vehicle, the situation of the pallets on the loading platform, the situation of the receiving area and the shipping area, and the operation result of the control system 101, and the like.

[0077] Next, the operation determination unit 407 of the forklift 103 determines the operation of the forklift 103 (step 702). Specifically, when the situation grasped by the situation grasping unit 406 is input, the operation determination unit 407 determines the operation of the forklift 103 based on the input. Here, the operation of the forklift 103 includes, for example, movement to the position of the pallet, inserting the fork claws into the pallet and lifting it up, movement of the pallet to the destination, and placing the pallet at the destination.

[0078] Specifically, for example, the operation determination unit 407 may hold a function that receives the situation grasped by the situation grasping unit 406 as an input and outputs an operation to be executed, and may use the function to execute step 702. This function may be any type, and may be implemented, for example, by a neural network.

[0079] Next, the operation execution unit 408 of the forklift 103 executes the determined operation (step 703). For example, when it is determined in step 702 that the forklift 103 should run, stop, steer, or raise or lower the fork, the operation execution unit 408 transmits an instruction to the drive unit 404 to execute the determined operation.

[0080] Thereafter, the process returns to step 701, and the processes from step 701 to 703 are repeatedly executed.

[0081] The operation determination unit 407 may perform a learning process to optimize the operation determination at any timing. For example, the operation determination unit 407 may hold a KPI set for the forklift 103 and change a parameter of a function so that the value is improved (i.e., the value indicates a high evaluation).

[0082] Here, an example of the learning process executed by the operation determination unit 407 will be described. Forklift 103The parameters for evaluating the performance of the forklift 103 are set so that the more efficiently or reliably the forklift 103 works, the higher the evaluation. The efficiency of the work may be measured, for example, by the amount of work (the amount of transported goods or the traveled distance, etc.) per unit time. The reliability of the work may be measured, for example, by the number of emergency stops per unit time.

[0083] The operation determination unit 407 may change the parameters of the function used in step 702, for example, based on the KPI calculated after the operation determined in step 702 is executed, so as to improve the KPI (e.g., improve work efficiency).

[0084] More specifically, for example, the operation determination unit 407 inputs the surrounding situation grasped by the situation grasping unit 406 based on the information acquired by the sensor unit 403 and the information received via the communication interface 402 into a function, and determines an operation based on the output. At this time, the grasped surrounding situation includes information transmitted from the control system 101 to the forklift 103, in addition to the recognition result of objects around the forklift 103. The information transmitted from the control system is the operation result of the control system 101, and for example, instructs the operation that the control system 101 judges that the forklift 103 should perform in order to improve the KPI of the control system 101.

[0085] However, simply following instructions from the control system 101 does not necessarily improve the KPI of the forklift 103. This is because, for example, due to differences in the type of sensor used and the installation position, the warehouse situation that the control system 101 can obtain does not necessarily match the surrounding situation that the forklift 103 can obtain, and the control system 101 does not know the KPI of the forklift 103. The KPI set for the forklift 103 is generally different from that set for the control system 101. Also, there are cases where different KPIs are set for each of the multiple forklifts 103.

[0086] For this reason, the operation determination unit 307 may change the parameters used in step 702 by, for example, changing the weighting given to the result of recognizing the operation result of the control system 101 among the surrounding conditions and other conditions so as to improve the KPI set for the forklift 103. By using the function learned in this way, in step 702, an operation that improves the KPI is determined.

[0087] By having the control system 101, the wing vehicle 102, and the forklift 103 each learn the functions in the manners shown in Figures 5 to 7, they can determine their respective operations so as to improve their respective KPIs, even if they do not know each other's KPIs.

[0088] For example, the control system 101 provides information to each wing vehicle 102 and each forklift 103 so as to improve KPI (e.g., warehouse throughput) based on the state of the warehouse ascertained from images captured by the camera 105. For example, when the control system 101 determines that the wing vehicle 102 has arrived at the receiving area 110, it may transmit information to each forklift 103 instructing it to go to the receiving area 110 for unloading.

[0089] The forklift 103 that receives this information determines its own operation so as to improve its own set KPI (e.g., work efficiency) based on this information, the surrounding situation grasped through its own sensor unit 403, and the task information 410. As a result, the forklift 103 may move to the receiving area 110 in accordance with the instruction included in the received information, or may not follow the instruction and perform another operation.

[0090] For example, if the forklift 103 is already transporting another pallet 104, it may disregard the instruction and continue transporting the pallet 104. Also, if the current location of the forklift 103 is far from the receiving area 110, it may determine that it is not appropriate to follow the instruction in order to improve the KPI set for itself, and may perform another operation without following the instruction.

[0091] Alternatively, for example, a warehouse may use multiple models of forklifts 103, and the conditions (such as weight) of the pallets 104 that can be transported may differ depending on the model. In such a case, if the forklift 103 is able to obtain information about the goods loaded on the wing vehicle 102 that it has arrived at, when the wing vehicle 102 carrying a pallet that it cannot transport arrives, the forklift 103 can decide not to unload the goods from that wing vehicle 102 even if it receives an instruction from the control system 101.

[0092] Alternatively, for example, when the control system 101 detects congestion of forklifts 103 in a specific area from an image in a warehouse, it may transmit an instruction to move the forklifts 103 involved in the congestion away from the area in order to eliminate the congestion and improve the KPI. The forklifts 103 that receive this instruction may move according to the instruction, or may determine that following the instruction will cause the KPI of the forklifts 103 to deteriorate and decide not to follow the instruction.

[0093] In this way, the instructions from the control system 101 to the forklift 103 are not intended to be compulsory, but rather to gently influence the decision of how the forklift 103 will operate.

[0094] Similarly, the wing vehicle 102 may decide to act in accordance with instructions received from the control system 101 via traffic lights 106, or may decide to act contrary to those instructions in order to improve the KPI set for itself.

[0095] In this way, when the control system 101 determines its own action and executes it (i.e., determines the contents of instructions for each lower-level autonomous entity and transmits them), some autonomous entities may act in accordance with the instructions, while other autonomous entities may act contrary to the instructions. As a result, the state inside the warehouse changes. The control system 101 grasps the changed state inside the warehouse and calculates the value of the KPI. Based on the previously grasped state, the action determined based on that, and the KPI value calculated based on the state after the action is executed, the control system 101 may learn the parameters of a function for determining an action from the grasped state so that the value of the KPI improves.

[0096] For example, when the wing vehicle 102 arrives at the receiving area 110, the initial control system 101 may instruct all forklifts 103 to move to the receiving area 110. However, as a result of learning, it is considered that the control system 101 will be more likely to send a movement instruction to a forklift 103 that is not transporting a pallet 104, a forklift 103 that is near the receiving area 110, or a forklift 103 that matches the conditions of the pallet loaded on the arriving wing vehicle 102.

[0097] In this way, even if the control system 101 does not know the characteristics (including KPIs) of each of the lower autonomous bodies in advance, it can issue instructions that are easy for each of the lower autonomous bodies to accept in light of the KPIs set for itself by gently influencing each autonomous body and learning the results. This makes it possible to determine the operation of each autonomous body so as to contribute to the improvement of the KPIs of the control system 101 (i.e., the entire autonomous body system 100) and also to the improvement of the KPIs of each autonomous body.

[0098] In order to carry out such learning, the control system 101 may transmit instructions that deliberately worsen the KPI values ​​set for itself to a certain extent, or instructions that are expected to be difficult for each autonomous entity to accept.

[0099] On the other hand, as for the lower-level autonomous bodies, for example, the forklift 103 may perform an action according to the surrounding situation acquired from the sensor unit 403 and the instruction from the control system 101 received by the communication interface 402 so as to improve its own KPI, or may perform an action contrary to the instruction. After performing the determined action, the forklift 103 calculates the value of the KPI set for itself based on the situation in which the action is reflected. The forklift 103 may learn the parameters of a function for determining an action from the grasped situation so as to improve the value of the KPI, based on the previously grasped situation, the action determined based on the previously grasped situation, and the value of the KPI calculated based on the situation after the action is executed.

[0100] Here, the parameters of the function may include weights assigned to the situation grasped via the sensor unit 403 and the instructions from the control system 101. In order to perform such learning, the forklift 103 may determine an operation that deliberately worsens the value of the KPI set for itself.

[0101] The wing vehicle 102 is similar to the forklift 103 described above.

[0102] As described above, by each of the upper and lower autonomous bodies learning, even if each autonomous body does not know the characteristics of the other autonomous bodies (in other words, even if the lower autonomous bodies are black boxes from the perspective of the upper autonomous body), it is possible to determine the behavior of each autonomous body so that the KPI of each autonomous body improves and the KPI of the whole system also improves. This eliminates the need to integrate the entire information or guarantee the consistency and synchronization of the whole system, and realizes robust operation. In addition, even if an unknown lower autonomous body is added or the characteristics of an existing lower autonomous body change, the upper autonomous body can adapt to it, allowing the autonomous system to continuously evolve.

[0103] Furthermore, by using an information transmission means such as the traffic light 106, the lower autonomous body does not need to explicitly have an interface with the higher autonomous body. This makes it possible to easily connect autonomous bodies of a type that was not initially anticipated (for example, an autonomous body manufactured by a different manufacturer than the other autonomous bodies).

[0104] In the above example, instructions to the wing vehicle 102 are transmitted via the traffic light 106, and instructions to the forklift 103 are transmitted via the communication interface 402. In this example, the wing vehicle 102 does not need to have the communication interface 302. This makes it possible to transmit information from the control system 101 to the forklift 103 without having to determine detailed communication procedures in advance.

[0105] However, such an information transmission method is merely an example. For example, information may be transmitted to the wing vehicle 102 via the communication interface 302. Alternatively, traffic lights similar to the traffic light 106 may be provided at various locations within the warehouse, and information may be transmitted to the forklift 103 via the traffic lights.

[0106] In the above example, both the wing vehicle 102 and the forklift 103 have an automatic driving function or a driving assistance function, but at least some of them may not have either the automatic driving function or the driving assistance function. For example, any of the wing vehicles 102 may not have the processor 301, the communication interface 302, the sensor unit 303, and the storage device 305. In that case, the driver of the wing vehicle 102 (i.e., a human) operates the drive unit 304 to control the running of the wing vehicle 102 and the opening and closing of the wings. Also, the communication interface 302 and the sensor unit 303 are replaced by the driver's sensory organs (e.g., eyes and ears, etc.). Also, the functions of the situation grasping unit 306, the action determining unit 307, and the action executing unit 308 are replaced by the driver's intelligence.

[0107] Even in this case, the driver can receive instructions from the control system 101 by referring to the traffic lights 106. In this case, the KPI set in the wing vehicle 102 may be, for example, that the driver wants to finish work quickly or that the driver wants to perform work reliably.

[0108] Similarly, the forklift 103 may be operated by a human being. In this case, instructions may be transmitted from the control system 101 via a traffic light, or instructions received via the communication interface 402 may be displayed on a display unit (not shown) of the forklift 103 so that the driver can see and receive the instructions.

[0109] In addition, although the autonomous body system of the first embodiment includes a plurality of wing vehicles 102 and a plurality of forklifts 103 as subordinate autonomous bodies, in reality, it may include only one of them, or may further include another autonomous body. For example, when a robot that picks up items is installed in a warehouse, the robot may be added as a second type of autonomous body. Alternatively, when an automatic guided vehicle is used to transport shelves that store items in the inventory area 111 to a picking area or the like, such an automatic guided vehicle may be added as a second type of autonomous body. In that case, some KPI is set for the second type of autonomous body as with the wing vehicles 102 and the forklifts 103, and the same processing as described above is performed.

[0110] In addition, the multiple wing vehicles 102 may include multiple types of vehicles made by different manufacturers, and different KPIs may be set for each of the multiple wing vehicles 102. The same applies to the multiple forklifts 103.

[0111] According to the above-described first embodiment, the control system 101 provides information determined to be effective for improving the KPI of the control system 101 to the lower autonomous bodies (for example, the wing vehicle 102 and the forklift 103) according to the state of the warehouse grasped by the control system 101 via the camera 105 or the like. This allows, for example, the timing for the forklift 103 to access the loading platform of the wing vehicle 102 to be accurately known in accordance with the actual state of the wing vehicle 102 and its surroundings. This reduces the waiting time of the forklift 103, and shortens the work time for unloading and loading. In addition, the timing for the wing vehicle 102 to depart can be accurately known in accordance with the actual state of unloading and loading of goods. This shortens the stopping time of the wing vehicle 102.

[0112] In an autonomous system in which there is no higher-level autonomous entity such as the control system 101 and lower-level autonomous entities directly exchange information, it is difficult to obtain the above-mentioned effects for the following reasons. That is, in a system without a higher-level autonomous entity, each wing vehicle 102 and each forklift 103 determines an operation for improving the KPI set for each vehicle based on information acquired by the vehicle's own sensor. However, since only information within a limited range can be acquired from each sensor, a discrepancy occurs between the acquired information and the actual warehouse situation, and as a result, the KPI may not improve.

[0113] On the other hand, according to this embodiment, the control system 101 can grasp the situation of the entire warehouse (i.e., the entire autonomous system) and based on that, can notify each autonomous body of the desired operation to improve the overall KPI, thereby improving the overall KPI.

[0114] Also, in order to directly exchange information between the wing vehicle 102 and the forklift 103, it is necessary to determine in advance the means of communication between them. However, in reality, there are cases where the managers are different, for example, the forklift 103 is managed by the owner of a warehouse, whereas the wing vehicle 102 is managed by a transportation company, making it practically difficult to determine the means of communication in advance. However, according to this embodiment, by using an information transmission means such as a traffic light 106, it becomes possible to provide the necessary information without determining the means of communication in advance. EXAMPLES

[0115] Next, a second embodiment of the present invention will be described. Except for the differences described below, each part of the autonomous system of the second embodiment has the same function as each part of the autonomous system of the first embodiment shown in Figs. 1 to 7 to which the same reference numerals are attached, and therefore the description thereof will be omitted.

[0116] FIG. 8 is an explanatory diagram showing the overall configuration of an autonomous body system 800 according to a second embodiment of the present invention.

[0117] The autonomous body system 800 of the second embodiment is a robot system that performs a picking task of taking out an item 805 stored on a shelf 804 and packing it into a sorting box 806. This autonomous body system 800 includes an arm / lifter integrated management system 801, an arm robot 802, and a lifter 803 as autonomous bodies.

[0118] The arm robot 802 is, for example, a vertical articulated robot, which picks up an item 805 stored on a shelf 804 and packs the item 805 into a box 806. For this purpose, the arm robot 802 grasps the situation of the items 805 stacked on the shelf 804 as the surrounding situation. For grasping this situation, a sensor (for example, a camera, etc.) installed in the arm robot 802 is used. The movement of the arm robot 802 is determined by the intelligence of the arm robot 802 (for example, a movement determination unit 307 of the arm robot 802 described later).

[0119] The configuration of the arm robot 802 is similar to that of the wing vehicle 102 of the first embodiment shown in Fig. 3 or the forklift 103 of the first embodiment shown in Fig. 4, except for the following differences. Here, the configuration of the arm robot 802 will be described with reference to Fig. 3.

[0120] The communication interface 302 of the arm robot 802 communicates with the arm / lifter integrated management system 801 via the network 210. The sensor unit 303 of the arm robot 802 is, for example, a camera attached to the arm robot 802. Based on an image captured by this camera in the direction of the shelf 804, the situation grasping unit 306 of the arm robot 802 grasps the situation of the items 805 stacked on the shelf 804. Furthermore, the situation grasping unit 306 of the arm robot 802 grasps a message transmitted from the arm / lifter integrated management system 801 (for example, information indicating that the lifter 803 has been lifted to a desired height, which will be described later).

[0121] Further, the driving unit 304 of the arm robot 802 includes a motor (not shown) that drives a joint 808 of an arm 807 of the arm robot 802 and a hand unit 809 at the tip of the arm 807, and the like.

[0122] The task information 310 includes, for example, information for identifying an item to be picked up by the arm robot 802. The operation determination unit 307 determines the operations of each joint 808 and the hand unit 809 for picking up a desired item 805, based on, for example, information from the arm / lifter integrated management system 801 acquired via the communication interface 302, a state grasped from the result of sensing by the sensor unit 303, and the task information 310. Then, the operation execution unit 308 executes the determined operation via the drive unit 304.

[0123] The motion determination unit 307 may also perform a learning process to optimize the motion determination at any timing. The flow of these processes is similar to the flow of the processes of the wing car 102 shown in Fig. 6. The KPI set for the arm robot 802 is, for example, the efficiency of the work.

[0124] The lifter 803 has a function of moving the arm robot 802 to any height within a predetermined range. For this purpose, the lifter 803 grasps the situation of the shelf 804 as the surrounding situation. To grasp this situation, a sensor (e.g., a camera, etc.) installed on the lifter 803 is used. The operation of the lifter 803 is determined by the intelligence of the lifter 803 (e.g., an operation determination unit 407 of the lifter 803 described later).

[0125] The configuration of the lifter 803 is similar to that of the wing vehicle 102 of the first embodiment shown in Fig. 3 or the forklift 103 of the first embodiment shown in Fig. 4, except for the following differences. Here, the configuration of the lifter 803 will be described with reference to Fig. 4.

[0126] The communication interface 402 of the lifter 803 communicates with the arm / lifter integrated management system 801 via the network 210. The sensor unit 403 of the lifter 803 is, for example, a camera attached to the lifter 803. Based on an image captured by this camera in the direction of the shelf 804, the situation grasping unit 406 grasps the situation of the shelf 804. Furthermore, the situation grasping unit 406 of the lifter 803 grasps a message transmitted from the arm / lifter integrated management system 801 (for example, information indicating the height to which the arm robot 802 should be lifted, as described later). Furthermore, the drive unit 404 of the lifter 803 includes a motor (not shown) that drives the lifter 803 in the up and down directions, and the like.

[0127] The task information 410 may include, for example, information indicating the position of a shelf storing an item to be taken out by the arm robot 802. The operation determination unit 407 determines the height of the lifter 803 appropriate for the arm robot 802 to take out a desired item 805, based on, for example, information from the arm / lifter integrated management system 801 acquired via the communication interface 402, a state grasped from the result of sensing by the sensor unit 403, and the task information 410. Then, the operation execution unit 408 executes the determined operation (for example, lifting the arm robot 802 to the determined height) via the drive unit 304.

[0128] The operation determination unit 407 may also perform a learning process to optimize the operation determination at any timing. The flow of these processes is similar to the flow of the processes of the forklift 103 shown in Fig. 7. Note that the KPI set for the lifter 803 is, for example, the low number of vertical movements (i.e., the lower the number of vertical movements, the higher the evaluation).

[0129] The arm / lifter integrated management system 801 is a computer that communicates with and manages an arm robot 802 and a lifter 803. The configuration of the arm / lifter integrated management system 801 is similar to the configuration of the control system 101 of the first embodiment shown in Fig. 2, except for the following differences. Here, the configuration of the arm / lifter integrated management system 801 will be described with reference to Fig. 2.

[0130] The communication interface 202 of the arm / lifter integrated management system 801 communicates with the arm robot 802 and the lifter 803 via the network 210. Furthermore, one or more cameras 105 are connected to the network 210. However, the cameras 105 may be connected to the arm / lifter integrated management system 801 without going through the network 210. Each camera 105 may photograph the entire autonomous body system 800, may photograph a predetermined part of the autonomous body system 800, or may photograph the surroundings of the autonomous body system 800 (e.g., the shelves 804 and the items 805). The camera 105 is an example of a sensor for grasping the situation of the autonomous body system 800, and other types of sensors may be used.

[0131] The situation grasping unit 204 grasps the situation of the autonomous body system 800 based on the image captured by the camera 105. The situations grasped here include, for example, the situation of the shelf 804, the situation of the item 805 stored on the shelf 804, the situation of the arm robot 802, and the situation of the lifter 803.

[0132] The task information 208 may include, for example, information for identifying an item to be taken out by the arm robot 802 and information indicating the position of the shelf on which the item is stored. The operation determination unit 205 determines the position of the target item 805 on the shelf 804 based on the situation determined from the image captured by the camera 105 and the task information 208. Then, the operation determination unit 205 determines the height of the lifter 803 (i.e., the height to which the arm robot 802 should be lifted) appropriate for the arm robot 802 to take out the target item 805. The operation execution unit 206 transmits the determined height to the lifter 803 via the communication interface 202. Then, the operation execution unit 206 transmits to the arm robot 802 that the lifter 803 has lifted the arm robot 802 to the determined height.

[0133] The arm robot 802 performs an action to grasp the target article 805. As a result, if the target article 805 cannot be grasped, the situation grasping unit 204 grasps that fact according to information from the camera 105. Based on the grasped information, the action determination unit 205 determines that the height of the lifter 803 was not appropriate, and determines a more appropriate height for the lifter 803. The action execution unit 206 transmits the height to the lifter 803.

[0134] Furthermore, the operation determination unit 205 may perform a learning process to optimize the operation determination at any timing. The flow of these processes is similar to the flow of processes of the control system 101 shown in Fig. 5. Note that the KPI set in the arm / lifter integrated management system 801 is, for example, the efficiency of work of the autonomous system 800 as a whole.

[0135] The autonomous body system 800 of this embodiment may be a part of a larger autonomous body system. For example, the autonomous body system 800 of the second embodiment may be incorporated in the warehouse of the autonomous body system 100 of the first embodiment as a robot that picks up items from a shelf. In this case, the arm / lifter integrated management system 801 of the second embodiment operates as a higher-level autonomous body with respect to the arm robot 802 and the lifter 803, and operates as a lower-level autonomous body with respect to the control system 101.

[0136] In the above example, the arm robot 802 is treated as one lower autonomous body, but in reality, the arm robot 802 may be an autonomous body system consisting of a higher-level autonomous body and a lower-level autonomous body. For example, the lower-level autonomous body may have a sensor unit that grasps the state of each joint 808, a motion decision unit that decides an action based on information obtained from the sensor unit, and a drive unit that drives the joint according to the decided action. On the other hand, the higher-level autonomous body may have a sensor unit that grasps the state of the entire arm robot 802 and its surroundings, a motion decision unit that decides an action based on information obtained from the sensor unit, and a motion execution unit that transmits the decided action to each joint 808.

[0137] In this way, the autonomous body system to which the present invention is applied may have three or more hierarchical layers. Even in this case, as described in the above-mentioned first and second embodiments, each autonomous body determines its own operation from the information acquired by each autonomous body based on the result of learning so as to improve the KPI set for each autonomous body, and thus it is possible to perform an operation that improves the KPI of the autonomous body system as a whole while improving the KPI of each autonomous body. In addition, such an autonomous body system can be constructed as a flexible system that does not require detailed arrangements for the interface between the upper and lower autonomous bodies, does not require prior integration of information, and can adapt to changes in the configuration of the lower autonomous bodies.

[0138] According to the above-described second embodiment, the arm / lifter integrated management system 801 provides information determined to be effective for improving the KPI of the arm / lifter integrated management system 801 to the lower autonomous body (for example, the arm robot 802 and the lifter 803) according to the state of the warehouse grasped by the camera 105 or the like. This allows the lifter 803 to know, for example, to what height the arm robot 802 should be lifted, something that the lifter 803 itself cannot determine. Also, the arm robot 802 can know in advance approximately where the target item 805 is placed. Therefore, it is easy to find the target item 805 when it is lifted by the lifter 803. In other words, the time required to find the target item 805 can be shortened, and the target item 805 can be found reliably.

[0139] In an autonomous body system in which there is no higher-level autonomous body such as the arm / lifter integrated management system 801 and lower-level autonomous bodies directly exchange information, it is difficult to obtain the above-mentioned effect for the following reasons. That is, in a system without a higher-level autonomous body, the lifter 803 cannot know the state of the upper stage of the shelf 804 through the sensor unit 403 when the arm robot 802 has not yet been lifted. Similarly, the arm robot 802 that has not yet been lifted cannot grasp the items on the shelf through the sensor unit 303.

[0140] Meanwhile, according to this embodiment, the arm / lifter integrated management system 801 can grasp the situation of the entire autonomous body system 800 and its surroundings, and based on that, can notify the desired operation of each autonomous body to improve the overall KPI, thereby improving the overall KPI. EXAMPLES

[0141] Next, a third embodiment of the present invention will be described. Except for the differences described below, each part of the autonomous system of the third embodiment has the same function as each part of the autonomous system of the first embodiment shown in Figs. 1 to 7 to which the same reference numerals are attached, and therefore the description thereof will be omitted.

[0142] FIG. 9 is an explanatory diagram showing the overall configuration of an autonomous body system 900 according to a third embodiment of the present invention.

[0143] The autonomous body system 900 of the third embodiment is a parking lot system that includes, as autonomous bodies, an autonomous vehicle 902, a human-driven automobile 903, and an in-parking lot control system 901 that controls a parking lot 904 used by them.

[0144] The parking lot control system 901 monitors where and at what speed vehicles (e.g., including one or more autonomous vehicles 902 and one or more manually driven vehicles 903) are traveling in a large parking lot 904, and how they are occupying the roads in the parking lot (i.e., areas other than the parking area 906 for vehicles to move in). Furthermore, the parking lot control system 901 monitors which sections of the parking area 906 are vacant. The parking lot control system 901 grasps the above monitoring results and performs gentle control over the autonomous vehicles 902 and manually driven vehicles 903.

[0145] In the parking lot 904, for example, one or more guide signs 908 are provided. This is an example of a means for transmitting information from the parking lot control system 901 to the autonomous vehicle 902 and the human-driven vehicle 903.

[0146] The configuration of the autonomous vehicle 902 is similar to the configuration of the wing vehicle 102 of the first embodiment shown in Fig. 3, except for the following differences. Here, the configuration of the autonomous vehicle 902 will be described with reference to Fig. 3.

[0147] A communication interface 302 of the autonomous vehicle 902 communicates with the parking lot control system 901 via the network 210. A sensor unit 303 of the autonomous vehicle 902 is, for example, a camera or a distance sensor that is attached to the autonomous vehicle 902 and captures images of the surroundings of the autonomous vehicle 902. An example of a camera will be described here. A situation grasping unit 306 of the autonomous vehicle 902 grasps the situation around the vehicle in the parking lot 904 based on images captured by the camera.

[0148] Furthermore, the situation ascertaining unit 306 of the autonomous vehicle 902 ascertains information transmitted from the parking lot control system 901 (for example, the situation inside the parking lot 904, information on the destination section and the route there, etc.). The situation ascertaining unit 306 may ascertain this information from information acquired by the camera or the like of the sensor unit 303 (for example, the recognition result of the display content of the guide display board 908), or may ascertain this information from information acquired via the communication interface 302.

[0149] In addition, the drive unit 304 of the autonomous vehicle 902 includes, for example, a prime mover such as an internal combustion engine or an electric motor, a transmission device that transmits the power generated by the prime mover to the wheels, a steering device for driving the autonomous vehicle 902 in a desired direction, and the like.

[0150] The task information 310 includes, for example, information to search for a parking space and move the vehicle there. The operation determination unit 307 determines operations such as proceeding, stopping, and steering amount based on, for example, information from the parking lot control system 901 acquired via the communication interface 302 or the sensor unit 303, a state grasped from the result of sensing by the sensor unit 303, and the task information 310. Then, the operation execution unit 308 executes the determined operation via the drive unit 304.

[0151] Furthermore, the operation determination unit 307 may perform a learning process to optimize the operation determination at any timing. The flow of these processes is similar to the flow of the processes of the wing vehicle 102 shown in Fig. 6. Note that the KPI set for the autonomous vehicle 902 may be, for example, a desire to minimize fuel (or power) consumption, or a desire to be instructed on a parking lot for the vehicle and a route there.

[0152] The autonomous vehicle 902 may have a fully autonomous driving function that allows it to travel without human operation, or may have a driving assistance function that assists with part of the driving by a human. As in the case of the wing vehicle 102 of the first embodiment, at least a part of the functions of the situation grasping unit 306, the operation decision unit 307, and the operation execution unit 308 may be implemented as at least a part of the autonomous driving function or the driving assistance function.

[0153] The configuration of the human-driven automobile 903 is similar to that of the autonomous vehicle 902, except for the following differences. That is, the human-driven automobile 903 may not have the processor 301, the communication interface 302, the sensor unit 303, and the storage device 305. In the human-driven automobile 903, functions corresponding to the situation grasping unit 306, the action decision unit 307, and the action execution unit 308 in the autonomous vehicle 902 are replaced by the driver's intelligence. The task information 310 is stored by the driver. In addition, the sensor unit 303 is replaced by the driver's sensory organs (e.g., the eyes and ears).

[0154] The KPIs set for a human-driven automobile 903 may be similar to those set for an autonomous vehicle 902, but may also be the opposite, for example, that the driver does not want to be told where to park the vehicle and the route to get there, or that the driver wishes to park close to a human exit or in the shade.

[0155] The parking lot control system 901 is a computer that monitors a parking lot 904 and gently controls self-driving cars 902 and human-driven cars 903. The configuration of the parking lot control system 901 is similar to the configuration of the control system 101 of the first embodiment shown in Fig. 2, except for the following differences. The configuration of the parking lot control system 901 will now be described with reference to Fig. 2.

[0156] One or more cameras 105 are connected to the network 210, each of which captures at least a portion of the parking lot 904. Each camera 105 may be a fixed camera installed at an appropriate location, or may be a mobile camera such as a camera-equipped drone. The camera 105 is an example of a sensor for understanding the situation of the autonomous system 900, and other types of sensors may be used.

[0157] Furthermore, a guide display board 908 is connected to the network 210. This is an example of an information transmission means by which the parking lot control system 901 controls each vehicle. Other examples of such information transmission means include traffic lights similar to those in the first embodiment, which are provided in various places in the parking lot 904.

[0158] The situation grasping unit 204 grasps the situation of the autonomous body system 900 based on the images captured by the camera 105. The situation grasped here is, for example, where and at what speed each vehicle is traveling, and how the road surface in the parking lot is occupied.

[0159] The task information 208 may be, for example, to guide a vehicle entering the parking lot 904 to an empty section in the parking lot 904, and to guide a vehicle leaving any section to an exit of the parking lot 904. The operation determination unit 205 determines the control of each vehicle based on the situation of the parking lot 904 (for example, the situation of each vehicle in the parking lot 904) grasped from the image taken by the camera 105 and the task information 208 so as to prevent congestion in the parking lot 904 and to allow each vehicle to park quickly and close to its current position, and the operation execution unit 206 executes the control. The control is performed, for example, by displaying information on a guide display board 908. If the vehicle can receive information via the communication interface 302 connected to the network 210, the operation execution unit 206 may transmit the control information via the communication interface 202 and the network 210.

[0160] Although only one guide sign 908 is shown in FIG. 9, multiple guide signs 908 may actually be installed in the parking lot 904. In this case, each guide sign 908 may all display the same content (for example, all vacant spaces in the parking lot 904), or each may display different content. For example, each guide sign 908 may display only vacant spaces to which it is desired to guide vehicles traveling in the vicinity. Furthermore, each guide sign 908 may display information other than vacant spaces, such as the direction of travel of vehicles. What each guide sign 908 displays in what situation is determined based on the results of learning, which will be described later.

[0161] The operation determination unit 205 may perform a learning process to optimize the operation determination at any timing. Specifically, the operation determination unit 205 may learn from past records, etc., what kind of vehicle flow occurs when what kind of information is provided to the vehicle. The operation determination unit 205 may also search for a method of providing information (for example, a display method on the guide board 908) that performs appropriate control by simulation. At this time, the KPI set in the parking lot control system 901 may be, for example, the utilization efficiency of the parking lot 904 or the satisfaction of people who use the parking lot 904. Then, the operation execution unit 206 provides the vehicle with optimal information in an optimal manner. The flow of these processes is similar to the flow of the processes of the control system 101 shown in FIG. 5.

[0162] According to the above-described third embodiment, the parking position and the movement route thereto are optimized so that the KPI of the autonomous vehicle 902, for example, minimizing fuel consumption, is improved. Also, the parking position and the movement route thereto are optimized so that the KPI of the human-driven vehicle 903, for example, wanting to park quickly, wanting to park near a human exit, or wanting to park in the shade, is improved. Such an effect is difficult to obtain in an autonomous system that does not have a parking lot control system 901, even if each autonomous vehicle 902 and each human-driven vehicle 903 directly exchanges information with each other. EXAMPLES

[0163] Next, a fourth embodiment of the present invention will be described. Except for the differences described below, each part of the autonomous system of the fourth embodiment has the same function as each part of the first embodiment shown in Figs. 1 to 7 to which the same reference numerals are attached, and therefore the description thereof will be omitted.

[0164] FIG. 10 is an explanatory diagram showing the overall configuration of an autonomous body system 1000 according to a fourth embodiment of the present invention.

[0165] The autonomous body system 1000 of the fourth embodiment is a construction machine system including, as autonomous bodies, an excavator 1002, a dump truck 1003, and a person 1004, and a construction site control system 1001 that controls them.

[0166] The construction site control system 1001 monitors the interference and possibility of interference between autonomous bodies (e.g., an excavator 1002, a dump truck 1003, and a person 1004) at the construction site, and performs appropriate control (presentation of information) for each autonomous body. The construction site control system 1001 also monitors the status of things other than the autonomous bodies at the construction site (e.g., the status of excavated holes or the status of materials used in construction, etc.), and performs appropriate control (presentation of information) for each autonomous body.

[0167] In addition, for the sake of simplicity, FIG. 10 shows one excavator 1002, one dump truck 1003, and one person 1004 as autonomous bodies included in the autonomous body system 1000; however, these are merely examples, and the autonomous body system 1000 may include multiple excavators 1002, multiple dump trucks 1003, and multiple people 1004, or may not include at least any of them, or may include other construction machines, such as one or more bulldozers.

[0168] Furthermore, at least one of the one or more excavators 1002 and one or more dump trucks 1003 may be construction machinery that operates autonomously without human operation (or with the assistance of human operation) (hereinafter referred to as autonomous construction machinery), and the others may be general construction machinery operated by a person (hereinafter also referred to as general construction machinery).

[0169] The configuration of the autonomous construction machine may be the same as that of the wing vehicle 102 shown in Fig. 3 or the forklift 103 shown in Fig. 4. Here, refer to Fig. 3. A situation grasping unit 306 of the autonomous construction machine grasps the surrounding situation of the autonomous construction machine at the construction site as the surrounding situation via a sensor unit 303 of the autonomous construction machine, such as a camera.

[0170] The surrounding situation grasped in this way includes information provided from the construction site control system 1001. This information may include, for example, information on whether there is another autonomous body near the autonomous construction machine and whether another autonomous body is approaching. The operation determination unit 307 determines an operation, such as excavation or transportation, based on the grasped situation and task information 310 (for example, information on an operation such as excavation or transportation to be performed), and the operation execution unit 308 executes the determined operation via the drive unit 304.

[0171] The configuration of the general construction machine may be the same as that of the wing vehicle 102 shown in FIG. 3 or the forklift 103 shown in FIG. 4, for example. FIG. 3 is referred to here. However, the general construction machine does not have to have the processor 301, the communication interface 302, the sensor unit 303, and the storage device 305. The sensory organs (e.g., eyes and ears) of the driver of the general construction machine correspond to the sensor unit 303 of the autonomous construction machine. The intelligence of the driver of the general construction machine corresponds to the situation grasping unit 306, the operation decision unit 307, and the operation execution unit 308 of the autonomous construction machine. In addition, information equivalent to the task information 310 (e.g., information regarding the operation such as excavation or transportation to be performed) may be stored by the driver.

[0172] The operator of a general construction machine senses the surroundings of the construction site through his / her own senses. general The surrounding situation of the construction machine is grasped. The surrounding situation grasped in this manner includes information provided from the construction site control system 1001 (for example, information on the approach of other autonomous bodies similar to the above).

[0173] The operator determines an action, such as digging or hauling, based on the understood situation and information corresponding to task information 310, and executes the determined action via the drive unit 304.

[0174] The KPIs set for the autonomous construction machine and the general construction machine may be, for example, "want to work efficiently" or "want to work reliably." Different KPIs may be set for each construction machine.

[0175] The person 1004 grasps the surrounding situation using sensory organs such as the eyes and ears, decides on an action (for example, continuing the current work, avoiding surrounding objects (for example, other approaching autonomous bodies), or rechecking the surroundings), and performs the determined action. The surrounding situation grasped by the person 1004 includes information provided by the construction site control system 1001 (for example, information similar to the above).

[0176] The KPI of the person 1004 may be, for example, wanting to work efficiently, wanting to work reliably, or not getting tired, if the person 1004 is a worker performing construction work at a construction site. Also, if the person 1004 is a person other than a worker (for example, a person who observes the construction site or a person who delivers goods (for example, lunch boxes) to workers), the KPI may be, for example, wanting to achieve each objective efficiently and reliably.

[0177] The configuration of the construction site control system 1001 is the same as that of the control system 101 of the first embodiment shown in FIG. 2. The situation grasping unit 204 of the construction site control system 1001 grasps the overall situation of the construction site as the surrounding situation (for example, the situation of each autonomous body such as the excavator 1002, the dump truck 1003, and the person 1004, as well as the situation of holes and materials in the construction site, etc.) via, for example, one or more cameras 105. Here, each camera 105 may be a fixed camera installed at an appropriate location, or may be a mobile camera such as a drone with a camera. This camera 105 is an example of a sensor for grasping the situation of the autonomous body system 1000, and other types of sensors may be used.

[0178] The task information 208 of the construction site control system 1001 may be, for example, information on work (e.g., excavation, transportation, etc.) to be performed at the construction site. The operation decision unit 205 of the construction site control system 1001 decides the control contents of each autonomous body based on the situation of the construction site grasped from the image taken by the camera 105 and the task information 208 so that the coordinated work of each autonomous body is performed efficiently and reliably. Here, the coordinated work of each autonomous body is, for example, putting the soil excavated by the excavator 1002 into the dump truck 1003, and shaping the hole created by the excavation by the person 1004.

[0179] Then, the operation execution unit 206 performs the determined control. The control is performed by presenting information to each autonomous entity. For example, the traffic light 106 similar to that in the first embodiment may be installed in the construction site, or the necessary information may be provided to each autonomous entity by voice.

[0180] The operation determination unit 205 may perform a learning process to optimize the operation determination at any timing. Specifically, the operation determination unit 205 may learn from past records, etc., what kind of information is provided to each autonomous entity, what kind of operation each entity will perform, and what kind of behavior will result overall. The operation determination unit 205 may also search for a method of providing information that performs appropriate control by simulation. At this time, the KPI set in the construction site control system 1001 may be, for example, the work efficiency of the construction site. Then, the operation execution unit 206 provides each autonomous entity with optimal information in an optimal manner. The flow of these processes is similar to the flow of processes of the control system 101 shown in FIG. 5.

[0181] According to the above-described fourth embodiment, the operation of the autonomous construction machine is optimized in terms of efficiency in accordance with other autonomous bodies and the situation at the site. Similarly, the operation of the general construction machine is optimized in terms of efficiency in accordance with other autonomous bodies and the situation at the site. The operation of the worker and the person other than the worker is optimized in terms of efficiency and fatigue level in accordance with other autonomous bodies and the situation at the site.

[0182] It is difficult to obtain such an effect in an autonomous body system that does not have the construction site control system 1001, even if each autonomous body directly exchanges information with each other. This is due to, for example, the following reasons. That is, there is a possibility that a discrepancy may occur between the situation that each autonomous body grasps and the actual situation. In addition, although it is necessary to determine the means of communication between each autonomous body in advance, this is practically difficult because the provider and operator of each autonomous body are generally different. EXAMPLES

[0183] Next, a fifth embodiment of the present invention will be described. Except for the differences described below, each part of the autonomous system of the fifth embodiment has the same function as each part of the autonomous system of the first embodiment shown in Figs. 1 to 7 to which the same reference numerals are attached, and therefore the description thereof will be omitted.

[0184] FIG. 11 is an explanatory diagram showing the overall configuration of an autonomous body system 1100 according to a fifth embodiment of the present invention.

[0185] The autonomous system 1100 of the fifth embodiment is a screw tightening robot system that combines two workpieces (parts to be processed) and tightens them with screws. The autonomous system 1100 includes a screw tightening monitoring system 1101, a first hand robot 1102, and a second hand robot 1103 as autonomous bodies.

[0186] The first hand robot 1102 places a first workpiece 1104 on a work table 1107, and then places a second workpiece 1105 on the first workpiece 1104 and leaves them stationary.

[0187] The second hand robot 1103 inserts screws 1106 into screw holes 1108 in the first workpiece 1104 and the second workpiece 1105 , and a screw tightening tool 1109 attached to the tip of the arm of the second hand robot 1103 tightens the screws 1106 .

[0188] The screw tightening monitoring system 1101 monitors the position of the second workpiece 1105, any misalignment, and any screw tightening failures, and performs appropriate control (presentation of information) on the first hand robot 1102 and the second hand robot 1103.

[0189] The configurations of the first hand robot 1102 and the second hand robot 1103 may be similar to, for example, the configuration of the wing vehicle 102 shown in Fig. 3 or the forklift 103 shown in Fig. 4. Furthermore, for the parts omitted in the following description, the description of the arm robot 802 in the second embodiment may be applied. Here, Fig. 3 will be referred to for the configuration of the first hand robot 1102, and Fig. 4 will be referred to for the configuration of the second hand robot 1103.

[0190] The situation grasping unit 306 of the first hand robot 1102 grasps the surrounding situation, such as the state of the first workpiece 1104, the state of the second workpiece 1105, and the state of the worktable 1107, via a sensor unit 303 of the first hand robot 1102, such as a camera attached to the first hand robot 1102.

[0191] The situation grasped by the situation grasping unit 306 also includes information provided by the screw tightening monitoring system 1101. This information includes, for example, information on the deviation of the fixing position of each workpiece and the completion or failure of the screw tightening work. The action determining unit 307 determines an action, such as removing each workpiece from a storage location and fixing it in a predetermined position on the workbench 1107, based on the grasped situation and task information 310 (for example, information on the screw tightening work to be performed), and the action executing unit 308 executes the determined action via the driving unit 304.

[0192] The KPI set for the first hand robot 1102 is, for example, to perform work efficiently.

[0193] The situation grasping unit 406 of the second hand robot 1103 grasps the surrounding situation, such as the status of the screw hole 1108, the screw 1106, and the screw tightening tool 1109, via a sensor unit 403 of the second hand robot 1103, such as a camera attached to the second hand robot 1103.

[0194] The situation grasped by the situation grasping unit 406 also includes information provided by the screw tightening monitoring system 1101. This information includes, for example, whether each workpiece is in a state in which the screw tightening work can be started, and information on the completion or failure of the screw tightening work. The operation determination unit 407 determines an operation, such as inserting the screw 1106 into the screw hole 1108 and tightening the screw with the screw tightening tool 1109, based on the grasped situation and task information 410 (for example, information on the screw tightening work to be performed), and the operation execution unit 408 executes the determined operation via the drive unit 404.

[0195] The KPIs set for the second hand robot 1103 are, for example, to perform the work efficiently and to successfully tighten the screws.

[0196] The configuration of the screw tightening monitoring system 1101 is similar to the configuration of the control system 101 of the first embodiment shown in Fig. 2. The situation grasping unit 204 of the screw tightening monitoring system 1101 grasps, as the surrounding situation, for example, the state of the first workpiece 1104 and the second workpiece 1105 (for example, whether the screw holes of each workpiece overlap each other) and the situation of the first hand robot 1102 and the second hand robot 1103, for example, via one or more cameras 105. Here, the camera 105 is, for example, one or more cameras installed at appropriate locations that can capture images of each workpiece and the hand robot. The camera 105 is an example of a sensor for grasping the situation of the autonomous body system 1100, and other types of sensors may be used.

[0197] The task information 208 of the screw tightening monitoring system 1101 is, for example, information about a screw tightening operation to be performed. The operation decision unit 205 of the screw tightening monitoring system 1101 decides the control contents of the first hand robot 1102 and the second hand robot 1103 based on the task information 208 and the situation grasped from the image captured by the camera 105.

[0198] For example, the operation determination unit 205 may grasp the status of the two workpieces (e.g., whether the screw holes are placed with each other overlapping, etc.) and may decide to notify the first hand robot 1102 of an instruction to correct the position of the second workpiece 1105 depending on the status. Also, the operation determination unit 205 may decide to notify the second hand robot 1103 of the status when the screw tightening work can be performed, for example, because the screw holes of the two workpieces overlap. Furthermore, the operation determination unit 205 may grasp whether the screw tightening has been successful or unsuccessful, and may decide to notify the second hand robot 1103 to try again if it has failed.

[0199] Then, the operation execution unit 206 performs the determined control. The control is performed by presenting information to each autonomous entity (for example, transmitting information via the network 210).

[0200] The motion determination unit 205 may perform a learning process for optimizing the determination of the motion at any timing. Specifically, the motion determination unit 205 may learn from past records, etc., what information is provided to each hand robot to result in the result (success or failure) of the screw tightening work, and what is the length of time required for the work. In this case, the KPI set in the screw tightening monitoring system 1101 may be, for example, "want to work efficiently" and "want to successfully tighten the screws." The flow of these processes is similar to the flow of processes of the control system 101 shown in FIG. 5.

[0201] According to the above-described fifth embodiment, the operation of the first hand robot 1102 becomes appropriate in terms of successful screw tightening. Also, the operation of the second hand robot 1103 becomes efficient (e.g., there is less waiting time and there is less possibility of failure).

[0202] It is difficult to obtain such an effect even if the hand robots directly exchange information with each other in an autonomous system that does not have the screw tightening monitoring system 1101. This is due to, for example, the following reasons. That is, there is a possibility that a discrepancy occurs between the situation grasped by each hand robot and the actual situation. Also, there is a possibility that each hand robot cannot observe the situation.

[0203] The present invention is not limited to the above-mentioned embodiment, but includes various modifications. For example, the above-mentioned embodiment has been described in detail for a better understanding of the present invention, and is not necessarily limited to those including all of the configurations described. Also, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. Also, it is possible to add, delete, or replace a part of the configuration of each embodiment with another configuration.

[0204] In addition, the above-mentioned configurations, functions, processing units, processing means, etc. may be realized in part or in whole by hardware, for example, by designing them as integrated circuits. In addition, the above-mentioned configurations, functions, etc. may be realized in software by a processor interpreting and executing a program that realizes each function. Information such as a program, table, file, etc. that realizes each function can be stored in a storage device such as a non-volatile semiconductor memory, a hard disk drive, or an SSD (Solid State Drive), or in a computer-readable non-transitory data storage medium such as an IC card, an SD card, or a DVD.

[0205] In addition, the control lines and information lines are shown as those considered necessary for the explanation, and not all control lines and information lines in the product are shown. In reality, it can be considered that almost all components are connected to each other. [Explanation of symbols]

[0206] 100, 800, 900, 1000, 1100 Autonomous Systems 101 Control System 102 Wing car 103 Forklift 104 Palettes 105 Camera 106 Traffic Lights 801 Lifter Integrated Management System 802 Arm Robot 803 Lifter 901 Parking Lot Control System 902 Self-driving cars 903 Automobiles 904 Parking lot 906 Parking Area 908 Information board 1001 Construction site control system 1002 Excavator 1003 Dump Truck 1004 people 1101 Screw tightening monitoring system 1102 First Hand Robot 1103 Second Hand Robot 1104 First Work 1105 Second Work 1106 Screw

Claims

1. An autonomous body system having a plurality of autonomous bodies, Each of the plurality of autonomous bodies is a non-human autonomous body having a sensor unit, a situation grasping unit that grasps a situation based on a sensing result by the sensor unit, an action determination unit that determines an action based on the grasped situation, and an action execution unit that executes the determined action, The plurality of autonomous entities in the autonomous entity system include one or more first autonomous entities and two or more second autonomous entities; The situation grasped by the situation grasping unit of the first autonomous body includes a situation of the second autonomous body grasped based on a sensing result by the sensor unit of the first autonomous body, The situation grasped by the situation grasping unit of the second autonomous body includes a result of an operation executed by an operation executing unit of the first autonomous body, which is grasped based on a sensing result by the sensor unit of the second autonomous body; The operation decision unit of the second autonomous body may decide an operation according to a result of the operation executed by the operation execution unit of the first autonomous body, or may decide an operation contrary to the result of the operation executed by the operation execution unit of the first autonomous body; An autonomous body system characterized in that the operation decision unit of the first autonomous body determines the operation of the first autonomous body so that an evaluation parameter calculated based on the situation grasped by the situation grasping unit of the first autonomous body becomes a value indicating a high evaluation.

2. 2. The autonomous system of claim 1, The autonomous body system is characterized in that the operation decision unit of the first autonomous body holds a function that decides an operation based on the situation grasped by the situation grasping unit of the first autonomous body, and changes a parameter of the function so that the evaluation parameter becomes a value indicating a high evaluation.

3. An autonomous body system having a plurality of autonomous bodies, Each of the plurality of autonomous bodies is a non-human autonomous body having a sensor unit, a situation grasping unit that grasps a situation based on a sensing result by the sensor unit, an action determination unit that determines an action based on the grasped situation, and an action execution unit that executes the determined action, The plurality of autonomous entities in the autonomous entity system include one or more first autonomous entities and two or more second autonomous entities; The situation grasped by the situation grasping unit of the first autonomous body includes a situation of the second autonomous body grasped based on a sensing result by the sensor unit of the first autonomous body, The situation grasped by the situation grasping unit of the second autonomous body includes a result of an operation executed by an operation executing unit of the first autonomous body, which is grasped based on a sensing result by the sensor unit of the second autonomous body; The operation decision unit of the second autonomous body determines the operation of the second autonomous body so that an evaluation parameter calculated based on the situation grasped by the situation grasping unit of the second autonomous body becomes a value indicating a high evaluation, thereby sometimes determining an operation that conforms to the result of the operation performed by the operation execution unit of the first autonomous body, and sometimes determining an operation that is contrary to the result of the operation performed by the operation execution unit of the first autonomous body.

4. 4. The autonomous system of claim 3, The situation grasped by the situation grasping unit of the second autonomous entity includes a result of the operation executed by the operation executing unit of the first autonomous entity and other situations; The autonomous body system is characterized in that the operation decision unit of the second autonomous body holds a function that decides an operation based on the result of the operation executed by the operation execution unit of the first autonomous body and other circumstances, and changes a parameter of the function so that the evaluation parameter becomes a value indicating a high evaluation.

5. 5. The autonomous system of claim 4, An autonomous system, characterized in that the parameters of the function include weights assigned to the result of the operation executed by the operation execution unit of the first autonomous body and other situations.

6. A method for controlling an autonomous body system having a plurality of autonomous bodies, comprising the steps of: Each of the plurality of autonomous bodies is an autonomous body other than a human that includes a sensor unit, The control method includes a situation grasping step in which each of the plurality of autonomous bodies grasps a situation based on a sensing result by the sensor unit, an operation determination step in which the autonomous bodies determine an operation based on the grasped situation, and an operation execution step in which the autonomous bodies execute the determined operation, The plurality of autonomous entities in the autonomous entity system include one or more first autonomous entities and two or more second autonomous entities; The situation grasped by the first autonomous body in the situation grasping procedure includes a situation of the second autonomous body grasped based on a sensing result by the sensor unit of the first autonomous body, the situation grasped by the second autonomous body in the situation grasping procedure includes a result of an operation executed in the operation execution procedure of the first autonomous body, which is grasped based on a sensing result by the sensor unit of the second autonomous body; In the action decision procedure, the second autonomous body may decide an action according to a result of the action executed in the action execution procedure of the first autonomous body, or may decide an action contrary to a result of the action executed in the action execution procedure of the first autonomous body; A control method characterized in that, in the operation decision procedure, the first autonomous body determines the operation of the first autonomous body so that an evaluation parameter calculated based on the situation grasped in the situation grasping procedure of the first autonomous body becomes a value indicating a high evaluation.

7. 7. The control method according to claim 6, A control method characterized in that the first autonomous body holds a function that determines an action based on the situation grasped in the situation grasping procedure of the first autonomous body, and in the action determination procedure, a parameter of the function is changed so that the evaluation parameter becomes a value indicating a high evaluation.

8. A method for controlling an autonomous body system having a plurality of autonomous bodies, comprising the steps of: Each of the plurality of autonomous bodies is an autonomous body other than a human that includes a sensor unit, The control method includes a situation grasping step in which each of the plurality of autonomous bodies grasps a situation based on a sensing result by the sensor unit, an operation determination step in which the autonomous bodies determine an operation based on the grasped situation, and an operation execution step in which the autonomous bodies execute the determined operation, The plurality of autonomous entities in the autonomous entity system include one or more first autonomous entities and two or more second autonomous entities; The situation grasped by the first autonomous body in the situation grasping procedure includes a situation of the second autonomous body grasped based on a sensing result by the sensor unit of the first autonomous body, the situation grasped by the second autonomous body in the situation grasping procedure includes a result of an operation executed in the operation execution procedure of the first autonomous body, which is grasped based on a sensing result by the sensor unit of the second autonomous body; A control method characterized in that, in the operation decision procedure, the second autonomous body determines an operation of the second autonomous body so that an evaluation parameter calculated based on the situation grasped in the situation grasping procedure of the second autonomous body becomes a value indicating a high evaluation, thereby sometimes determining an operation that follows the result of the operation performed in the operation execution procedure of the first autonomous body, and sometimes determining an operation that is contrary to the result of the operation performed in the operation execution procedure of the first autonomous body.

9. 9. The control method according to claim 8, the situation grasped by the second autonomous entity in the situation grasping procedure includes a result of the operation executed in the operation execution procedure of the first autonomous entity and other situations; The second autonomous entity is A control method characterized by retaining a function that determines an action based on the result of the action executed in the action execution procedure of the first autonomous body and other circumstances, and changing a parameter of the function in the action decision procedure so that the evaluation parameter becomes a value that indicates a high evaluation.

10. 10. The control method according to claim 9, A control method characterized in that the parameters of the function include weights assigned to the results of the operation executed in the operation execution procedure of the first autonomous body and other situations.

Citation Information

Patent Citations

  • Control system and the control method for robot device

    JP2006000954A

  • Simulation device of a plurality of robots

    JP2015208811A

  • Traffic flow control device, traffic flow control system, road sign, traffic light, traffic flow control method and program

    JP2018067111A

  • Autonomous community energy management system and method

    JP2019110742A