Predictive delivery device, predictive delivery method, and predictive delivery program
The predictive delivery device addresses the challenge of anticipating and preparing for rescue situations by using sensing data to predict needs and dispatching equipped vehicles, ensuring timely and effective resource allocation for rescue operations.
Patent Information
- Application Number
- PCT/JP2023/020237
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-05-31
- Publication Date
- 2024-12-05
AI Technical Summary
Existing systems lack the capability to predict the occurrence of a rescue situation and deploy an autonomous vehicle with the necessary equipment before it is needed, as they do not address the prediction of rescue events or planning for the types and numbers of required equipment.
A predictive delivery device that includes a prediction unit to forecast rescue situations based on sensing data from target facilities, a delivery unit to dispatch equipped mobile objects, and a system that selects and prepares the appropriate rescue supplies, ensuring a rescue vehicle is positioned before a person requires assistance.
Enables the timely and effective deployment of rescue supplies and equipment by predicting rescue situations and dispatching the necessary resources, allowing for quick provision of medical and sanitary aid when needed.
Smart Images

Figure JP2023020237_05122024_PF_FP_ABST
Abstract
Description
Predictive delivery device, predictive delivery method, and predictive delivery program
[0001] The present disclosure relates to a technology for deploying a rescue vehicle before a person in need of rescue appears.
[0002] Patent Literature 1 discloses deploying an autonomous vehicle based on external information when a person in need of rescue appears. However, Patent Literature 1 does not disclose deploying an autonomous vehicle by predicting the appearance of a person in need of rescue. Furthermore, Patent Literature 1 does not disclose planning the types and numbers of equipment required for rescue.
[0003] Japanese Patent Application Laid-Open No. 2022-94806
[0004] The present disclosure aims to enable the deployment of a rescue vehicle before a person in need of rescue appears.
[0005] The predictive delivery device disclosed herein includes a prediction unit that predicts the occurrence of a rescue-requiring situation at a target facility based on sensing data obtained by sensing the target facility, and a delivery unit that delivers a target mobile object equipped with rescue supplies to the target facility when the occurrence of the rescue-requiring situation is predicted.
[0006] According to the present disclosure, it is possible to deploy a rescue vehicle before a person in need of rescue appears.
[0007] FIG. 1 is a configuration diagram of a predictive delivery device 100 according to embodiment 1. FIG. 2 is a diagram showing an example of the configuration of a predictive delivery system 200 according to embodiment 1. FIG. 3 is a diagram showing an overview of a predictive delivery method according to embodiment 1. FIG. 4 is a flowchart of the predictive delivery method according to embodiment 1. FIG. 5 is a hardware configuration diagram of the predictive delivery device 100 according to embodiment 1.
[0008] In the embodiments and drawings, the same or corresponding elements are denoted by the same reference numerals. The description of elements denoted by the same reference numerals as those already described will be omitted or simplified as appropriate. Arrows in the drawings primarily indicate the flow of data or the flow of processing.
[0009] First Embodiment A predictive delivery device 100 will be described with reference to FIGS.
[0010] *** Description of Configuration *** The configuration of the predictive delivery device 100 will be described with reference to Figure 1. The predictive delivery device 100 is a computer equipped with hardware such as a processor 101, memory 102, auxiliary storage device 103, communication device 104, and input / output interface 105. These pieces of hardware are connected to each other via signal lines.
[0011] The processor 101 is an IC that performs arithmetic processing and controls other hardware. For example, the processor 101 is a CPU. IC is an abbreviation for Integrated Circuit. CPU is an abbreviation for Central Processing Unit.
[0012] The memory 102 is a volatile or non-volatile storage device. The memory 102 is also called a primary storage device or a main memory. For example, the memory 102 is a RAM. Data stored in the memory 102 is saved in the secondary storage device 103 as needed. RAM is an abbreviation for Random Access Memory.
[0013] The auxiliary storage device 103 is a non-volatile storage device. For example, the auxiliary storage device 103 is a ROM, a HDD, a flash memory, or a combination of these. Data stored in the auxiliary storage device 103 is loaded into the memory 102 as needed. ROM is an abbreviation for Read Only Memory. HDD is an abbreviation for Hard Disk Drive.
[0014] The communication device 104 is a receiver and a transmitter. For example, the communication device 104 is a communication chip or NIC. Communication of the predictive distribution device 100 is performed using the communication device 104. NIC is an abbreviation for Network Interface Card.
[0015] The input / output interface 105 is a port to which an input device and an output device are connected. For example, the input / output interface 105 is a USB terminal, the input devices are a keyboard and a mouse, and the output device is a display. Input and output of the predictive delivery device 100 is performed using the input / output interface 105. USB is an abbreviation for Universal Serial Bus.
[0016] The predictive delivery device 100 includes elements such as an acquisition unit 111, a prediction unit 112, an output unit 113, and a delivery unit 114. These elements are realized by software.
[0017] The auxiliary storage device 103 stores a predictive delivery program for causing the computer to function as an acquisition unit 111, a prediction unit 112, an output unit 113, and a delivery unit 114. The predictive delivery program is loaded into the memory 102 and executed by the processor 101. The auxiliary storage device 103 also stores an OS. At least a portion of the OS is loaded into the memory 102 and executed by the processor 101. The processor 101 executes the predictive delivery program while running the OS. OS is an abbreviation for Operating System.
[0018] Input and output data for the predictive delivery program are stored in the storage unit 120. The memory 102 functions as the storage unit 120. However, a storage unit such as the auxiliary storage unit 103, a register in the processor 101, or a cache memory in the processor 101 may function as the storage unit 120 instead of or together with the memory 102.
[0019] The predictive delivery program can be recorded (stored) in a computer-readable manner on a non-volatile recording medium such as an optical disk or flash memory.
[0020] An example of the configuration of the predictive delivery system 200 will be described with reference to FIG. 2 . However, the configuration of the predictive delivery system 200 is not limited to the configuration shown in FIG. 2 . The predictive delivery system 200 is a system that uses a predictive delivery device 100. The predictive delivery device 100 is located in a back-of-house 210 of a facility 220. The facility 220 is a facility (target facility) where sensing and predictive delivery are performed. The facility 220 has multiple locations 221. Various sensors 222 for sensing are installed in the facility 220. Examples of sensors include a weather sensor, a thermograph, a thermometer, a hygrometer, and a camera. For example, the various sensors 222 are installed in each location 221. However, the various sensors 222 may also be installed outside the facility 220. Data obtained by the various sensors 222 is referred to as sensing data. The sensing data indicates, for example, weather, body temperature, humidity, or congestion. A charging station 211 is installed in the back-of-house 210. The charging station 211 is a location (waiting location) where multiple transport robots 201 wait. Multiple chargers are installed in the charging station 211. The transport robots 201 wait while charging using the chargers. The transport robots 201 are autonomously moving mobile objects and are called automated transport robots (AMRs). The transport robots 201 are equipped with equipment 213. The equipment 213 is items used for rescue (rescue supplies). The predictive delivery device 100 predicts the occurrence of a rescue-requiring situation at each location 221 in the facility 220 based on the sensing data. A rescue-requiring situation is a situation in which rescue is necessary. In other words, a rescue-requiring situation is a situation in which a person (a person requiring rescue) needs to be rescued. When the occurrence of a rescue-requiring situation is predicted, the predictive delivery device 100 selects a transport robot 201 (target mobile object) equipped with the necessary equipment 213 and transmits a movement instruction to the selected transport robot 201. Upon receiving the movement instruction, the transport robot 201 moves to the location where it is predicted that a rescue-requiring situation will occur (predicted occurrence location) and waits there. If there is no transport robot 201 equipped with the necessary equipment 213, a worker 202 takes the necessary equipment from the equipment shelf 212 and supplies it to the transport robot 201. Then, the transport robot 201 moves to the predicted occurrence location and waits there.When a rescue situation occurs, the necessary equipment 213 in the transport robot 201 is used. After that, the transport robot 201 returns to the facility 220 and waits while charging at the charging station 211.
[0021] ***Explanation of Operation*** The operational procedures of the predictive delivery device 100 correspond to the predictive delivery method. Also, the operational procedures of the predictive delivery device 100 correspond to the processing procedures of the predictive delivery program.
[0022] An overview of the predictive delivery method will be described with reference to Fig. 3. The prediction unit 112 predicts the occurrence of a rescue-requiring situation at a target facility based on sensing data 121 obtained by sensing the target facility. When a rescue-requiring situation is predicted to occur at the target facility, the delivery unit 114 delivers a target mobile vehicle equipped with rescue supplies to the target facility.
[0023] The predictive delivery method will be described in detail with reference to FIG. 4. The flow shown in FIG. 4 is executed repeatedly. A plurality of mobile objects are prepared. Each mobile object is equipped with one or more types of rescue equipment and moves autonomously. Examples of the mobile object are a robot, a drone, or a vehicle. The mobile object includes a storage space, a position sensor, a communication device, a computer (calculating device), and the like. The storage space is a space where the rescue equipment is stored. The position sensor is a sensor for positioning. Examples of the position sensor include a satellite positioning receiver and an inertial measurement unit. Note that the mobile object may be rideable by a person. In other words, the mobile object may deliver rescue personnel such as a doctor or nurse along with the rescue equipment.
[0024] In step S110, the acquisition unit 111 acquires sensing data 121. The sensing data 121 is obtained by sensing the target facility. Specifically, the sensing data 121 is obtained by sensing multiple locations within the target facility. For example, the sensing data 121 indicates weather information, temperature information, humidity information, images, thermography information, etc. for each location within the target facility. For example, the acquisition unit 111 receives measurement data from each sensor installed at each location within the target facility. A collection of the received measurement data becomes the sensing data 121.
[0025] Furthermore, the acquisition unit 111 acquires location information of each mobile object. For example, each mobile object acquires location information using a location sensor and transmits the location information to the predictive delivery device 100. Then, the acquisition unit 111 receives the location information of each mobile object.
[0026] In step S120, the prediction unit 112 predicts the occurrence of a rescue-requiring situation in the target facility based on the sensing data 121. When it is predicted that a rescue-requiring situation will occur in the target facility, the prediction unit 112 predicts the type and number of rescue supplies required based on the sensing data 121. When it is predicted that a rescue-requiring situation will occur in the target facility, the prediction unit 112 predicts the location where the rescue-requiring situation will occur among multiple locations in the target facility based on the sensing data 121.
[0027] Prediction is performed as follows. The prediction unit 112 inputs the sensing data 121 and calculates a prediction model. This results in a prediction result. The prediction result indicates whether a rescue-requiring incident will occur at the target facility. If a rescue-requiring incident is predicted to occur at the target facility, the prediction result indicates the location where the rescue-requiring incident will occur, the type of rescue equipment required, and the number of rescue equipment required. Rescue equipment includes sanitary and medical supplies, first aid supplies, water, ice packs, mist generators, etc. The prediction model estimates the risk of heatstroke and the congestion level of each location, and makes a prediction based on the estimation results. The congestion level of each location is estimated using images acquired by cameras installed at each location. The prediction model is a trained model and is generated in advance by machine learning correct answer data. For example, the correct answer data is a prediction result based on past sensing data and past risk information, to which a correct answer label has been assigned. The correct answer label indicates, for example, "risk of heatstroke." The prediction model is generated by machine learning the correct answer data so that the information of the prediction result approaches the information of the correct label. Examples of machine learning are support vector machines, deep learning, etc.
[0028] In step S130, the output unit 113 and the delivery unit 114 determine whether a rescue-requiring situation is predicted to occur at the target facility. If a rescue-requiring situation is predicted to occur at the target facility, the process proceeds to step S140. If a rescue-requiring situation is not predicted to occur at the target facility, the process ends.
[0029] In step S140, the output unit 113 outputs the prediction result. The prediction result indicates the location where a rescue-requiring situation will occur, the type of rescue equipment required, and the number of rescue equipment required. Furthermore, the output unit 113 outputs the location information of each moving object. For example, the output unit 113 displays the prediction result and the location information of each moving object on a display.
[0030] In step S150, the delivery unit 114 determines whether there is a mobile object equipped with the necessary rescue equipment. In other words, the delivery unit 114 determines whether there is a mobile object equipped with the predicted number or more of the predicted types of rescue equipment. A mobile object equipped with the necessary rescue equipment is referred to as a relevant mobile object.
[0031] For example, whether or not a relevant moving object exists is determined as follows: The moving object is equipped with a detection sensor. The detection sensor detects information (type and number) of rescue equipment equipped on the moving object. The acquisition unit 111 receives the detection information from each moving object. The delivery unit 114 determines whether or not a relevant moving object exists based on each detection information.
[0032] If the relevant moving object is present, the process proceeds to step S160, whereas if the relevant moving object is not present, the process proceeds to step S170.
[0033] In step S160, the delivery unit 114 selects a target mobile object from among the plurality of mobile objects. The target mobile object is a mobile object to be delivered to a target location.
[0034] When there are two or more relevant moving objects, the delivery unit 114 selects one of the relevant moving objects as the target moving object based on the location information of each moving object. For example, the delivery unit 114 selects the relevant moving object closest to the target location as the target moving object.
[0035] The description of step S160 continues. The delivery unit 114 transmits a movement command to the target mobile object to the target location within the target facility. The target mobile object receives the movement command and moves to the target location within the target facility. As a result, the target mobile object is delivered to the target location within the target facility. After step S160, the processing ends.
[0036] In step S170, the delivery unit 114 outputs a replenishment request message. The replenishment request message is a message for requesting a worker to replenish the necessary rescue supplies to any of the multiple mobile objects. For example, the delivery unit 114 displays the replenishment request message on a display.
[0037] The worker checks the replenishment request message, selects one of the plurality of moving objects as a target moving object, and replenishes the target moving object with the necessary rescue supplies. For example, the worker selects the target moving object by referring to the location information of each moving object.
[0038] After the necessary rescue supplies have been replenished, the target mobile object receives a movement command. For example, the mobile object is equipped with a user interface. A worker operates the user interface of the target mobile object to input a movement command to the target mobile object. The worker may specify the target mobile object to the predictive delivery device 100. In this case, the delivery unit 114 transmits the movement command to the target mobile object. The target mobile object then moves to the target location within the target facility. As a result, the target mobile object is delivered to the target location within the target facility. After step S170, the processing ends.
[0039] ***Effects of First Embodiment*** According to the first embodiment, it is possible to predict the occurrence of people needing rescue and the demand for supplies based on sensing data, and to deploy mobile objects before people needing rescue appear. In other words, according to the first embodiment, it is possible to quickly provide sanitary medical supplies, first aid supplies, and the like.
[0040] *** Supplementary Information to the First Embodiment *** The predictive delivery device 100 utilizes sensing data obtained from devices (sensors) installed in equipment or facilities to predict required services using demand forecasting AI (such as LSTM or LightGBM). Based on the prediction results, the predictive delivery device 100 then plans the types and quantities of supplies required and dispatches mobile vehicles. LSTM is an abbreviation for Long Short Term Memory. GBM is an abbreviation for Gradient Boosting Machine.
[0041] The hardware configuration of the predictive delivery device 100 will be described with reference to Figure 5. The predictive delivery device 100 includes a processing circuit 109. The processing circuit 109 is hardware that implements an acquisition unit 111, a prediction unit 112, an output unit 113, and a delivery unit 114. The processing circuit 109 may be dedicated hardware, or may be a processor 101 that executes a program stored in memory 102.
[0042] When the processing circuit 109 is dedicated hardware, the processing circuit 109 may be, for example, a single circuit, a multiple circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof. ASIC is an abbreviation for Application Specific Integrated Circuit. FPGA is an abbreviation for Field Programmable Gate Array.
[0043] In the processing circuit 109, some functions may be realized by dedicated hardware, and the remaining functions may be realized by software or firmware.
[0044] Thus, the functions of the predictive delivery device 100 can be realized by hardware, software, firmware, or a combination of these.
[0045] The first embodiment is an example of a preferred embodiment and is not intended to limit the technical scope of the present disclosure. The first embodiment may be implemented in part or in combination with other embodiments. The procedures described using flowcharts and the like may be modified as appropriate.
[0046] The "part" of each element of the predictive delivery device 100 may be read as a "process," a "step," a "circuit," or a "circuitry."
[0047] 100 Predictive delivery device, 101 Processor, 102 Memory, 103 Auxiliary storage device, 104 Communication device, 105 Input / output interface, 109 Processing circuit, 111 Acquisition unit, 112 Prediction unit, 113 Output unit, 114 Delivery unit, 120 Memory unit, 121 Sensing data, 200 Predictive delivery system, 201 Transport robot, 202 Worker, 210 Backyard, 211 Charging station, 212 Equipment shelf, 213 Equipment, 220 Facility, 221 Location, 222 Sensor.
Claims
1. A predictive delivery device comprising: a prediction unit that predicts the occurrence of a rescue-requiring situation at a target facility based on sensing data obtained by sensing the target facility; and a delivery unit that delivers a target mobile object equipped with rescue supplies to the target facility when the occurrence of the rescue-requiring situation is predicted.
2. The predictive delivery device according to claim 1, wherein the prediction unit predicts the type and number of rescue supplies required based on the sensing data when the occurrence of the rescue-requiring situation is predicted.
3. A predictive delivery device as described in claim 2, in which a plurality of autonomously moving bodies equipped with one or more types of rescue equipment are provided, and the delivery unit selects a mobile body from the plurality of moving bodies that is equipped with a predicted number or more of the predicted type of rescue equipment as the target mobile body, and delivers the target mobile body to the target facility by sending a movement command to the target mobile body.
4. The predictive delivery device according to claim 3, wherein, when there is no mobile body equipped with the rescue supplies of the predicted type in the predicted number or more, the delivery unit outputs a replenishment request message requesting a worker to replenish the necessary rescue supplies to any one of the plurality of mobile bodies.
5. A predictive delivery device as described in any one of claims 1 to 4, wherein the sensing data is obtained by sensing a plurality of locations within the target facility, and when it is predicted that the rescue-requiring incident will occur, the prediction unit predicts the location among the plurality of locations within the target facility where the rescue-requiring incident will occur based on the sensing data, and the delivery unit delivers the target moving object to the predicted location within the target facility.
6. A predictive delivery device as described in any one of claims 1 to 5, wherein the prediction unit predicts the occurrence of the rescue-requiring situation by inputting the sensing data and using a trained model.
7. The predictive delivery device according to any one of claims 1 to 6, further comprising an output unit that outputs the predicted result of the occurrence of the rescue-requiring situation.
8. A predictive delivery method, comprising: predicting the occurrence of a rescue-requiring situation at a target facility based on sensing data obtained by sensing the target facility; and, when the occurrence of the rescue-requiring situation is predicted, delivering a target mobile object equipped with rescue supplies to the target facility.
9. A predictive delivery program for causing a computer to execute a predictive process for predicting the occurrence of a rescue-requiring situation at a target facility based on sensing data obtained by sensing the target facility, and a delivery process for delivering a target mobile object equipped with rescue supplies to the target facility when the occurrence of the rescue-requiring situation is predicted.
Citation Information
Patent Citations
Disaster correspondence supporting system
JP2003281324A
Control apparatus for vehicle, control method of the vehicle, and computer program for controlling vehicle
JP2022159896A
Work system, control server and work method
JP2023042074A
