Search

EP-4372667-B1 - UNMANNED MACHINE MANAGEMENT SYSTEM, UNMANNED MACHINE MANAGEMENT METHOD, DATA ACCUMULATION DEVICE, DATA ANALYSIS DEVICE, DISASTER PREDICTION DEVICE, UNMANNED MACHINE CONTROL DEVICE, AND INFORMATION PROVISION DEVICE

EP4372667B1EP 4372667 B1EP4372667 B1EP 4372667B1EP-4372667-B1

Inventors

  • KONISHI YOSHIAKI

Dates

Publication Date
20260506
Application Date
20220331

Claims (13)

  1. An unmanned vehicle management system comprising: a collection unit (200) configured to collect video data acquired by an unmanned vehicle (100) and natural disaster data related to natural disasters from information sources, wherein the video data is captured by a camera device of the unmanned vehicle (100) in a monitored area and includes first video data collected during normal times and second video data collected during a disaster, wherein the natural disaster data is from at least one of a satellite device, a weather radar device, and a sensor device deployed in the monitored area and is data for predicting natural disasters and includes first natural disaster data collected during normal times and second natural disaster data collected during the disaster; a storage unit (300) configured to store the video data and the natural disaster data collected by the collection unit (200); an analysis unit (400) configured to extract feature values of the first video data and the first natural disaster data collected by the collection unit (200), and predict a high-risk area where a risk of natural disaster occurrence is higher than in other areas; a prediction unit (530) configured to compare the second video data and the second natural disaster data collected during the disaster with the first video data and the first natural disaster data collected during the normal times using the feature values extracted by the analysis unit, and predict a disaster occurrence area where a disaster will occur, the disaster occurrence area including an area where the disaster will expand and an area where a secondary disaster will occur; and a deployment unit (520) configured to determine deployment of the unmanned vehicle (100) and a rescuer based on the high-risk area and the disaster occurrence area, wherein the deployment unit (520) is configured to determine to deploy a first unmanned vehicle in an area where the rescuer has been deployed, deploy a second unmanned vehicle on an evacuation route from the area where the rescuer has been deployed to an evacuation site, and deploy a third unmanned vehicle in an area where a secondary disaster is expected to occur around the area where the rescuer has been deployed, and the prediction unit (530) is further configured to predict the occurrence of the disaster by comparing the second video data and the second natural disaster data acquired by the first unmanned vehicle with the first video data and the first natural disaster data, predict the occurrence of the disaster by comparing the second video data and the second natural disaster data acquired by the second unmanned vehicle with the first video data and the first natural disaster data, and predict the occurrence of the secondary disaster by comparing the second video data and the second natural disaster data acquired by the third unmanned vehicle with the first video data and the first natural disaster data.
  2. The unmanned vehicle management system according to claim 1, wherein the deployment unit (520) is configured to, when a disaster occurs, determine to deploy the unmanned vehicle (100) and the rescuer in the high-risk area and the disaster occurrence area.
  3. The unmanned vehicle management system according to claim 1 or 2, wherein the collection unit (200) is configured to collect residential area data regarding an area where a resident is present, the storage unit (300) is configured to store the residential area data, and the deployment unit (520) is configured to, when a disaster occurs, determine to deploy the unmanned vehicle (100) and the rescuer in the area where the resident is present.
  4. The unmanned vehicle management system according to claim 1 or 2, wherein the storage unit (300) is configured to store information indicating a rescue area where a disaster victim has been rescued in the past, and the deployment unit (520) is configured to determine to deploy the unmanned vehicle (100) and the rescuer in the rescue area.
  5. The unmanned vehicle management system according to any one of claims 1 to 4, comprising: an information presentation unit (540) configured to be worn by the rescuer, wherein the information presentation unit (540) is configured to, based on a direction in which the rescuer is looking, switch information based on the video data acquired by the first unmanned vehicle (100), information based on the video data acquired by the second unmanned vehicle (100), and information based on the video data acquired by the third unmanned vehicle (100), and superimpose the switched information onto a video or scenery.
  6. The unmanned vehicle management system according to claim 1 or 2, wherein the deployment unit (520) is configured to determine to deploy a plurality of unmanned vehicles (100) in each of the high-risk area and the disaster occurrence area, the plurality of unmanned vehicles (100) are configured to acquire an amount of moisture in a ground using an optical sensor, while moving in parallel.
  7. The unmanned vehicle management system according to claim 1 or 2, wherein the deployment unit (520) is configured to determine to deploy in the high-risk area, the disaster occurrence area, or a residential area where a resident is present, a fourth unmanned vehicle (100) having a relay function to relay a signal from a base station fixedly installed, and the unmanned vehicle (100) deployed in the high-risk area, the disaster occurrence area, or the residential area is configured to transmit video data to the base station via the fourth unmanned vehicle (100).
  8. An unmanned vehicle management method comprising: collecting video data acquired by an unmanned vehicle (100) and natural disaster data related to natural disasters from information sources, wherein the video data is captured by a camera device of the unmanned vehicle (100) in a monitored area and includes first video data collected during normal times and second video data collected during a disaster, wherein the natural disaster data is from at least one of a satellite device, a weather radar device, and a sensor device deployed in the monitored area and is data for predicting natural disasters and includes first natural disaster data collected during normal times and second natural disaster data collected during the disaster; storing the video data and the natural disaster data; extracting feature values of the first video data and the first natural disaster data, and predicting a high-risk area where a risk of natural disaster occurrence is higher than in other areas; comparing the second video data and the second natural disaster data collected during the disaster with the first video data and the first natural disaster data collected during the normal times using the feature values extracted, and predicting a disaster occurrence area where a disaster will occur, the disaster occurrence area including an area where the disaster will expand and an area where a secondary disaster will occur; and determining deployment of the unmanned vehicle (100) and a rescuer based on the high-risk area and the disaster occurrence area, wherein the method further comprises: deploying a first unmanned vehicle (100) in an area where the rescuer has been deployed; deploying a second unmanned vehicle (100) on an evacuation route from the area where the rescuer has been deployed to an evacuation site; deploying a third unmanned vehicle (100) in an area where a secondary disaster is expected to occur around the area where the rescuer has been deployed; predicting the occurrence of the disaster by comparing the second video data and the second natural disaster data acquired by the first unmanned vehicle with the first video data and the first natural disaster data, predicting the occurrence of the disaster by comparing the second video data and the second natural disaster data acquired by the second unmanned vehicle with the first video data and the first natural disaster data, and predicting the occurrence of the secondary disaster by comparing the second video data and the second natural disaster data acquired by the third unmanned vehicle with the first video data and the first natural disaster data.
  9. The unmanned vehicle management method according to claim 8, further comprising: when a disaster occurs, determining to deploy the unmanned vehicle (100) and the rescuer in the high-risk area and the disaster occurrence area.
  10. The unmanned vehicle management method according to claim 8, further comprising: collecting residential area data regarding an area where a resident is present; storing the residential area data; and when a disaster occurs, determining to deploy the unmanned vehicle (100) and the rescuer in the area where the resident is present.
  11. The unmanned vehicle management method according to claim 8, further comprising: storing information indicating a rescue area where a disaster victim has been rescued in the past; and determining to deploy the unmanned vehicle (100) and the rescuer in the rescue area.
  12. The unmanned vehicle management method according to any one of claims 8 to 12, further comprising: based on a direction in which the rescuer is looking, switching information based on the video data acquired by the first unmanned vehicle (100), information based on the video data acquired by the second unmanned vehicle (100), and information based on the video data acquired by the third unmanned vehicle (100), and superimposing the switched information onto a video or scenery.
  13. The unmanned vehicle management method according to claim 8, further comprising: determining to deploy a plurality of unmanned vehicles (100) in each of the high-risk area and the disaster occurrence area; and acquiring, by the plurality of unmanned vehicles (100), an amount of moisture in a ground using an optical sensor, while the plurality of unmanned vehicle (100) are moving in parallel.

Description

[Technical Field] The present disclosure relates to an unmanned vehicle management system and an unmanned vehicle management method. [Background Art] Conventionally, techniques for using unmanned vehicles for the purpose of understanding disaster situations, for example, have been known. For example, a disaster activity support system described in Patent Document 1 supports disaster activities using mobile objects by communicating information with a terminal mounted on a mobile object that moves to a site and performs disaster activities when a disaster occurs. The disaster activity support system collects from the terminal local information acquired by the mobile object, calculates an activity point of the mobile object and a travel route from the current location to the activity point, based on the local information and map information, and transmits a result of the calculation to the terminal as instruction information. A system described in Patent Document 2 is a system for inspecting water facilities that are difficult for workers to reach, and the system transmits an operation control signal generated by a control device to a drone and displays data acquired by the drone. An information processing device described in Patent Document 3 combines a plurality of first captured images captured by a plurality of flying objects, and controls the flight of the plurality of flying objects based on an operation for changing a first image range that is the image range of the composite image. KR20170101519A discloses a disaster monitoring apparatus using an unmanned aerial vehicle and a method thereof. According to the present invention, the disaster monitoring apparatus using an unmanned aerial vehicle comprises a communication part, an image data analysis part, a disaster situation prediction part and a disaster situation counteracting part. The communication part transmits a control signal to an unmanned aerial vehicle and receives collection information including at least one among image data, voice data, and sensing data from the unmanned aerial vehicle. The image data analysis part analyzes the image data by at least one among a mechanical learning based-analysis method, an analysis method through combination of a plurality of image data and an analysis method through relation with the sensing data. The disaster situation prediction part applies an analysis result of the image data and weather information received from the outside to a prediction model to generate a prediction result. The disaster situation counteracting part transmits disaster alarm information, which is generated based on a scenario corresponding to the analysis result and the prediction result, to an external integral alarm system or outputs the disaster alarm information. KR102161917B1 discloses an information processing system using an unmanned aerial vehicle for rescue in a mountain area and a method thereof. The information processing system comprises: an unmanned aerial vehicle that analyzes image information obtained from at least one camera while flying over an arbitrary area to determine whether the arbitrary area is a disaster area or whether there is a rescuee, and detects the position of the rescuee when it is determined that there is the rescuee; a rescuer terminal that transmits real-time position information of the rescuer; and a control server that receives image information and position information of the rescuee and the real-time position information of the rescuer, sets a rescue route of the rescuer and an evacuation route of the rescuee according to a presorted rescue scenario so that the rescuer and the rescuee can meet each other, provides the rescue route to the rescuer terminal, and provides the evacuation route to the unmanned aerial vehicle, wherein the control server updates the position information of the rescuee and the position information of the rescuer, correct the rescue route and the evacuation route based on the updated information, and the unmanned aerial vehicle guide flies along the evacuation route. [Prior Art Documents] [Patent Documents] [Patent Document 1] Japanese Patent Application Publication No. 2013-134663[Patent Document 2] Japanese Patent Application Publication No. 2021-64878[Patent Document 3] Japanese Patent Application Publication No. 2019-115012 [Summary of the Invention] [Problems to be Solved by the Invention] In order to avoid or reduce disasters during natural disasters, for example, it is necessary to appropriately deploy unmanned vehicles and rescuers in areas with high disaster risk. However, Patent Documents 1 to 3 do not describe appropriate deployment of unmanned vehicles and rescuers. The present disclosure has been made to solve the above-described problems, and has an object to provide an unmanned vehicle management system, an unmanned vehicle management method, a data storage device, a data analysis device, a disaster prediction device, an unmanned vehicle control device,