Road Closure Management System
The road closure management system automates obstacle detection and removal using UAVs, ensuring efficient and timely road closures by reducing manual checks and enabling rapid response to obstacles.
Patent Information
- Application Number
- JP2023009369
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2023-01-25
- Publication Date
- 2026-11-11
- Estimated Expiration
- 2043-01-25
AI Technical Summary
Traditional methods require manual patrols to check for obstacles at road closure locations, which are time-consuming and can delay construction and sightseeing, and existing technologies cannot detect obstacles or request their removal.
A road closure management system using an unmanned aerial vehicle equipped with a camera and determination means to photograph and determine obstacles, and take automatic actions such as generating warnings or communicating with designated contacts to remove obstacles.
Automated obstacle detection and removal processes save labor and time, ensuring smooth road closures by eliminating the need for human patrols and enabling rapid response to obstacles.
Smart Images

Figure 0007948007000001 
Figure 0007948007000002 
Figure 0007948007000003
Abstract
Description
TECHNICAL FIELD
[0001] The present invention relates to a road closure management system for managing road closure locations.
BACKGROUND ART
[0002] For example, when a road closure is required for construction work or tours at a power plant, advance measures are taken to prevent obstacles such as vehicles and materials from being placed at road closure locations including construction sites and tour routes. That is, notification sheets or emails are distributed in advance to make the relevant parties aware, and road closure locations are clearly specified on bulletin boards and site layout maps. However, since obstacles may remain on the day, patrols must be conducted to check for the presence of obstacles, and if any obstacles are present, contacts and requests must be made to have them removed.
[0003] On the other hand, a traffic control device is known that can rapidly detect the occurrence of unexpected events such as road closures (see, for example, Patent Document 1). This device performs machine learning based on training data that uses video data from past occurrences of unexpected events as input data and outputs data indicating that an unexpected event has occurred, and when road video data is input, it outputs information indicating that an unexpected event has occurred.
PRIOR ART DOCUMENTS
PATENT DOCUMENTS
[0004]
PATENT DOCUMENT 1
SUMMARY OF THE INVENTION
PROBLEMS TO BE SOLVED BY THE INVENTION
[0005] Traditionally, on the day of a road closure, people had to patrol to check for obstacles, which required a great deal of effort and time. Moreover, if there were obstacles, it was necessary to contact and request their removal, which delayed the start of construction and caused visitors to wait, hindering smooth construction and sightseeing. Furthermore, while the device described in Patent Document 1 can detect the occurrence of sudden events such as road closures, it cannot detect whether there are obstacles at the road closure location or request their removal.
[0006] Therefore, the present invention aims to provide a road closure management system that manages road closure locations in order to facilitate road closures. [Means for solving the problem]
[0007] To solve the above problems, the invention of claim 1 is a road closure management system characterized by comprising: an unmanned aerial vehicle that flies around a road closure location; a photographing means mounted on the unmanned aerial vehicle for photographing the area around the road closure location; and a determination means that determines whether or not there are obstacles that would hinder the road closure at the road closure location based on the images taken by the photographing means.
[0008] The invention of claim 2 is characterized in that, in the road closure management system described in claim 1, the determination means determines that there is an obstacle and provides a means for taking measures against the obstacle.
[0009] The invention of claim 3 is a road closure management system according to claim 2, wherein the action means, after the action has been taken, determines that there is no obstacle, and the action It is characterized by releasing the function.
[0010] The invention of claim 4 is characterized in that, in the road closure management system described in claim 2, the action means communicates obstacle information, including the location of the obstacle, to a predetermined contact when the determination means determines that the obstacle is present.
[0011] The invention of claim 5 is characterized in that, in the road closure management system described in claim 2, the system is equipped with a voice generating means mounted on the unmanned aerial vehicle and which generates sound, and the action means generates a warning sound from the voice generating means when the determination means determines that there is an obstacle.
[0012] The invention of claim 6 is characterized in that, in the road closure management system described in claim 2, the system is further equipped with a map display on which the road closure locations are marked on a map, and the action means causes the map display to show obstacle information, including the location of the obstacle, when the determination means determines that the obstacle is present.
[0013] The invention of claim 7 is characterized in that, in the road closure management system described in claim 1, the determination means uses a machine learning model for obstacle determination that has been trained on past performance data, such that when an image captured by the photography means is input, it outputs whether or not there is an obstacle. [Effects of the Invention]
[0014] According to the invention described in claim 1, an unmanned aerial vehicle flies around the area of a road closure and takes photographs, and based on these images, it is automatically determined whether or not there are any obstacles that would hinder the road closure. As a result, there is no need for human patrol, saving labor and time, and obstacle removal can be carried out appropriately and quickly based on the determination result. As a result, road closures can be carried out smoothly.
[0015] According to the invention described in claim 2, if an obstacle is detected, the countermeasure means automatically takes action against the obstacle, making it possible to take appropriate and rapid action against the obstacle. As a result, road closures can be carried out smoothly.
[0016] According to the invention described in claim 3, if the obstacle is removed after measures have been taken against it, the measures are automatically released, making it possible to complete unnecessary measures quickly and to carry out road closures smoothly.
[0017] According to the invention described in claim 4, when an obstacle is detected, obstacle information, including the location of the obstacle, is communicated to a designated contact, enabling that contact to take appropriate and prompt action regarding the obstacle. As a result, road closures can be implemented smoothly.
[0018] According to the invention described in claim 5, if an obstacle is detected, a warning sound is emitted from the sound generating means mounted on the unmanned aerial vehicle, making it possible to encourage the removal or evacuation of the obstacle. As a result, road closures can be carried out smoothly.
[0019] According to the invention described in claim 6, when an obstacle is detected, obstacle information, including the location of the obstacle, is displayed on a map display that shows the road closure area on the map. This makes it possible to alert people to the presence of an obstacle and encourage them to move out of the way or remove the obstacle. As a result, road closures can be carried out smoothly.
[0020] According to the invention described in claim 7, a machine learning model for obstacle detection is used to output whether or not there are obstacles that would hinder the road closure, making it possible to more accurately determine the presence or absence of obstacles. As a result, road closures can be carried out smoothly. [Brief explanation of the drawing]
[0021] [Figure 1] This is a schematic diagram showing a road closure management system according to an embodiment of the present invention. [Figure 2] Figure 1 is a schematic block diagram showing the management computer for the road closure management system. [Figure 3] This figure shows a first example of an image captured by the camera of the road closure management system shown in Figure 1. [Figure 4] This is a diagram showing a second example of an image captured by the camera of the traffic restriction management system in Figure 1. [Figure 5] This is a diagram showing a third example of an image captured by the camera of the traffic restriction management system in Figure 1. [Figure 6] This is a diagram showing a fourth example of an image captured by the camera of the traffic restriction management system in Figure 1. [Figure 7] This is a functional block diagram showing a schematic configuration of the obstacle determination learning model of the management computer in Figure 2. [Figure 8] This is a diagram showing a first image example of the on-site arrangement board of the traffic restriction management system in Figure 1. [Figure 9] This is a diagram showing a second image example of the on-site arrangement board of the traffic restriction management system in Figure 1. [Figure 10] This is a flowchart showing the operation of the traffic restriction management system in Figure 1. [Figure 11] This is a flowchart showing the continuation of Figure 10. [Mode for Carrying Out the Invention]
[0022] Hereinafter, the present invention will be described based on the illustrated embodiments.
[0023] Figure 1 is a schematic configuration diagram showing a traffic restriction management system 1 according to an embodiment of the present invention. This traffic restriction management system 1 is a system for managing traffic restriction points PS, and in this embodiment, the case where traffic restriction points PS are closed to traffic within the premises of a power plant P will be mainly described. In addition, there may be a plurality of traffic restriction points PS at the same time.
[0024] The traffic restriction management system 1 mainly includes a drone 2, a management computer 3, a staff terminal 4, and an on-site arrangement board (map display) 5, and the management computer 3 is capable of communicating with the drone 2, the staff terminal 4, and the on-site arrangement board 5 respectively. Note that reference numeral M in Figure 1 denotes a construction staff member.
[0025] Drone 2 is an unmanned aerial vehicle that flies around the road closure area PS, and in this embodiment, it automatically flies along a predetermined flight route. That is, it is equipped with GPS (Global Positioning System) and, for example, flies autonomously along a flight route set by the monitoring computer 3 (described later) and stored in Drone 2, while confirming its own position with GPS. In this case, it may also be equipped with a function to avoid obstacles to flight depending on the flight environment.
[0026] Furthermore, Drone 2 is equipped with an automatic hovering function that allows it to hover at a predetermined hovering position, and is designed to hover near the road closure point (PS) to perform the photography described later. Multiple such Drones 2 may be provided and flown simultaneously along different flight routes. Alternatively, instead of automatic flight, the drones may be remotely controlled by a person.
[0027] This drone 2 is equipped with a camera (shooting means) 21 that takes pictures of the area around the road closure point PS, and a speaker (sound generation means) 22 that generates sound. That is, camera 2 System 1 is activated when drone 2 hovers near the road closure point PS, and takes pictures of the area around the road closure point PS. The images taken here can be a video of a predetermined duration (e.g., several tens of seconds) or still images, but if still images are taken, they are taken continuously for a predetermined duration at predetermined time intervals (e.g., every few seconds). The captured images, along with the location information (latitude and longitude) of the shooting location, i.e., the road closure point PS, are then transmitted in real time to the monitoring computer 3.
[0028] Speaker 22 outputs a predetermined message sound in response to an audio output command from the monitoring computer 3. That is, it stores multiple message sounds in the memory installed in the drone 2 and outputs the message corresponding to the audio output command. For example, upon receiving audio output command 1, it outputs the message, "This area is closed to traffic from XX to XX. Please leave." Upon receiving audio output command 2, it outputs the message, "This area is closed to traffic from XX to XX. Please remove any objects." It also stops outputting audio upon receiving a stop command.
[0029] The personnel terminal 4 is a communication terminal used by various personnel, and in this embodiment, the communication terminal is a smartphone (multifunctional mobile phone). Furthermore, the personnel include not only personnel from the departments that want to close off the road, such as construction personnel and tour personnel, but also personnel from all departments, and as will be described later, the contact information (email address, etc.) of these personnel's personnel terminal 4 is pre-stored in the monitoring computer 3. When the personnel terminal 4 receives the obstacle information for communication, which will be described later, from the monitoring computer 3, it is displayed on the display of the personnel terminal 4.
[0030] The site layout board 5 is a display / electronic bulletin board that displays road closures (PS) and other information on the site layout map (map), and is installed in a location within office C where it can be seen by many people. Alternatively, the displays on the personal computers of each person in charge at the power plant P may be used as map displays to show similar information and images. In response to display commands from the monitoring computer 3, the site layout board 5 displays road closures (PS) and obstacles on the site layout map (a diagram showing the arrangement of facilities and roads within the power plant P), as shown in Figures 8 and 9 described later.
[0031] The monitoring computer 3 is located in office C and is a computer that determines whether or not there are obstacles that would hinder road closures. As shown in Figure 2, it mainly comprises an input unit 31, a display unit 32, a communication unit 33, a storage unit 34, management tasks (determination means, action means) 35, a learning task 36, and a central processing unit 37 that controls these.
[0032] The input unit 31 is an interface for inputting various information and commands, specifically for inputting the location and date of road closures, the person in charge, and the activation command for management task 35. The display unit 32 is a display for displaying various data and information, specifically for displaying images taken by drone 2 and the contents of the site layout board 5. The communication unit 33 is an interface for communicating with the outside world via the internet or telephone network, specifically for receiving images taken by drone 2, sending voice output commands to drone 2, and sending display commands to the site layout board 5. In addition, personnel can access the management computer 3 from terminals 4 and other devices to input road closure application information, as described later.
[0033] The memory unit 34 mainly comprises a road closure database (road closure information storage means) 341, an obstacle detection learning model 342, and an obstacle detection performance database 343. Here, the road closure database 341 will be described, and the obstacle detection learning model 342 and the obstacle detection performance database 343 will be described later.
[0034] The road closure database 341 is a database that stores information about road closures for each road closure location PS identification information. It stores the location of the road closure (latitude and longitude of the road closure location PS), the date and time of the road closure and its duration, the name of the person in charge of the department that wants to close the road, and their contact information, which are entered in the input unit 31, etc. It also stores the names and contact information of the persons in charge of road closures for all departments.
[0035] Management task 35 is a task program that determines whether or not there are obstacles that would prevent road closure and takes measures against such obstacles. It is activated when it receives captured images from drone 2 or when an activation command is input from input unit 31 specifying captured images.
[0036] First, based on the images captured by the camera 21 of drone 2, it is determined whether or not there are any obstacles that could obstruct (or potentially obstruct) the road closure at the road closure point PS. That is, the captured images are analyzed to determine whether or not there are any obstacles, and if obstacles are found, the type of obstacle is determined. In this case, the determination may be made by comparing the captured images with images (reference images) of the area around the road closure point PS that have been previously captured and stored, in a state where there are no obstacles. Here, an obstacle is anything that could hinder or obstruct the road closure, and includes not only objects and vehicles, but also people.
[0037] For example, in the first example, in the case of the image shown in Figure 3, it is determined that a vehicle is stopped as an obstacle at the road closure point (PS). In the second example, in the case of the image shown in Figure 4, it is determined that there is no obstacle at the road closure point (PS). In the third example, in the case of the image shown in Figure 5, it is determined that a motorcycle is traveling as an obstacle at the road closure point (PS). In the fourth example, in the case of the image shown in Figure 6, it is determined that a vehicle is traveling as an obstacle at the road closure point (PS), and that there is also a drum.
[0038] In this type of determination, when a captured image is input, a machine learning model 342 for obstacle detection is used, which is trained on past performance data to output the presence or absence of obstacles, as well as the type and state of the obstacles. This obstacle detection model 342 is created by the learning task 36.
[0039] In other words, the learning task 36 creates an obstacle detection learning model 342 using a known machine learning algorithm such as a neural network, based on past performance data recorded and stored in the obstacle detection performance database 343. This obstacle detection performance database 343 is a database in which performance data is recorded and stored, including images taken around road closure points PS as input information, and the presence, type, and condition of obstacles actually confirmed at road closure points PS. The past performance data includes not only actual images taken and data on the presence, type, and condition of obstacles actually confirmed, but also data created through pre-training, etc.
[0040] As shown in Figure 7, this learning task 36 uses machine learning and deep learning with a neural network to create a neural network based on actual data recorded in the obstacle detection database 343. For example, it uses images taken around the road closure point PS as the input layer, the presence or absence of obstacles at the road closure point PS, and if present, their type and state as the output layer, and the analysis processing from the input layer to the output layer as the hidden layer. Then, the learning task 36 uses the actual data of the obstacle detection learning model 342 as training data to learn various parameters in the hidden layer. In other words, the learning task 36 learns various parameters in the hidden layer so that the presence or absence of obstacles, and the type and state of obstacles, are output appropriately based on images taken around the road closure point PS.
[0041] Next, if an obstacle is detected, measures and countermeasures are taken against the obstacle. While measures may be taken immediately upon detection of an obstruction, in this embodiment, measures are taken according to the type of obstruction only after a predetermined period of time (e.g., 10 seconds) has been detected. This allows measures to be taken only when there is a high probability that the obstruction will hinder the road closure, thereby reducing unnecessary measures and post-processing. Furthermore, if the type of obstruction is a movable object such as a vehicle or person, and it is in motion, no measures may be taken, or measures may be taken only if it has been in the road closure area (PS) for a predetermined period of time.
[0042] As the first measure, if an obstacle is detected, the obstacle information is sent to the designated contact person. Here, the obstacle information includes the location and type of the obstacle, and the location of the obstacle indicates which road closure point PS it is. That is, based on the shooting location received from the drone 2 along with the captured image and the location of the road closure stored in the road closure database 341, the road closure point PS is determined and its identification information is displayed.
[0043] Furthermore, the designated contact is, firstly, the contact information (person in charge terminal 4) of the person in charge of the department that requested the road closure, which is stored in the road closure database 341. Obstacle information, such as what kind of obstacle is present at which road closure location PS and that the person should be careful, is then linked to this contact. In addition, if it is possible to determine which department the obstacle belongs to (for example, if it can be determined from a nearby building or from signs marked on the obstacle), the contact information of the person in charge of that department is also contacted. In this case, obstacle information, such as what kind of obstacle is present at which road closure location PS and that it should be removed, is linked. Here, if there is only one road closure location PS, it is not necessary to specify which road closure location PS it is.
[0044] As a second measure, if an obstacle is detected, a warning voice is emitted from speaker 22. Specifically, if the obstacle is determined to be a person or vehicle, voice output command 1 is sent to drone 2, and as described above, speaker 22 emits the message, "This area is closed from XX to XX. Please leave." Also, if the obstacle is determined to be an object, voice output command 2 is sent to drone 2, and as described above, speaker 22 emits the message, "This area is closed from XX to XX. Please remove the object."
[0045] As a third measure, if an obstacle is detected, the obstacle information for display is shown on the site layout board 5. That is, a display command including the location and type of the obstacle is transmitted to the site layout board 5, causing the obstacle information to be displayed on the site layout board 5. Here, the location of the obstacle is the road closure point PS where the obstacle has been detected, and by transmitting the identification information of the road closure point PS determined as described above to the site layout board 5, the site layout board 5 can identify which road closure point PS it is. Upon receiving this display command, as shown in Figure 9, a mark indicating the type of obstacle (obstacle mark) is displayed at the corresponding road closure point PS in the site layout diagram on the site layout board 5. Here, the circle with a car in Figure 9 indicates that the obstacle is a vehicle, and the circle with a person in Figure 9 indicates that the obstacle is a person.
[0046] Such measures are canceled if it is determined that there are no obstacles after the measures have been taken. In other words, if the obstacles are gone after the measures have been taken, the measures are canceled. Specifically, information that the obstacles have been removed is sent to the contact person in charge of the department that requested the road closure, and a stop command is sent to speaker 22 (drone 2) to stop the audio output. In addition, a display command is sent to the site layout board 5 to erase the obstacle mark. Here, this display command includes a request to erase the obstacle mark and the location of the obstacle (identification information of the road closure location PS), and in response, the corresponding obstacle mark is erased on the site layout board 5.
[0047] Next, the operation, function, and method of managing road closures using the road closure management system 1, as described above, will be explained based on Figures 10 and 11.
[0048] First, the person in charge of the department that wants to close the road accesses the management computer 3 from the person in charge's terminal 4 or similar device and enters the road closure application information. This road closure application information includes the location of the road closure, the date and time of the road closure and the duration of the closure, the name of the person in charge of the department that wants to close the road, and their contact information. Alternatively, the road closure application information may be entered from the input section 31 of the management computer 3. Once this road closure application information is entered, identification information is generated for each road closure location PS and stored in the road closure database 341, and a display command is sent to the site layout board 5, so that the road closure locations PS, the date and time of the road closure, and the person in charge are displayed on the site layout map as shown in Figure 8. Here, in Figures 8 and 9, the date and time of the road closure and the person in charge are omitted, and two road closure locations PS1 and PS2 are displayed.
[0049] Next, the flight route including the road closure point PS is set by the management computer 3 and stored in the drone 2. A predetermined time before the road closure date and time (for example, one hour before), the drone 2 flies to the first road closure point PS (step S1). Subsequently, the drone 2 takes pictures of the area around the road closure point PS with the camera 21 and transmits the captured images to the management computer 3, at which point the management task 35 is activated to determine whether or not there are any obstacles.
[0050] As a result, if there are no obstacles (if "N" is selected in step S2), the process proceeds to step S6; if there are obstacles (if "Y" is selected in step S2), measures are taken regarding the obstacles. Specifically, as described above, a warning sound is generated from speaker 22 (step S3), obstacle information for posting is displayed on the site layout board 5 (step S4), and obstacle information for communication is sent via email to the designated person in charge (step S5). In this case, as described above, these measures are taken if the obstacles remain for a predetermined period of time.
[0051] Next, if there is another road closure PS after a predetermined time has elapsed (for example, a few minutes) (if the answer is "Y" in step S6), drone 2 flies to the next road closure PS (step S7), and steps S2 to S5 are repeated in the same manner. However, if multiple drones 2 are to be flown to different road closure PS locations, this repetitive process does not need to be performed.
[0052] On the other hand, if the flight to all road closure locations (PS) and the assessment are completed (if "N" is obtained in step S6), and there are no obstacles at any of the road closure locations (if "Y" is obtained in step S8), a link email is sent to the contact person in charge of the department that requested the road closure, informing them that there are no obstacles (step S9). After that, drone 2 returns to its designated location (step S16).
[0053] In response to this, if there is an obstacle at any of the road closure points PS (if the answer is "N" in step S8), drone 2 will fly to the road closure point PS where the obstacle was found (step S10). If there is only one road closure point PS, the drone will remain in place. Then, in the same manner as above, it will determine whether there is still an obstacle, and if there is an obstacle (if the answer is "Y" in step S11), it will continue to issue a warning from speaker 22 (step S12).
[0054] On the other hand, if the obstacle is removed (if "N" is selected in step S11), the audio output from speaker 22 is stopped (step S13), the obstacle mark is erased from the site layout board 5 (step S14), and a notification is sent via email to the designated person in charge that the obstacle has been removed (step S15). After that, drone 2 returns to the designated location (step S16).
[0055] As described above, according to this road closure management system 1 and road closure management method, the drone 2 flies around the road closure location PS and takes photographs, and based on those images, the road closure is determined by the obstruction. The system automatically determines whether or not there are obstacles. This eliminates the need for human patrols, reducing labor and time, and allows for the appropriate and rapid removal of obstacles based on the determination results. Furthermore, since the system uses a machine learning-based obstacle detection model 342 to output whether or not there are obstacles that would hinder the road closure, it is possible to determine the presence or absence of obstacles more accurately. As a result, road closures can be carried out smoothly, meaning that construction work, sightseeing, etc., can be carried out smoothly.
[0056] Furthermore, if an obstacle is detected, the management computer 3 automatically takes action against the obstacle, enabling appropriate and prompt action against the obstacle. In other words, when an obstacle is detected, obstacle information, including the location of the obstacle, is communicated to a designated person, allowing that person to take appropriate and prompt action against the obstacle.
[0057] Furthermore, if an obstacle is detected, a warning voice will be emitted from the speaker 22 mounted on the drone 2, prompting the removal or evacuation of the obstacle. In addition, if an obstacle is detected, obstacle information, including the location of the obstacle, will be displayed on the site layout board 5, which is marked with the road closure area PS on the map, allowing for warnings about the presence of obstacles and prompting the removal or evacuation of the obstacles. In this way, road closures can be carried out smoothly.
[0058] Furthermore, once the obstacle is removed after measures have been taken against it, the measures are automatically released, allowing unnecessary measures to be completed quickly and road closures to be carried out smoothly.
[0059] Although embodiments of this invention have been described in detail above, the specific configuration is not limited to these embodiments, and any design changes, etc., that do not depart from the gist of this invention are also included. For example, the above embodiment described a case where road closures are made within the premises of a power plant P, but it can also be applied to road closures in other locations, such as factories or universities. Furthermore, although the case in which the management task 35 serves as both a determination means and a corrective action means has been described, these may be made into separate tasks or provided on a separate computer.
[0060] Furthermore, while the above embodiment described and illustrated the case where the road closure point PS is relatively small and short in length, it is of course applicable to cases where the road closure point PS is relatively large and long in length, in which case the flight monitoring range and shooting range of drone 2 will be wider and longer. [Explanation of Symbols]
[0061] 1. Road Closure Management System 2. Drones (unmanned aerial vehicles) 21. Camera (Method of taking pictures) 22. Speaker (means of generating sound) 3. Management Computer 35 Management tasks (determination means, action means) 341 Road Closure Database 342 Obstacle detection learning model 343 Obstacle Detection Performance Database 4. Terminal of the person in charge 5. Site layout board (map display) P Power Plant PS Road Closure Locations
Claims
1. An unmanned aircraft flying around the area of the road closure, A photographic means mounted on the aforementioned unmanned aerial vehicle for photographing the area around the road closure, A determination means that determines whether or not there are any obstacles that would hinder the road closure at the road closure location, based on the image captured by the aforementioned photographic means, A road closure management system characterized by having the following features.
2. If the determination means determines that the obstacle exists, the system includes means for taking measures against the obstacle. The road closure management system according to feature 1.
3. The aforementioned action means releases the action if the determination means determines that there is no obstacle after the action has been taken. The road closure management system according to feature 2.
4. The aforementioned means, when the determination means determines that the obstacle exists, communicates the obstacle information, including the location of the obstacle, to a predetermined contact. The road closure management system according to feature 2.
5. The aforementioned unmanned aerial vehicle is equipped with a sound generating means that generates sound, The aforementioned action means generates a warning sound from the sound generating means when the determination means determines that the obstacle is present. The road closure management system according to feature 2.
6. The map display shows the aforementioned road closure locations marked on the map, The aforementioned means, when the determination means determines that the obstacle exists, causes the obstacle information, including the location of the obstacle, to be displayed on the map display. The road closure management system according to feature 2.
7. The determination means uses a machine learning model for obstacle detection, which is based on past performance data, so that when an image captured by the shooting means is input, it outputs whether or not the obstacle is present. The road closure management system according to claim 1, characterized in that it is a road closure management system.
Citation Information
Patent Citations
Supervisory camera system
JP1999284988A
Patrol inspection support system and patrol inspection support control program
JP2019009919A
Vehicle security device
JP2020093618A
Traffic control device, and traffic control method
JP2022080556A
Information processing system, information processing device, server device, program, or method
JP2022091787A
