Entrance gate guidance system, entrance gate guidance method, and entrance gate guidance program

Abstract

To guide an approach path that allows passage through an entrance gate in the shortest time when guiding a vehicle to one of the approach paths leading to the plurality of entrance gates.SOLUTION: An entrance gate guidance system includes: first imaging devices 101 for capturing vehicles approaching an approach path leading to each one of entrance gates; second imaging devices 102 for capturing vehicles traveling toward a branch part to the approach path leading to each one of the entrance gates; and guidance display units. When there is a vehicle traveling toward the branch part, the system performs: inferring a required time for gate passing procedures at the entrance gate with respect to each one of the vehicles approaching the approach path by using a learning model obtained by machine learning of correlation between input data including presence / absence of a permanent entry permit, vehicle types, and presence / absence of an entry history and output data including the required time for the gate passing procedures of the vehicle at the entrance gate; inferring a total time to the passing of the rearmost vehicle through the entrance gate in each one of the approach paths; and allowing the guidance display units to guide the approach path with the shortest total time by displaying.SELECTED DRAWING: Figure 6

Description

[Technical field]

[0001] The present invention relates to an entrance / exit gate guidance system, an entrance / exit gate guidance method, and an entrance / exit gate guidance program that can be used in facilities such as power plants equipped with entrance / exit gates for multiple vehicles, and in particular, that can shorten the waiting time until gate passage procedures are completed at the entrance / exit gate. [Background technology]

[0002] Every morning, many commuter and construction-related vehicles enter the power plant. When passing through the limited number of entrance and exit gates, security guards have to check people and their baggage, which takes a considerable amount of time, resulting in long traffic jams in front of the gates.

[0003] Entrants are broadly divided into two categories: "temporary entrants" and "regular entrants." A "temporary entrant" is someone who does not have a gate pass, and based on a prior application for entry, security guards will verify the entrant's identity by checking a driver's license or other identification and will also inspect the entrant's baggage. In addition, if the entrant brings in any items, the contents of those items (tools, equipment, etc.) will also be checked. In addition, "regular entrants" are those who possess a gate pass, and their identity is verified by checking the photo on the gate pass they possess, and their baggage is also inspected. In addition, if they bring in any items, the contents of those items are checked based on their prior application. For this reason, it takes a shorter time for a "regular entrant" to go through the gate procedures than a "temporary entrant."

[0004] In addition, the gate procedures differ depending on the type of vehicle, and inspections take a considerable amount of time for large vehicles such as those used at the goods loading dock or buses carrying passengers. Furthermore, if there is a history of entry, the applicant will have experience with the gate procedures, and will often be able to enter smoothly with relatively few application errors. Summary of the Invention [Problem to be solved by the invention]

[0005] Despite these circumstances, people trying to pass through the gates have no idea about the situation of the vehicles ahead of them, so they have no choice but to rely on their own intuition when deciding which approach route to use. As a result, they tend to select an approach route (lane) that looks like it will have a smoother flow of traffic through the entrance / exit gates, which results in vehicles concentrating on certain approach routes and worsening traffic congestion.

[0006] Even if the number of vehicles waiting ahead is the same, there can be a large difference in the time it takes to pass through the entrance / exit gate, and once vehicles have lined up at the selected entrance, they cannot change entrances if there are vehicles following behind, which can result in the following vehicles lining up at a different entrance completing the gate procedures first. This creates many stressful situations for those trying to pass through the gate, and some kind of improvement is needed.

[0007] The present invention has been made in consideration of the above circumstances, and has as its main objective the provision of an entrance / exit gate guidance system, an entrance / exit gate guidance method, and an entrance / exit gate guidance program that, when guiding a vehicle to one of the entrance / exit gates (lanes) leading to a plurality of entrance / exit gates, guides the vehicle to the entrance / exit gate that will allow the vehicle to pass through the entrance / exit gate in the shortest time possible, thereby eliminating the need to rely on the driver's intuition in selecting an entrance / exit gate, thereby easing vehicular congestion, and reducing the stress of those attempting to pass through the gate. [Means for solving the problem]

[0008] In order to achieve the above object, the present invention provides an entrance / exit gate guidance system that can be used in a facility where a vehicle passes through a gate procedure at any one of a plurality of entrance / exit gates, the entrance / exit gate guidance system comprising: a first photographing device that photographs a vehicle entering an access road leading to each entrance / exit gate; A second photographing device photographs a vehicle traveling toward a branching section to an approach road leading to each entrance / exit gate; a guide display device including a display board installed at a position visible from a vehicle traveling toward the junction, a display device mounted on a vehicle traveling toward the junction, or a portable display device carried by a passenger in a vehicle traveling toward the junction; Equipped with a learning model storage unit that stores a learning model that has been machine-learned to learn a correlation between input data including the presence or absence of a permanent entry permit, vehicle type, and the presence or absence of an entry history, and output data including the time required for the vehicle to pass through the gate at the entrance / exit gate; an approaching vehicle information storage unit that determines whether or not a vehicle has a permanent entry permit, the vehicle type, and the vehicle number of each vehicle that passes through the branching section and enters one of the approach roads, based on an image captured by the first photographing device or the second photographing device, and stores the information; a gate procedure time required estimation means for estimating, when a vehicle traveling toward the junction is detected by the second photographing device, a time required for the gate procedure at the entrance / exit gate, for each vehicle waiting at each approach road recognized by the first photographing device, using the learning model from information on the presence or absence of a permanent entry permit, information on the vehicle type, and information on the presence or absence of an entry history recognized from the vehicle number, which are stored in the approaching vehicle information storage unit; a total time estimation means for estimating a total time until the rearmost vehicle at each approach road passes through the entrance / exit gate based on the required time estimated by the gate procedure required time estimation means; an approach route determination means for determining an approach route having the shortest total time estimated by said total time estimation means as a recommended approach route; a recommended approach route guidance means for displaying the recommended approach route determined by the approach route determination means on the guidance display; The present invention is characterized in that it has the following features.

[0009] Therefore, when a vehicle is detected traveling toward the branch point to the entrance road leading to the entrance gate, the gate procedure required time estimation means estimates the required time for the gate procedure at the entrance / exit gate for each vehicle waiting at each entrance road, using the learning model stored in the learning model storage unit, from the presence or absence of a permanent entry permit and vehicle type information stored in the entering vehicle information storage unit, and the presence or absence of an entry history ascertained from the vehicle number, a total time estimation means for estimating a total time required for the last vehicle on each approach road to pass through the entrance / exit gate based on the time required for each waiting vehicle to complete the gate procedure at the entrance / exit gate estimated by the gate procedure required time estimation means; The approach route determination means determines the approach route having the shortest total time estimated by the total time estimation means as the recommended approach route; The recommended approach route guiding means causes the guidance display to guide the vehicle along the recommended approach route determined by the approach route determining means, Passengers of vehicles heading towards a junction can drive their vehicles into the approach road (lane) indicated by the guide display, and wait in the approach road (lane) that will take them to the entrance / exit gate in the shortest time.

[0010] Here, in order to improve the accuracy of estimating the time required for the gate procedure, the number of passengers in the vehicle may be added as input data. In addition, when the road before the branching point is divided into a lane for regular visitors and a lane for temporary visitors, the multiple entrance / exit gates are divided into multiple entrance / exit gates for regular visitors and multiple entrance / exit gates for temporary visitors, and the gate procedure time required estimation means, the total time estimation means, the approach route determination means, and the recommended approach route guidance means are preferably provided separately for regular visitors and temporary visitors. In this way, regular entrants can line up in the access road (lane) with the shortest waiting time among the regular entrant entrance / exit gates, and first-time entrants can also line up in the access road (lane) with the shortest waiting time among the first-time entrant entrance gates. Effect of the Invention

[0011] As described above, according to the present invention, when guiding a vehicle to one of the access roads (lanes) leading to a plurality of entrance / exit gates, it is possible to guide the vehicle to the access road that allows the vehicle to pass through the entrance / exit gate in the shortest time, thereby easing vehicular congestion and reducing the stress of those trying to pass through the gate. [Brief description of the drawings]

[0012] [Figure 1] 1 is a schematic diagram showing an entrance / exit gate guide system according to the present invention; [Diagram 2] FIG. 2 is a diagram illustrating an outline of an entrance / exit gate and the branching section just before it. [Diagram 3] FIG. 1 is a block diagram illustrating an example of a machine learning device. [Figure 4] 1 is a table showing an example of learning data. [Diagram 5] 1 is a flowchart showing an example of a machine learning method performed by the machine learning device according to the present invention. [Figure 6] FIG. 1A is a block diagram showing an example of an entrance / exit gate guide device, and FIG. 1B is a table showing waiting vehicle information stored in a waiting vehicle database. [Figure 7] 4 is a flowchart showing an example of an entrance / exit gate guidance method by the entrance / exit gate guidance device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a schematic diagram showing the configuration of an entrance / exit gate guide system 1 according to the present invention, which is adopted in entrance / exit gates 100 (100I, 100II, 100III, 100IV) of a power plant or the like. As shown in the figure, this entrance / exit gate guidance system 1 includes an imaging device 2, an image analysis device 3, a machine learning device 4, an entrance / exit gate guidance device 5, and a guidance display 7.

[0014] The photographing device 2 comprises a first photographing device 101 installed at each entrance / exit gate 100 (100I, 100II, 100III, 100IV) and a second photographing device 102 installed near the branching point 111 to the access roads 110I, 110II, 110III, 110IV leading to each entrance / exit gate 100I, 100II, 100III, 100IV.

[0015] The first photographing device 101 photographs from near the entrance / exit gate to the upstream side, and can photograph the appearance and license plate of each vehicle entering (waiting) on ​​the corresponding approach road, as well as objects and people placed inside the vehicle that can be seen through the glass. The second photographing device 102 photographs from near the branch point to the upstream side, and can detect vehicles approaching the branch point and photograph the appearance and license plate of the vehicle, as well as objects and people placed inside the vehicle that can be seen through the glass.

[0016] In the learning phase, the photographing device 2 (101, 102) is connected to the machine learning device 4 via the image analysis device 3 and used to collect data for learning, and in the inference phase, the photographing device 2 (101, 102) is connected to the entrance / exit gate guidance device 5 via the image analysis device 3 and used to collect data for inference.

[0017] Here, the photographing device 2 (101, 102) used in the learning phase and the photographing device used in the inference phase may be the same or different, and are used to analyze the images taken by each to obtain various information (presence or absence of a permanent entry permit, vehicle type, vehicle number, number of passengers, etc.).

[0018] The photographing devices 2 (101, 102) may be cameras, video cameras, digital cameras, etc., and the first photographing device 101 is provided near the entrance / exit gates of each of the approaches 110I, 110II, 110III, 110IV to photograph all vehicles waiting upstream of the gates of the corresponding approaches. The first photographing device 101 may be provided for each entrance / exit gate, or multiple devices may be provided for each entrance / exit gate to photograph from different directions and heights. This first photographing device 101 photographs objects placed on the top of the dash panel visible through the glass of the vehicle, license plates, objects and people mounted on the vehicle, etc. The first photographing device 101 may be fixed to a pole provided beside each approach and photographed from a desired height, or may be fixed to a beam or the like provided above the gate.

[0019] The second photographing device 102 is provided near the junction 111 to photograph vehicles traveling from the upstream side toward the junction 111, and may be provided as a single unit or as multiple units to photograph from different directions and heights. This second photographing device 102 photographs objects placed on the upper part of the dash panel visible through the glass of the vehicle traveling toward the junction 111, the license plate, objects and people mounted on the vehicle, etc. These second photographing devices 102 may also be fixed to a pole provided at the side of the junction 111, or may be fixed to a utility pole or building if there is one nearby.

[0020] Based on the image captured by the first photographing device 101, the image analyzing device 3 determines the number of vehicles waiting on the corresponding access road, the type of vehicle (model), recognition of objects placed on the top of the vehicle's dash panel, the vehicle number displayed on the vehicle's license plate, the number of passengers in the vehicle, etc. In addition, based on the image captured by the second photographing device 102, the image analyzing device 3 recognizes the presence or absence of vehicles traveling toward the junction 111, the type of vehicles traveling toward the junction 111, objects placed on the top of the vehicle's dash panel, the vehicle number displayed on the vehicle's license plate, the number of passengers on board, etc. The image analysis device 3 is configured to output the above information obtained by image analysis to the machine learning device 4 and the entrance / exit gate guidance device 5 in association with information unique to each vehicle. That is, in the machine learning device 4 described later, the information is stored as learning data in the learning data storage unit 42 and used for machine learning. Also, in the entrance / exit gate guidance device 5 described later, the information is stored in the waiting vehicle database 51 as vehicle information of waiting vehicles and used as input data for inference.

[0021] The machine learning device 4 operates as a subject of the learning phase, and generates, by machine learning, a learning model 6 that is used to infer the time required for each vehicle to pass through the gate procedure. The entrance / exit gate guidance device 5 operates as the main body of the inference phase, and uses a trained learning model 6 generated by the machine learning device 4 to infer the time required for gate passage procedures for each vehicle waiting at each entrance road from information obtained by analyzing images from the photographing device 2, and determines the entrance road that will allow vehicles to pass through the gate in the shortest time when waiting to enter the gate passage procedures. The guidance display 7 displays the approach route determined by the entrance / exit gate guidance device 5 (the approach route that allows passing through the gate in the shortest time), and in this example is composed of a display board such as an electronic bulletin board installed in a position visible from vehicles traveling toward the junction 111.

[0022] The machine learning device 4 is configured by a general-purpose or dedicated computer. This computer may be a stationary computer or a portable computer, and may also be a client computer, a server computer, or a cloud computer. As shown in FIG. 3, the machine learning device 4 includes a learning data acquisition unit 41, a learning data storage unit 42, a machine learning unit 43, and a trained model storage unit 44.

[0023] The learning data acquisition unit 41 functions as an interface unit that acquires learning data including input data via the communication network 10. The learning data acquisition unit 41 is connected via the communication network 10 to the image analysis device 3, the worker terminal 8 used by the worker, and the like. The learning data storage unit 42 is a database that stores a plurality of sets of learning data acquired by the learning data acquisition unit 41 .

[0024] The machine learning unit 43 performs machine learning using the learning data stored in the learning data storage unit 42. The machine learning unit 43 inputs multiple sets of learning data to the learning model 6, thereby causing the learning model 6 to learn the correlation between the input data included in the learning data and the time required for the vehicle to pass through the gate procedure corresponding to that input data, and generates a learned model. There are various methods for generating a trained model, but it is preferable to employ, for example, a neural network as supervised learning by the machine learning unit 43.

[0025] The trained model storage unit 44 is a database that stores the trained learning model 6 generated by the machine learning unit 43. The training model 6 stored in the trained model storage unit 44 is provided to the entrance / exit gate guidance device 5 via any communication network or storage medium. The learning data storage unit 42 and the trained model storage unit 44 may be configured as a single storage device or as separate storage devices.

[0026] The learning data includes, as input data, at least the presence or absence of a permanent entry permit, the vehicle type, and the presence or absence of an entry history. Once a permanent entry permit is issued, it is required to be placed on the dashboard of the vehicle so that it can be recognized from the outside, and its presence or absence can be confirmed from either the image captured by the first photographing device 101 or the image captured by the second photographing device 102.

[0027] The vehicle type is external shape information of the vehicle that can be identified from either an image captured by the first image capturing device 101 or an image captured by the second image capturing device 102. The entry history is used to determine whether a vehicle has previously entered the facility, and can be determined by recognizing the vehicle number printed on the license plate of the vehicle and comparing it with a database that stores past entry records. This information can also be obtained from images captured by the first image capture device 101 or the second image capture device 102.

[0028] The learning data also includes the time required to complete the gate procedures at the entrance / exit gate as output data (teaching data) associated with the input data. Here, the correlation between the input data included in the learning data (presence or absence of a permanent entry permit, vehicle type, and presence or absence of entry history) and the time required for gate procedures will be described.

[0029] (Correlation between the presence or absence of a permanent entry permit and the time required for gate procedures) Entrants are broadly divided into two categories: "regular entrants" and "temporary entrants." A "regular entrant" is someone who has a gate pass, and for vehicles, the vehicle number is checked against the entry permit and a copy of the vehicle inspection certificate. In addition, passengers are identified by the facial photo on the gate pass they possess, and their baggage is inspected. Furthermore, if any items are brought in, the contents of those items (tools, equipment, etc.) are checked based on prior application.

[0030] A "temporary entrant" is someone who does not have a gate pass, and for vehicles, they must present a prior entry application form and have the vehicle number checked against a copy of the vehicle inspection certificate, etc. Passengers are also checked for identity using a photo ID such as a driver's license, and the purpose of entry is confirmed, and baggage is inspected. If any items are brought in, the contents of those items (tools, equipment, etc.) are checked.

[0031] When comparing the time required for gate procedures with and without a permanent entry permit (gate pass), it has been found that even if the vehicle type is the same, the time required for a "temporary entrant" to go through the gate procedures is nearly twice as long as the time required for a "permanent entrant." For this reason, the presence or absence of a permanent entry permit can be used to infer the time required for gate procedures.

[0032] (Correlation between vehicle type and time required for gate procedures) There are various vehicle classifications. For example, vehicles can be classified into "large" vehicles such as large buses, "special" vehicles with special structures such as concrete mixers, and "ordinary" vehicles such as passenger cars. For special vehicles such as concrete mixers, it is necessary to check whether they are loaded with concrete and what equipment and materials they contain, so the time required for gate entry procedures is longer than for ordinary vehicles. Furthermore, large buses carrying temporary entrants often contain people without gate passes, such as inspectors, and passengers must present identification to verify their identity, which makes the gate procedures even longer. As the time required to pass through the gate procedures varies depending on the vehicle type, identifying the vehicle type can be used to infer the time required to pass through the gate procedures.

[0033] At this time, if it is recognized that the time required for gate procedures is significantly affected by the number of passengers in the vehicle, the number of passengers in the vehicle may be further added as input data.

[0034] (Correlation between the presence or absence of a campus entry history and the time required for gate procedures) Furthermore, if someone has a campus entry history, they have experience with the gate procedures, so they can enter smoothly without making any mistakes in their application. If they do not have a campus entry history, they will not understand the gate procedures, and not only will there be incomplete documents, but the process at the time of entry will also be delayed, and the gate procedures will tend to take longer than for someone with a campus entry history. Therefore, the presence or absence of an entry history can be used to infer the time required for gate procedures.

[0035] (Machine learning methods) The machine learning device 4 forms a learning model using, for example, a neural network model. This neural network model is a publicly known model, which learns the correlation between the input data and the output data by inputting input data included in the learning data to an input layer and comparing output data output from an output layer as an inference result with output data (teacher data) included in the learning data.

[0036] First, as a preliminary preparation for having the machine learning device 4 perform machine learning, the learning data acquisition unit 41 acquires a desired number of learning data and stores the acquired multiple sets of learning data in the learning data storage unit 42. The number of learning data prepared here is appropriately set in consideration of the inference accuracy required for the learning model 6.

[0037] There are various ways to prepare training data, for example: If data on the time required for the gate procedure remains in the records of past gate procedures, the presence or absence of a permanent entry permit, the type of vehicle (normal, large, special), and the presence or absence of an entry history are confirmed for the vehicle at the time of the gate procedure, and the data is acquired as learning data together with the time required for the gate procedure and stored in the learning data storage unit 42. If learning data is to be collected in the future, the presence or absence of a permanent entry permit, the type of vehicle, and the presence or absence of an entry history are confirmed as input data for the vehicle that is about to enter, and the number of passengers in the vehicle is also confirmed as necessary. In addition, the time required for the gate procedure for the vehicle at the entrance / exit gate is measured as output data (the time from arriving at the entrance / exit gate where the gate procedure is performed to leaving is measured), and this required time is input from the operator terminal 8 and linked to the input data to create a set of learning data. In order to avoid bias toward a particular vehicle model, a large number of such sets of learning data are prepared as shown in FIG. 4, and are stored in advance in the learning data storage unit 42 as learning data for generating a learning model.

[0038] After the above-mentioned preparations, the machine learning unit 43 performs, for example, a series of processes in the flowchart shown in FIG. 5 to generate a trained model. In step S100, the machine learning unit 43 acquires one set of learning data from the multiple sets of learning data stored in the learning data storage unit 42. The learning data may be acquired in a predetermined order or randomly.

[0039] Next, in step S110, the machine learning unit 43 performs supervised machine learning. That is, the machine learning unit 43 inputs input data included in the acquired set of learning data to a prepared learning model to output an inference result, compares this inference result with output data (teacher data) included in the learning data acquired in step S100, and performs machine learning using a known neural network method. In this way, the machine learning unit 43 causes the learning model 6 to learn the correlation between the input data and the output data (the time required for the gate procedure).

[0040] Thereafter, using this learning model under learning, the above-mentioned operations are performed on the remaining prepared learning data, and learning is continued in the machine learning unit 43. At this time, the machine learning unit 43 determines whether or not to continue machine learning based on the error between the output data and the teacher data, the number of learnings, the remaining amount of unlearned learning data stored in the learning data storage unit 42, etc. (step S120).

[0041] That is, if the machine learning unit 43 determines in step S120 to continue machine learning (YES), it performs steps S100 to S110 on the learning model 6 being trained using untrained learning data. On the other hand, if the machine learning unit 43 determines in step S120 not to continue machine learning (NO), it stores the generated trained learning model 6 in the trained model storage unit 44 in step S130 and ends the machine learning.

[0042] (Entrance / Exit Gate Guidance Device) Next, we will explain the entrance / exit gate guidance device 5, which uses the trained model generated using the above-mentioned method to guide vehicles that are about to wait in line for gate procedures to the optimal entrance / exit gate (the entrance / exit gate that can be accessed in the shortest time). The entrance / exit gate guidance device 5 is configured with a general-purpose or dedicated computer. This computer may be a stationary computer or a portable computer, and may also be a client computer, a server computer, or a cloud computer. As shown in FIG. 6(a), the entrance / exit gate guidance device 5 includes a waiting vehicle database 51, an input data acquisition unit 52, an inference unit 53, a learned model storage unit 54, and an output processing unit 55.

[0043] The waiting vehicle database 51 is a database that stores information about vehicles (hereinafter, referred to as waiting vehicles) that have entered an approach road leading to one of the entrance / exit gates via the junction 111 and are waiting, but have not yet passed through the entrance / exit gate. For example, as shown in FIG. 6(b), for each waiting vehicle, information (presence or absence of a permanent entry permit, vehicle type, presence or absence of entry history, number of passengers, etc.) obtained by analyzing images taken by the first photographing device 101 and the second photographing device 102 by the image analyzing device 3 is stored together with an assigned waiting vehicle ID. The waiting vehicle ID is assigned, for example, in the order in which the vehicle passed through the junction 111, and vehicles that have passed through the entrance / exit gate are removed from the list. This waiting vehicle database 51 corresponds to an entering vehicle information storage unit.

[0044] The input data acquisition unit 50 identifies vehicles waiting at each approach road and acquires input data corresponding to each of those vehicles. It may be configured to identify vehicles lined up at the approach road corresponding to each entrance / exit gate using the first photographing device and read the input data for those vehicles from the waiting vehicle DB. This input data acquisition unit 52 is configured to operate when the second photographing device 102 detects a vehicle traveling toward the junction 111. Therefore, when a vehicle is traveling toward the junction 111, the input data acquisition unit identifies vehicles waiting on each entrance road at that time, reads various information about the waiting vehicles from the waiting vehicle DB, and prepares for inference by the inference unit 53.

[0045] The inference unit 53 inputs the input data acquired by the input data acquisition unit 52 into the learning model 6 stored in the trained model storage unit 54, and performs inference processing (inferring the time required for each waiting vehicle to pass through the entrance / exit gate). The trained learning model 6 trained by the machine learning device 4 is used for the inference processing.

[0046] The inference unit 53 may not only have the function of performing inference processing using the learning model 6, but may also include a preprocessing function of adjusting the input data acquired by the input data acquisition unit 52 to a desired format, etc., and inputting it into the learning model 6 as a preprocessing of the inference processing.

[0047] The trained model storage unit 54 is a database that stores trained learning models 6 used in the inference process of the inference unit 53. The number of learning models 6 stored in the trained model storage unit 54 is not limited to one. For example, multiple learning models 6 with different numbers of input data or different machine learning techniques may be stored and used by selecting them appropriately.

[0048] An example of a case in which the number of input data is different would be when data on the number of passengers in a vehicle, which affects the required time, is added to the input data of a learning model that estimates the time required for gate procedures.

[0049] The output processing unit 55 uses the inference results of the inference unit 53, i.e., the time required for each waiting vehicle to complete the gate procedures at the entrance and exit gates, to perform a process of determining the optimal gate (the gate that can be passed through in the shortest time) for a vehicle traveling toward the junction 111.

[0050] FIG. 7 shows a flowchart of an example of an entrance / exit gate guidance method using the entrance / exit gate guidance device 5. First, in step S200, the presence or absence of a vehicle approaching the junction 111 is determined from an image captured by the second photographing device. If there is no vehicle approaching the junction 111, no guidance process for the approach road is performed and the system enters a standby state. On the other hand, if it is determined that there is a vehicle traveling toward the junction, the input data acquisition unit 52 identifies all vehicles currently waiting on each approach road, and acquires information stored in the waiting vehicle database 51 for each of them (i.e., data associated with each waiting vehicle analyzed by the image analysis device based on images captured by the imaging device (data on whether or not the vehicle has a permanent entry permit, data on the vehicle type, and data on whether or not the vehicle has an entry history)).

[0051] Next, in step S210, the inference unit 53 inputs input data for each of all waiting vehicles into the input layer of the learning model 6, and infers the output data output from the output layer of the learning model 6, i.e., the time required for the gate procedures for the input corresponding vehicles.

[0052] Then, in step S220, the output processing unit 55 infers the total time required for the gate procedures of the vehicles waiting at each approach road (lane), that is, the total time until the rearmost vehicle at each approach road (lane) passes through the entrance / exit gate. At this time, the travel time from when the preceding vehicle completes the gate procedures to when the succeeding vehicle arrives at the entrance / exit gate may be adjusted.

[0053] Then, in step S230, the approach road (lane) with the shortest total time is selected based on the result of the inference in step S220. Then, the selected approach road (lane) with the shortest total time is designated as the recommended approach road and a guidance display ("Please proceed to XX lane gate") is displayed on the display board (guide display 7) (step S240).

[0054] Therefore, according to the above-mentioned entrance guidance system, when guiding a vehicle to one of the entrance roads (lanes) leading to multiple entrance / exit gates, it is possible to guide the vehicle to the entrance road that will allow it to pass through the entrance / exit gate in the shortest time, eliminating the need for the driver to rely on their intuition when selecting an entrance road, thereby easing vehicle congestion and reducing stress for those attempting to enter the premises.

[0055] In the above configuration, the time required for the gate procedure is inferred using the relevant information linked to the vehicle (whether or not there is a permanent entry permit, the vehicle type, and whether or not there is an entry history) as input data, but if a specific gate procedure is required for all passengers in the vehicle, the time required for the gate procedure may vary greatly depending on the number of passengers. Therefore, the number of passengers may be added as a gate procedure to improve the correlation accuracy with the required time.

[0056] In the above, a configuration has been shown in which the branching point just before each access road is treated as one lane, and traffic is distributed to each access road (lane) from there; however, if the road just before the branching point is divided, for example, into a lane for permanent entrants and a lane for temporary entrants, and the multiple vehicle entrance / exit gates are also divided into multiple entrance / exit gates for permanent entrants and multiple entrance / exit gates for temporary entrants, the above systems may be operated separately for the gates for permanent entrants and temporary entrants. In this way, regular entrants can line up in the access road (lane) with the shortest waiting time among the regular entrant entrance / exit gates, and first-time entrants can also line up in the access road (lane) with the shortest waiting time among the first-time entrant entrance gates.

[0057] Although the above describes an example in which the entrance / exit gate guide system is used at the entrance / exit gate of a power plant, the system can also be adopted at vehicle entrance / exit gates of other facilities. In addition, an example has been shown in which the guidance display 7 is configured as a display board installed in a position visible from vehicles traveling toward the junction 111, but if software that displays and provides guidance on recommended approach routes from the entrance / exit gate guidance system is installed on a display device mounted on a vehicle traveling toward the junction 111, such as a navigation system, or on a portable display device carried by a passenger in a vehicle traveling toward the junction 111, such as a smartphone, then these vehicle-mounted display devices or portable display devices may be used as guidance displays to provide guidance and provide guidance on the recommended approach route.

[0058] In the above embodiment, a case has been described in which a neural network is adopted as a specific method of machine learning by the machine learning unit 43, but the machine learning unit 43 may adopt any other machine learning method. For example, tree types such as decision trees and regression trees, ensemble learning such as bagging and boosting, neural network types (including deep learning) such as recurrent neural networks and convolutional neural networks, clustering types such as hierarchical clustering, non-hierarchical clustering, k-nearest neighbors and k-means, multivariate analyses such as principal component analysis, factor analysis and logistic regression, support vector machines, etc. may be adopted.

[0059] In addition, the above-mentioned machine learning device 4 and entrance / exit gate guidance device 5 can also be provided in the form of a program (machine learning program or entrance / exit gate guidance program) for executing each step of the above-mentioned machine learning method and entrance / exit gate guidance method. [Explanation of symbols]

[0060] 1. Entrance / exit gate guidance system 4. Machine learning equipment 5. Entrance / exit gate guide device 6 Learning Model 7 Information display 51 Waiting vehicle database 100, 100I, 100II, 100III, 100IV Entrance gates 101 First camera 102 Second camera 110I,110II,110III,110IV approach road 111 Branch

Claims

1. An entrance / exit gate guidance system that can be used in a facility where a vehicle passes through a gate procedure at any one of a plurality of entrance / exit gates, a first photographing device that photographs a vehicle entering an access road leading to each of the entrance / exit gates; A second photographing device photographs a vehicle traveling toward a branching portion to an approach road leading to each entrance / exit gate; a guidance display configured by a display board installed at a position visible from a vehicle traveling toward the junction, a display device mounted on a vehicle traveling toward the junction, or a portable display device carried by a passenger in a vehicle traveling toward the junction; Equipped with a learning model storage unit that stores a learning model that has been machine-learned to learn a correlation between input data, including the presence or absence of a permanent entry permit, vehicle type, and the presence or absence of an entry history, and output data, including the time required for the vehicle to pass through the gate procedure at the entrance / exit gate; an approaching vehicle information storage unit that determines whether or not a vehicle has a permanent entry permit, the vehicle type, and the vehicle number of each vehicle that passes through the branching section and enters one of the approach roads, based on an image captured by the first photographing device or the second photographing device, and stores the information; a gate procedure time required estimation means for estimating, when a vehicle traveling toward the junction is detected by the second photographing device, a time required for the gate procedure at the entrance / exit gate, for each vehicle waiting at each approach road recognized by the first photographing device, using the learning model from information on the presence or absence of a permanent entry permit, information on the vehicle type, and information on the presence or absence of an entry history recognized from the vehicle number, which are stored in the approaching vehicle information storage unit; a total time estimation means for estimating a total time until the rearmost vehicle at each approach road passes through the entrance / exit gate based on the required time estimated by the gate procedure required time estimation means; an approach route determination means for determining an approach route having the shortest total time estimated by said total time estimation means as a recommended approach route; a recommended approach route guidance means for causing the guidance display device to display the recommended approach route determined by the approach route determination means; An entrance / exit gate guide system comprising:

2. 2. The entrance / exit gate guidance system according to claim 1, wherein the number of passengers in a vehicle is added as the input data.

3. The area before the branch is divided into a lane for regular visitors and a lane for temporary visitors. The plurality of entrance / exit gates are divided into a plurality of regular entry entrance / exit gates and a plurality of temporary entry entrance / exit gates, The entrance / exit gate guidance system according to claim 1 or 2, characterized in that the gate procedure time estimation means, the total time estimation means, the approach route determination means, and the recommended approach route guidance means are provided separately for regular visitors and for temporary visitors.

4. An entrance / exit gate guidance method that can be used in a facility that allows a vehicle to pass through a gate procedure at any one of a plurality of entrance / exit gates, comprising: a first photographing device that photographs a vehicle entering an access road leading to each of the entrance / exit gates; A second photographing device photographs a vehicle traveling toward a branching portion to an approach road leading to each entrance / exit gate; a guidance display configured by a display board installed at a position visible from a vehicle traveling toward the junction, a display device mounted on a vehicle traveling toward the junction, or a portable display device carried by a passenger in a vehicle traveling toward the junction; Equipped with an entering vehicle information storage step of determining whether or not a vehicle has a permanent entry permit, the vehicle type, and the vehicle number of each vehicle that passes through the branching section and enters one of the entry roads, based on an image captured by the first photographing device or the second photographing device, and storing the information; a gate procedure required time estimation step of estimating the required time for gate procedures at the entrance / exit gate when the second photographing device detects a vehicle traveling toward the junction, using a learning model that has been machine-learned to determine the correlation between input data including the presence or absence of a permanent entry permit, vehicle type, and presence or absence of entry history for each vehicle waiting at each approach road recognized by the first photographing device, and output data including the required time for gate procedures at the entrance / exit gate, based on the information on the presence or absence of a permanent entry permit and vehicle type stored in the entering vehicle information storage step, and the information on the presence or absence of an entry history recognized based on the vehicle number; a total time estimation step of estimating a total time until the rearmost vehicle at each approach road passes through the entrance / exit gate based on the required time estimated in the gate procedure required time estimation step; an approach route determination step of determining the approach route with the shortest estimated total time as a recommended approach route; a recommended approach route display step of displaying the recommended approach route determined in the approach route determination step on the guidance display; An entrance / exit gate guidance method comprising the steps of:

5. An entrance / exit gate guidance program for causing a computer to execute each step of the entrance / exit gate guidance method according to claim 4.

Images