Traffic monitoring device, traffic monitoring system, traffic monitoring method and program

The traffic monitoring system uses optical fiber sensing and machine learning to provide a wide-area, real-time view of traffic conditions, accurately detecting and predicting traffic events for effective traffic management.

JP7578140B2Active Publication Date: 2024-11-06NEC CORP

Patent Information

Application Number
JP2022539825
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2020-07-28
Publication Date
2024-11-06
Estimated Expiration
2040-07-28

AI Technical Summary

Technical Problem

Existing traffic monitoring systems struggle to provide a wide-area, real-time view of traffic situations, including accurate detection of traffic conditions such as average speed and the occurrence of abnormal events like traffic jams and road obstacles.

Method used

A traffic monitoring system utilizing optical fiber sensing technology to detect vehicle positions and changes in vehicle positions over time, combined with machine learning models to generate historical and predictive traffic information, enabling real-time detection and prediction of traffic conditions and events.

Benefits of technology

Enables real-time, wide-area monitoring of traffic conditions and accurate prediction of traffic events, allowing for effective traffic control measures to mitigate congestion and obstacles.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007578140000001
    Figure 0007578140000001
  • Figure 0007578140000002
    Figure 0007578140000002
  • Figure 0007578140000003
    Figure 0007578140000003
Patent Text Reader

Abstract

The purpose of the present invention is to holistically ascertain traffic conditions over a broad range in real time. A traffic monitoring device (102) comprises: a position acquisition unit (106) that acquires position information indicating vehicle positions on a roadway; and a history generation unit (107) that generates history information indicating changes in the vehicle positions over time on the basis of the position information. The position acquisition unit (106) acquires the position information that is obtained, for instance, on the basis of optical fiber sensing utilizing optical fibers laid in the roadway. The traffic monitoring device (102) may further comprise an abnormality detection unit (107) that detects abnormal events on the roadway on the basis of: event-specific history patterns that indicate patterns in the changes in the vehicle positions over time, which correspond to individual types of abnormal events on the roadway; and the changes in the vehicle positions over time.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a traffic monitoring device, a traffic monitoring system, a traffic monitoring method, and a program. [Background technology]

[0002] Patent Document 1 discloses a technology for grasping the distribution of moving objects in a target section on a route from point A to point B. The route guidance device described in Patent Document 1 grasps the distribution of moving objects by calculating the number of moving objects present in each of small sections obtained by dividing the target section into several hundred meters at a certain cross section based on the detection results of vehicle detectors.

[0003] The vehicle detector described in Patent Document 1 is described as being installed in several units over small sections primarily for road traffic purposes to detect the speed of automobiles traveling on the road and the volume of passing traffic, and as being widely used in the field of road traffic control.

[0004] However, vehicle detectors (traffic counters) only collect information on vehicles passing through a certain point. but It is not always possible to accurately obtain information such as the average vehicle speed in the section that includes that point. Therefore, Patent Document 2 discloses a technology that supports appropriate use of probe information.

[0005] The traffic information management system described in Patent Document 2 calculates a first traffic condition for each of multiple sections of a road using information acquired from vehicle detectors installed on the road, and calculates a second traffic condition using probe information acquired from vehicles traveling on the road.The traffic information management system then displays on a display unit the first traffic condition and the second traffic condition for a section where the difference between the first traffic condition and the second traffic condition is equal to or greater than a predetermined threshold. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Publication No. 11-203594 [Patent Document 2] JP 2018-190117 A Summary of the Invention [Problem to be solved by the invention]

[0007] Generally, the traffic conditions in one section of a road can affect other sections, so for road control purposes, it is desirable to be able to grasp the traffic conditions of moving objects on the road from a wide-area bird's-eye view. Also, since the traffic conditions on a road change from moment to moment, for example, when a traffic accident occurs, congestion occurs in a relatively short time, it is desirable to be able to grasp the traffic conditions in real time.

[0008] However, the distribution of moving objects in a target section grasped by the technology described in Patent Document 1 is the number of moving objects present in each of the small sections into which the target section is divided, and it is difficult to grasp the traffic situation of moving objects on roads in a wide-area, bird's-eye view, and in real time.

[0009] In addition, even if the traffic information management system described in Patent Document 2 can inform the user of a section where the difference between the first traffic condition and the second traffic condition is equal to or greater than a threshold, it is difficult to obtain the traffic conditions in that section, for example, the average vehicle speed. Therefore, even with the technology described in Patent Document 2, it is difficult to grasp the traffic conditions of moving objects on a road in a wide area, in a bird's-eye view, and in real time.

[0010] The present invention has been made in consideration of the above-mentioned circumstances, and aims to provide a traffic monitoring device, a traffic monitoring system, a traffic monitoring method, and a program that are capable of grasping traffic conditions over a wide area in real time from a bird's-eye view. [Means for solving the problem]

[0011] In order to achieve the above object, a traffic monitoring device according to a first aspect of the present invention comprises: a position acquisition means for acquiring position information indicating a vehicle position, which is the position of the vehicle on a road; a history generating means for generating first history information indicating a change in the vehicle position over time based on the position information; an abnormality detection means for detecting an abnormal event on the road based on an event-specific history pattern indicating a pattern of a change over time of a vehicle position corresponding to a type of the abnormal event on the road and the change over time of the vehicle position indicated by the first history information; a prediction means for generating prediction information indicating a change over time in a predicted vehicle position on the road based on the first history information generated by the history generating means; 、 an abnormality prediction means for predicting a type of an abnormal event that will occur on the road based on the event-specific history pattern and the change over time indicated by the prediction information; Equipped with The event-specific history pattern includes at least one of a congestion pattern corresponding to a congestion as the abnormal event and an obstacle pattern corresponding to an obstacle on a road as the abnormal event, The congestion pattern indicates that the speed of the vehicle is slower than a reference speed and the inter-vehicle distance is shorter than a reference distance, The obstacle pattern indicates that the vehicle will slow down, change lanes, and then return to its original lane.

[0012] A traffic monitoring system according to a second aspect of the present invention comprises: The traffic monitoring device; an optical fiber laid on the road and having one end terminated with a termination treatment for suppressing reflection of an optical signal; a sensing device that inputs an optical signal into the optical fiber and observes a change in optical interference intensity, which is the intensity of light produced by interference between backscattered lights occurring in response to the input of the optical signal; The position acquisition means acquires the position information on the road based on the amount of change in the optical interference intensity observed by the sensing device.

[0013] A traffic monitoring method according to a third aspect of the present invention comprises: The computer obtaining position information indicating a vehicle position, the vehicle position being a position of the vehicle on a road; generating first history information indicating a change in the vehicle position over time based on the position information; Detecting an abnormal event on a road based on an event-specific history pattern indicating a pattern of change over time of a vehicle position corresponding to a type of the abnormal event on the road and the change over time of the vehicle position indicated by the first history information; generating prediction information indicating a change over time in a predicted vehicle position on the road based on the generated first history information; 、 predicting a type of abnormal event occurring on the road based on the event-specific history pattern and the change over time indicated by the prediction information; Including, The event-specific history pattern includes at least one of a congestion pattern corresponding to a congestion as the abnormal event and an obstacle pattern corresponding to an obstacle on a road as the abnormal event, The congestion pattern indicates that the speed of the vehicle is slower than a reference speed and the inter-vehicle distance is shorter than a reference distance, The obstacle pattern indicates that the vehicle will slow down, change lanes, and then return to its original lane.

[0014] A program according to a fourth aspect of the present invention comprises: obtaining position information indicating a vehicle position, the vehicle position being a position of the vehicle on a road; generating first history information indicating a change in the vehicle position over time based on the position information; Detecting an abnormal event on a road based on an event-specific history pattern indicating a pattern of change over time of a vehicle position corresponding to a type of the abnormal event on the road and the change over time of the vehicle position indicated by the first history information; generating prediction information indicating a change over time in a predicted vehicle position on the road based on the generated first history information; 、 predicting a type of abnormal event occurring on the road based on the event-specific history pattern and the change over time indicated by the prediction information; Run the command, The event-specific history pattern includes at least one of a congestion pattern corresponding to a congestion as the abnormal event and an obstacle pattern corresponding to an obstacle on a road as the abnormal event, The congestion pattern indicates that the speed of the vehicle is slower than a reference speed and the inter-vehicle distance is shorter than a reference distance, The obstacle pattern indicates that the vehicle will slow down, change lanes, and then return to its original lane. Effect of the Invention

[0015] According to the present invention, it is possible to grasp the traffic conditions over a wide area in real time from a bird's-eye view. [Brief description of the drawings]

[0016] [Figure 1] 1 is a diagram showing a configuration of a traffic monitoring system according to an embodiment of the present invention. [Diagram 2] 1 is a diagram showing a functional configuration of a traffic monitoring device according to an embodiment; [Diagram 3] 1 is a diagram showing a physical configuration of a traffic monitoring device according to an embodiment; [Figure 4A] 10 is a flowchart illustrating an example of a traffic monitoring process according to an embodiment. [Figure 4B] 10 is a flowchart illustrating an example of a traffic monitoring process according to an embodiment. [Diagram 5] FIG. 13 is a diagram for explaining a method of determining traffic parameters from changes in vehicle position over time. [Figure 6] FIG. 2 is a diagram showing an example of a congestion pattern, which is a pattern of changes in vehicle positions over time when congestion occurs; [Figure 7] FIG. 2 is a diagram showing an example of a stop pattern, which is a pattern of change over time in the vehicle position when the vehicle stops. [Figure 8] 1A shows an example of an obstacle pattern, which is a pattern of change in vehicle position over time, and FIG. 1B shows an example of a path of travel of a vehicle traveling on a road. [Figure 9] 13 is a flowchart illustrating an example of a traffic control process. [Figure 10] FIG. 13 is a diagram illustrating a functional configuration of a history generating unit according to one modified example. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. The same elements are designated by the same reference numerals throughout the drawings. Note that the same reference numerals are designated to similar components in all the drawings, and the description thereof will be omitted as appropriate.

[0018] <Configuration of the traffic monitoring system according to the present embodiment> As shown in Fig. 1, a traffic monitoring system 100 according to this embodiment is a system for monitoring the traffic of vehicles 101 traveling on a road R by using optical fiber sensing technology. The road R is typically a highway, but may be any other general road. The vehicles 101 are automobiles, motorcycles, buses, trucks, etc.

[0019] The traffic monitoring system 100 includes an optical fiber OF, a sensing device 102, a camera 103, and a traffic monitoring device 104.

[0020] The optical fiber OF is an optical fiber cable laid on the road R. The optical fiber OF is, for example, one core of a multi-core optical fiber cable for communication that is generally laid on the shoulders or median strips of expressways, and has one end connected to the sensing device 102 and the other end terminated to suppress reflection of optical signals. Note that a plurality of fiber cables of the multi-core optical fiber cable may be adopted as the optical fiber OF for optical fiber sensing.

[0021] The sensing device 102 inputs an optical signal into the optical fiber OF, and observes the amount of change in optical interference intensity, which is the intensity of light resulting from interference between backscattered lights occurring in response to the input of the optical signal.

[0022] More specifically, for example, an optical signal having a pulse waveform is input to the sensing device 102 from one end of the optical fiber OF. As a result, weak return light called backscattered light is generated from all positions of the optical fiber OF. The sensing device 102 observes the backscattered light.

[0023] When an environmental change occurs around the optical fiber OF, the structure and characteristic parameters of the quartz glass that constitutes the optical fiber change accordingly, and the signal quality of the backscattered light from the location where the change occurred also changes.

[0024] When an optical signal with high coherence is input and vibrations caused by a vehicle 101 traveling on a road R are transmitted to the optical fiber OF, the phase state of the backscattered light changes. This change in the phase state of the backscattered light is observed as a change in light intensity due to interference with other backscattered light received at the same time. That is, the sensing device 102 inputs an optical signal to the optical fiber OF and observes the amount of change in light interference intensity caused by the application of vibration.

[0025] The location of vibration occurrence is calculated from the round-trip time from inputting the optical signal to observing the backscattered light, and the propagation speed of the optical signal. The optical signal is repeatedly input at a constant frequency so that the backscattered light from the other end of the optical fiber OF (i.e., the farthest end as viewed from the sensing device 102) is not mixed with the next optical signal to be input. This makes it possible to accurately observe the transition of environmental changes such as vibrations occurring around the optical fiber OF in real time.

[0026] In this way, optical fiber sensing is a technology that uses optical fiber OF as a sensing medium to detect the location of vibrations, etc. This technology can use general optical fiber OF, which is a transmission medium for communication data, as a linear passive sensor, making it possible to grasp traffic conditions over a wide area in real time without installing new sensors.

[0027] The camera 103 is an example of an imaging means for capturing an image of the road R in order to grasp the traffic conditions on the road R. The camera 103 is, for example, a CCTV (closed-circuit television) camera installed at intervals on the road R.

[0028] The traffic monitoring device 104 repeatedly acquires observation information including the location of vibration occurrence on road R from the sensing device 102. The location of vibration occurrence corresponds to the position (vehicle position) of the vehicle 101 on road R. Therefore, the observation information includes position information indicating the vehicle position.

[0029] The traffic monitoring device 104 obtains changes in the vehicle position over time from the repeatedly acquired position information. Based on the changes in the vehicle position over time, the traffic monitoring device 104 calculates traffic parameters indicating the traffic conditions of the road, detects abnormal events that have occurred on the road R, predicts future changes in the vehicle position over time and the occurrence of abnormal events, and provides traffic control support including presenting traffic control measures.

[0030] Examples of abnormal events include traffic jams, vehicle stoppages, and road obstacles. An obstacle on the road is the presence of an object on the road that impedes the passage of the vehicle 101 (for example, a vehicle stopped due to an accident, a fallen load, or an object blown by a strong wind).

[0031] Examples of traffic parameters include the speed of vehicles 101 traveling on road R, the traffic density of road R, traffic volume, which is the number of vehicles 101 traveling per unit time at a predetermined point on road R, and occupancy, which indicates the proportion of road R occupied by vehicles 101.

[0032] The change in vehicle position over time is shown in a graph, as shown in FIG. 1, with the vehicle position on the horizontal axis and time on the vertical axis.

[0033] A solid line HP1 shown in FIG. 1 indicates the change over time of the vehicle position of a vehicle 101 traveling on an upbound lane, and a dotted line HP2 indicates the change over time of the vehicle position of a vehicle 101 traveling on a downbound lane (not shown).

[0034] Moreover, the solid line HP1 and the dotted line HP2 within the area surrounded by the dashed dotted line P are examples of changes in the vehicle position over time from the past to the present. Information indicating the changes in the vehicle position over time from the past to the present, that is, the history of the vehicle position, is referred to as history information.

[0035] The solid line HP1 and dotted line HP2 within the area surrounded by the dotted line F are examples of changes in the predicted future vehicle position over time, and information indicating changes in the predicted future vehicle position over time is referred to as predicted information.

[0036] In this embodiment, an example will be described in which history information of a vehicle 101 heading in one direction (a vehicle 101 traveling in an upbound lane) is generated. Then, an example will be described in which, based on the generated history information, traffic parameters are acquired, abnormal events are detected, changes in vehicle positions over time are predicted, occurrence of abnormal events is predicted, and traffic control measures are presented.

[0037] Similarly, for the vehicle 101 traveling in the opposite direction, the traffic parameter Get It is possible to detect abnormal events, predict changes in vehicle positions over time, predict the occurrence of abnormal events, and present traffic control measures.

[0038] <Functional configuration of the traffic monitoring device according to the present embodiment> The traffic monitoring device 104 of this embodiment includes an input unit 105, a position acquisition unit 106, a history generation unit 107, a first learning model memory unit 108, a traffic condition acquisition unit 109, an anomaly detection unit 110, an event-specific history pattern memory unit 111, a prediction unit 112, a second learning model memory unit 113, an anomaly prediction unit 114, a control unit 115, a display unit 116, a display control unit 117, an imaging control unit 118, and a communication unit 119.

[0039] The input unit 105 is a keyboard, a mouse, a touch panel, or the like for a user to input instructions and the like.

[0040] The position acquisition unit 106 acquires, from the sensing device 102, position information on the road R obtained based on optical fiber sensing using an optical fiber OF laid on the road R.

[0041] In detail, the position acquisition unit 106 repeatedly acquires, from the sensing device 102, position information on the road R that is obtained based on the amount of change in the optical interference intensity observed by the sensing device 102.

[0042] In this embodiment, an example will be described in which the location information is obtained based on optical fiber sensing, but the location information may be obtained based on information obtained from various sensors installed on the road R, such as a CCTV camera and a traffic volume meter (coil). Furthermore, the location information may be obtained based on probe information of ETC (Electronic Toll Collection System) 2.0.

[0043] The history generating unit 107 generates history information indicating changes over time in the vehicle position from the past to the present, based on the position information acquired by the position acquiring unit 106.

[0044] In detail, the position information is included in the observation information acquired based on the optical signal input at a certain frequency as described above. Therefore, the position information repeatedly acquired by the position acquisition unit 106 indicates discrete vehicle positions, although at relatively short time intervals.

[0045] The history generating unit 107 receives discrete vehicle positions as input, and generates history information that continuously indicates changes in the vehicle position over time, as shown by line HP1 in FIG. 1, in accordance with the first learning model.

[0046] The history generating unit 107 may generate history information indicating an approximate curve, an approximate straight line, or a combination thereof, by obtaining an approximate curve, an approximate straight line, or a combination thereof for discrete vehicle positions.

[0047] The first learning model storage unit 108 is a storage unit for storing in advance the first learning model to which the history generation unit 107 refers.

[0048] The first learning model is a trained learning model that has undergone machine learning to generate history information using location information included in observation information from the sensing device 102 as an input. Supervised learning may be employed for training the first learning model. In this case, the teacher data may be created based on probe information of the vehicle 101 that has actually traveled, an on-board camera, or the like.

[0049] The traffic condition acquisition unit 109 obtains the values ​​of the traffic parameters based on the history information.

[0050] The abnormality detection unit 110 detects an abnormal event on the road R based on the change over time of the vehicle position indicated by the history information generated by the history generation unit 107 and the event-specific history patterns.

[0051] The event-specific history pattern indicates a pattern of change over time in the vehicle position corresponding to the type of abnormal event on road R.

[0052] The event-specific history pattern storage unit 111 is a storage unit for storing in advance event-specific history pattern information indicating an event-specific history pattern, and is referred to by the abnormality detection unit 110 .

[0053] When an abnormal event is detected by the abnormality detection unit 110, the prediction unit 112 generates prediction information about the road R based on the type of the detected abnormal event and the history information generated by the history generation unit 107. For example, the prediction unit 112 receives the history information generated by the history generation unit 107 and the type of the abnormal event detected by the abnormality detection unit 110 as input, and outputs prediction information about the road based on the second learning model.

[0054] The second learning model storage unit 113 is a storage unit for storing in advance the second learning model to which the prediction unit 112 refers.

[0055] The second learning model is a trained learning model that has undergone machine learning to generate prediction information using the first historical information for training and the type of abnormal event detected as input. Supervised learning may be employed for training the second learning model. In this case, data indicating historical information and the type of abnormal event that has actually occurred in the past may be employed as the teacher data.

[0056] Regardless of whether an abnormal event is detected by the abnormality detection unit 110, the prediction unit 112 may output prediction information indicating a change over time in a predicted vehicle position for road R based on the history information generated by the history generation unit 107. In this case, the prediction unit 112 may use the history information generated by the history generation unit 107 as an input and generate prediction information for road R based on the second learning model.

[0057] In this case, the second learning model is a trained learning model that has undergone machine learning to generate prediction information using the first historical information for training as an input. Supervised learning may be employed for the training of the second learning model. In this case, data including actually occurring past historical information may be employed as the teaching data.

[0058] The abnormality prediction unit 114 predicts the type of abnormal event that will occur on the road R, based on the change over time indicated by the prediction information generated by the prediction unit 112 and the event-specific history pattern. The event-specific history pattern is indicated by the event-specific history pattern information stored in the event-specific history pattern storage unit 111.

[0059] That is, while the above-mentioned abnormality detection unit 110 employs history information as information indicating a change in the vehicle position over time, the abnormality prediction unit 114 employs prediction information as information indicating a change in the vehicle position over time. Except for this point, the abnormality prediction unit 114 and the abnormality detection unit 110 may have the same functions.

[0060] When the occurrence of an abnormal event is predicted by the abnormality prediction unit 114, the control unit 115 generates traffic control information based on the type of the abnormal event predicted to occur. The traffic control information includes traffic control measures for mitigating the abnormal event predicted to occur.

[0061] Examples of traffic control measures include at least one of the following: closing an interchange, taking an alternative route to avoid a location where an abnormal event is predicted to occur, changing road tolls, and dispatching emergency personnel. The dispatching plan for emergency personnel is a plan for dispatching personnel to deal with an abnormal event, and includes the number of personnel, the number of emergency vehicles, arrangements for emergency vehicles to head to the site, and the expected arrival time.

[0062] For example, the control unit 115 receives as input the type of abnormal event predicted to occur by the abnormality prediction unit 114 and historical information, and generates traffic control information including traffic control measures for mitigating the abnormal event based on a learning model.

[0063] The learning model may be a machine-learned model that uses the type of abnormal event and history information in the past as input to generate traffic control information including traffic control measures for mitigating the abnormal event, and is stored in a storage unit (not shown) in advance. Supervised learning may be used for the learning, and the teacher data may be, for example, traffic control information that uses the type of abnormal event and history information in the past as input and includes traffic control measures that were effective in mitigating the abnormal event.

[0064] The control unit 115 may store in advance control information that associates the type of abnormal event with a traffic control means, without relying on a learning model. In this case, the control unit 115 may acquire the traffic control means associated with the type of abnormal event predicted to occur from the control information, and generate traffic control information including the acquired traffic control means.

[0065] Moreover, the control unit 115 holds performance information in advance.

[0066] The performance information includes traffic control measures that have been adopted in the past for each type of abnormal event and an evaluation of the traffic control measures. The evaluation is expressed, for example, by an evaluation value indicating the degree to which the abnormal event was mitigated by the traffic control measures, or by whether the abnormal event was mitigated by the traffic control measures.

[0067] Then, based on the performance information, the control unit 115 generates evaluation information for the traffic control measure generated by the control unit 115. The evaluation information indicates an evaluation value indicating a predicted degree to which the traffic control measure generated by the control unit 115 will mitigate an abnormal event predicted to occur by the abnormality prediction unit 114, or whether or not the traffic control measure is predicted to mitigate the abnormal event.

[0068] Furthermore, the control unit 115 accepts a traffic control means selected by the user from among the traffic control means indicated by the traffic control information generated by the control unit 115, and generates decision information indicating the selected traffic control means.

[0069] The display unit 116 displays a screen. The display control unit 117 causes the display unit 116 to display a screen.

[0070] The display control unit 117 displays, for example, history information, values ​​of traffic parameters, prediction information, road maps showing the locations of current or future abnormal events, presentation of traffic control measures, traffic control information generated by the control unit 115, evaluation information, decision information, and the like on the screen of the display unit 116.

[0071] The imaging control unit 118 controls the camera 103 that captures the image of the road R.

[0072] For example, when an abnormal event is detected by the abnormality detection unit 110, the imaging control unit 118 causes the camera 103 to take an enlarged image of the location where the abnormal event occurred. For example, when the abnormality prediction unit 114 predicts the occurrence of an abnormal event, the imaging control unit 118 causes the camera 103 to take an enlarged image of the location where the abnormal event is predicted to occur.

[0073] For example, by displaying an image captured by the camera 103 on the screen of the display unit 116 under the control of the display control unit 117, the current situation of a place where an abnormal event has occurred or is predicted to occur can be grasped.

[0074] The communication unit 119 communicates with a driving control device 120 mounted on the vehicle 101. The driving control device 120 is preferably a device that controls the automatic driving of the vehicle 101.

[0075] For example, when an abnormal event is detected by the abnormality detection unit 110, the communication unit 119 transmits information indicating the location of the abnormal event to the operation control device 120. For example, when the abnormality prediction unit 114 predicts the occurrence of an abnormal event, the communication unit 119 transmits information indicating the location of the predicted abnormal event to the operation control device 120. For example, the communication unit 119 transmits decision information generated by the control unit 115 to the operation control device 120.

[0076] <Related to this embodiment traffic monitoring Physical configuration of the device From now on, the traffic monitoring device 10 according to this embodiment will be described. 4 An example of the physical configuration will be described with reference to the drawings.

[0077] Traffic monitoring device 10 4 Physically, as shown in FIG. 3, the system includes a bus 1010, a processor 1020, a memory 1030, a storage device 1040, a network interface 1050, and a user interface 1060.

[0078] The bus 1010 is a data transmission path for transmitting and receiving data among the processor 1020, the memory 1030, the storage device 1040, the network interface 1050, and the user interface 1060. However, the method of connecting the processor 1020 and the like to each other is not limited to a bus connection.

[0079] The processor 1020 is implemented by a central processing unit (CPU) or a graphics processing unit (GPU).

[0080] The memory 1030 is a main storage device realized by a RAM (Random Access Memory) or the like.

[0081] The storage device 1040 is an auxiliary storage device realized by a hard disk drive (HDD), a solid state drive (SSD), a memory card, a read only memory (ROM), or the like.

[0082] The storage device 1040 is 4 The memory units (first learning model memory unit 108, event-specific history pattern memory unit 111, second learning model memory unit 113) and the function of retaining information are realized.

[0083] The storage device 1040 also includes: traffic monitoring equipment 10 4 The memory 1030 stores program modules for implementing each of the functional units (position acquisition unit 106, history generation unit 107, traffic condition acquisition unit 109, abnormality detection unit 110, prediction unit 112, abnormality prediction unit 114, control unit 115, display control unit 117, imaging control unit 118, and communication unit 119). The processor 1020 loads each of these program modules onto the memory 1030 and executes them to implement each of the functional units corresponding to the program modules.

[0084] The network interface 1050 is an interface for connecting the traffic monitoring device 102 to a wired or wireless network, or a network configured by combining these. 4 The device communicates with the sensing device 102, the camera 103, and the driving control device 120 by being connected to a network via a network interface 1050.

[0085] The user interface 1070 is an interface through which information is input by the user and an interface that presents information to the user, and includes, for example, a mouse, keyboard, touch sensor, etc. as the input unit 105, and a liquid crystal display, etc. as the display unit 116.

[0086] In this way, the traffic monitoring device 10 4 The functions can be realized by the physical components working together to execute a software program. Therefore, the present invention may be realized as a software program (hereinafter, simply referred to as a "program") or as a non-transitory storage medium on which the program is recorded.

[0087] <Traffic monitoring process according to this embodiment> Hereinafter, the traffic monitoring process according to an embodiment of the present invention will be described with reference to the drawings.

[0088] 4A and 4B are flowcharts showing an example of a traffic monitoring process according to the present embodiment.

[0089] The traffic monitoring process is a process for monitoring traffic on the road R, and is performed by referring to position information repeatedly acquired, for example, at regular time intervals, from the sensing device 102. The traffic monitoring process is started, for example, in response to a user instruction from the input unit 103.

[0090] As shown in FIG. 4A, the position acquisition unit 106 acquires the vehicle position of the vehicle 101 traveling on the road R by acquiring position information from the sensing device 102 (step S101).

[0091] The history generating unit 107 obtains the change over time in the vehicle position from the past to the present, that is, the history of the vehicle position, based on the vehicle position acquired in step S101 (step S102).

[0092] Here, since the vehicle position can be acquired at a relatively short interval in step S101, the history generation unit 107 may be performed when the vehicle position for a predetermined period of time longer than the acquisition period of the vehicle position is acquired in step S101.

[0093] The traffic condition acquisition unit 109 acquires traffic parameters of the road R, such as the speed (vehicle speed) of the vehicle 101, traffic density, traffic volume, and occupancy rate, based on the history of the vehicle position acquired in step S102 (step S103).

[0094] As can be seen by referring to the example of the change in vehicle position over time in Fig. 5, the vehicle speed [km / h] is calculated by dividing the amount of change in the vehicle position (for example, in the figure, the distance from vehicle position P1 to vehicle position P2) by the time required for the change. In the figure, the vehicle speed is expressed by the slope of the change in the vehicle position over time, and among lines HPa, HPb, and HPc which show the change in the vehicle position over time, the vehicle 101 corresponding to line HPa is the fastest, and the vehicle 101 corresponding to line HPc is the slowest.

[0095] The traffic density [vehicles / km] is the number of vehicles 101 per unit section at a certain moment. In the figure, the traffic density [vehicles / km] is expressed by the number of lines HPa, HPb, and HPc that indicate the change in vehicle position over time and intersect with a horizontal line corresponding to a certain time within the area corresponding to the unit section.

[0096] The traffic volume [vehicles / h] is the number of vehicles 101 passing a certain point in a unit time. In the figure, the traffic volume [vehicles / h] is expressed by the number of lines HPa, HPb, and HPc that indicate the change in vehicle positions over time and intersect with a vertical line corresponding to a certain point within an area corresponding to a unit time.

[0097] The occupancy rate is the proportion of distance of a target section at a certain moment that is occupied by the vehicle 101 (spatial occupancy), or the proportion of time of a target time at a certain point that is occupied by the vehicle 101 (time occupancy). These can be calculated based on the traffic density, traffic volume, etc., with the length of the vehicle 101 (vehicle length) as the average length, for example.

[0098] The display control unit 117 causes the display unit 116 to display the values ​​of the transportation parameters calculated in step S103 (step S104).

[0099] The abnormality detection unit 110 detects an abnormal event occurring on the road R, based on the vehicle position history obtained in step S102 and the event-specific history pattern indicated by the event-specific history pattern information stored in the event-specific history pattern storage unit 111. Then, the abnormality detection unit 110 determines whether or not an abnormal event has occurred on the road R, based on the detection result (step S105).

[0100] In detail, for example, the event-specific history pattern includes at least one of a congestion pattern corresponding to congestion, a stop pattern corresponding to a stop of the vehicle, and an obstacle pattern corresponding to an obstacle on the road.

[0101] 6 shows an example of a congestion pattern, which is a pattern of changes over time in vehicle positions when congestion occurs. When congestion occurs, vehicles 101 travel at a speed slower than a reference speed that is appropriately determined, and the distance between the vehicles becomes shorter than a reference distance that is appropriately determined. Therefore, as shown in the figure, the congestion pattern is characterized by the slope of the line indicating the change in vehicle position of each vehicle 101 indicating a low speed, and the distance between vehicle positions P4 to P9 of each vehicle 101 at time T1 being short.

[0102] FIG. 7 shows an example of a stop pattern, which is a pattern of change over time in the vehicle position when the vehicle 101 stops. When the vehicle 101 stops, the vehicle position of the vehicle 101 does not change over time, so the stop pattern includes a feature that the vehicle position remains constant over time. In the example shown in the figure, the vehicle 101 gradually decelerates and stops at vehicle position P10 at time T2. Note that a stop that continues for longer than a predetermined time may also be included in the stop pattern.

[0103] FIG. 8(a) shows an example of an obstacle pattern, which is a pattern of change over time in the vehicle position when an obstacle occurs on the road.

[0104] For example, as shown in Fig. 8(b), assume that an on-road obstacle F is at position P11 in the driving lane on road R. In this case, a vehicle 101 traveling in the driving lane decelerates from position 101a in front of the on-road obstacle F in order to move to an adjacent lane to avoid the on-road obstacle F. Then, after moving to position 101b in the adjacent lane and passing the on-road obstacle F, the vehicle 101 returns to the driving lane and accelerates as shown in Fig. 8(b).

[0105] 8(a) includes a feature that, when a road obstacle F is in a driving lane on a road R, the multiple vehicles 101 travel at a low speed within a predetermined range from the road obstacle F and then accelerate. Note that, in order to determine whether the deceleration on the road R is for avoiding the road obstacle F, the obstacle pattern may include deceleration to a speed slower than a predetermined speed.

[0106] 6 to 8 are merely examples of event-specific history patterns corresponding to the types of abnormal events, i.e., congestion, stop of the vehicle 101, and obstacle on the road. Event-specific history pattern information including a plurality of event-specific history patterns for each type of abnormal event may be stored in advance in the event-specific history pattern storage unit 111. Then, the abnormality detection unit 110 finds the similarity between each event-specific history pattern and the history of the vehicle position by collating them (for example, pattern matching).

[0107] When the similarity is greater than a predetermined threshold, the abnormality detection unit 110 detects an abnormal event occurring on the current road R, and therefore determines that an abnormal event has occurred on road R (step S105; Yes). At this time, the type of abnormal event occurring on road R is also identified from the type of abnormal event corresponding to the event-specific history pattern with a high similarity.

[0108] On the other hand, if the similarity is equal to or less than the threshold, the abnormality detection unit 110 does not detect any abnormal event occurring on the current road R, and therefore determines that no abnormal event has occurred on the road R (step S105; No).

[0109] As described above, various patterns of event-specific history patterns are assumed for each event. Therefore, the abnormality detection unit 110 may detect an abnormal event based on a learning model using a change in the vehicle position over time as an input, and generate information indicating the type of the abnormal event when an abnormal event is detected. In this case, the learning model may be a trained learning model that has undergone machine learning to generate information indicating the type of abnormal event according to the similarity between the change in the vehicle position over time and the event-specific history pattern using an input of the change in the vehicle position. This learning may be supervised learning, and the learning model may be stored in advance in a storage unit that replaces the event-specific history pattern storage unit 111.

[0110] In addition, the abnormality detection unit 110 may detect an abnormal event and identify the type of the detected abnormal event based on whether or not one or more of the characteristics of the event-specific history pattern for each type of abnormal event is included in the change in the vehicle position over time.

[0111] If it is determined that no abnormal event has occurred (step S105; No), the position acquisition unit 106 performs the process of step S101.

[0112] If it is determined that an abnormal event has occurred (step S105; Yes), the display control unit 117 causes the display unit 116 to display a road map showing the location where the abnormal event has occurred (step S106). In addition, the imaging control unit 118 causes the camera 103 to take an enlarged image of the location where the abnormal event has occurred (step S107).

[0113] Reference is now made to FIG. 4B. The prediction unit 112 predicts a future change over time in the vehicle position on the road R based on the history of the vehicle position obtained in step S102 and the type of the abnormal event detected in step S105 (step S108).

[0114] The predicted change over time in the future vehicle position here is, for example, the change over time in the vehicle position indicated by a solid line HP1 in the area surrounded by a two-dot chain line F in FIG.

[0115] Please refer to Figure 4B. The abnormality prediction unit 114 predicts an abnormal event occurring on the road R, based on the change over time of the future vehicle position predicted in step 108 and the event-specific history pattern indicated by the event-specific history pattern information stored in the event-specific history pattern storage unit 111. Then, based on the prediction result, the abnormality prediction unit 114 determines whether or not an abnormal event is predicted to occur on the road R (step S109).

[0116] The method of predicting the occurrence of an abnormal event in step S109 is generally similar to that in step S105, except that the vehicle position history is replaced with the future change in vehicle position over time.

[0117] Then, in step S109, similarly to step S105, if the similarity is greater than a predetermined threshold, the abnormality prediction unit 114 determines that an abnormal event is predicted to occur on road R (step S109; Yes). At this time, the type of abnormal event predicted to occur on road R is also identified from the type of abnormal event corresponding to the event-specific history pattern with high similarity.

[0118] Moreover, when the similarity is equal to or less than the threshold value, the abnormality prediction unit 114 determines that the occurrence of an abnormal event on the road R is not predicted (step S109; No).

[0119] If it is determined that the occurrence of an abnormal event is not predicted (step S109; No), referring again to FIG. 4A, the position acquisition unit 106 performs the process of step S101.

[0120] Returning to Fig. 4B, when it is determined that the occurrence of an abnormal event is predicted (step S109; Yes), the display control unit 117 displays a road map indicating a predicted occurrence location, which is a location where the abnormal event is predicted to occur, on the screen of the display unit 116 (step S110). In addition, the communication unit 119 transmits predicted occurrence location information indicating the predicted occurrence location of the abnormal event to the driving control device 120 (step S111).

[0121] The control unit 115 performs traffic control processing (step S112) for traffic control to mitigate abnormal events that are currently occurring and abnormal events that are predicted to occur in the future.

[0122] FIG. 9 is a flowchart showing the details of the traffic control process (step S112).

[0123] The control unit 115 generates traffic control information based on the type of the abnormal event determined to be predicted to occur in step S109 (step S201). The display control unit 117 causes the display unit 116 to display the traffic control information generated in step S201 (step S202).

[0124] For example, assume that the abnormal event predicted to occur is a traffic jam. In this case, the traffic control measures included in the traffic control information include closing an interchange before the predicted location of the traffic jam, temporarily raising the toll from the toll gate, presenting an alternative route that avoids the predicted location of the traffic jam, etc. The alternative route may be searched for based on map information and the predicted location of the traffic jam.

[0125] Furthermore, when congestion due to a traffic accident is predicted, the traffic control measures included in the traffic control information may include a dispatch plan for emergency personnel to handle the traffic accident.

[0126] The control unit 115 extracts traffic control means for which there is a choice from among the traffic control means included in the traffic control information generated in step S201 (step S 203 ).

[0127] An example of a traffic control measure with options is one in which multiple alternative routes are presented as a traffic control measure.

[0128] The control unit 115 performs step S 203 For the traffic control measures extracted in step S204, evaluation information is generated for each option (step S204). The display control unit 117 causes the display unit 116 to display the evaluation information generated in step S204 (step S205).

[0129] The control unit 115 judges whether or not a user instruction regarding the adoption or rejection of the traffic control measures displayed in step S202 has been accepted (step S206). If it is judged that a user instruction has not been accepted (step S206; No), the control unit 115 continues the process of step S205 until the instruction is accepted.

[0130] When it is determined that the user's instruction has been accepted (step S206 ;Yes ), the control unit 115 outputs the traffic control means adopted by the user as decision information (step S207).

[0131] That is, the user can select the traffic control means to be actually adopted by referring to the traffic control means displayed in step S202 and the evaluation displayed in step S205. Then, by specifying the traffic control means to be adopted from the input unit 105, the control unit 115 generates decision information including the traffic control means to be adopted.

[0132] It is also possible to skip step S203 and perform steps S204 to S205 for all or a predetermined portion of the traffic control measures included in the traffic control information generated in step S201.

[0133] The display control unit 117 causes the display unit 116 to display the decision information output in step S207 (step S208). Faith The unit 119 transmits the determination information to the automobile driving control device 120 (step S209).

[0134] Here, step S 2 At 09 Decision information The automobile equipped with the driving control device 120 to which the information is transmitted is an example of the vehicle 101, and is preferably an autonomous automobile. This allows the driving control device 120 to refer to the decision information, determine a desired driving route, and control the traveling of the automobile.

[0135] When the traffic control process (step S112) is completed, the position acquisition unit 106 performs the process of step S101, as shown in Fig. 4B and A. The traffic monitoring process is continuously executed until, for example, a user's instruction to end the process is received from the input unit 105.

[0136] So far, one embodiment of the present invention has been described.

[0137] According to this embodiment, history information indicating changes in the vehicle position over time is generated based on position information indicating the vehicle position on the road.

[0138] The vehicle position on the road can be ascertained from sensors installed over a wide area, etc. Furthermore, the history information can be used to ascertain the continuous changes in the vehicle position over time, both in terms of location and time.

[0139] This makes it possible to grasp the traffic conditions over a wide area in real time.

[0140] According to this embodiment, an abnormal event occurring on road R is detected based on the event-specific history pattern and the change in vehicle position over time.

[0141] By utilizing the change in vehicle position over time, it is possible to grasp in real time any abnormal events occurring on road R. In addition, since the change in vehicle position over time has a characteristic pattern for each type of abnormal event, it is possible to accurately detect abnormal events occurring on road R by detecting abnormal events using the event-specific history patterns.

[0142] Therefore, it becomes possible to accurately grasp abnormal phenomena occurring on road R in real time.

[0143] According to this embodiment, a road map showing the location where the abnormal event occurred is displayed on the screen. This allows the user to easily recognize the location where the event occurred. In addition, the camera 103 is caused to take an enlarged image of the location where the event occurred. This allows the user to accurately recognize the current state of the location where the event occurred in real time.

[0144] According to this embodiment, the change in the vehicle position on the road R over time is predicted based on history information indicating the change in the vehicle position over time.

[0145] A vehicle 101 that changes its position in a similar manner under similar road conditions is likely to change its position in a similar manner in the future, so it is possible to accurately predict changes in the vehicle's position over time on road R based on historical information.

[0146] This will enable accurate prediction of traffic conditions over a wide area.

[0147] According to this embodiment, an abnormal event occurring on road R is predicted based on the predicted change in vehicle position on road R over time, the event-specific history patterns, and the change in vehicle position over time.

[0148] As described above, it is possible to accurately predict the change over time of the predicted vehicle position on road R from a wide-area and bird's-eye view. In addition, the change over time of the vehicle position has a characteristic pattern for each type of abnormal event, just like in the case of detecting an abnormal event that is currently occurring, so that it is possible to accurately predict an abnormal event that will occur on road R by using the event-specific history patterns.

[0149] Therefore, it becomes possible to predict abnormal events occurring on road R with high accuracy.

[0150] According to this embodiment, a road map showing the predicted location of an abnormal event is displayed on the screen. This allows the user to easily recognize the predicted location of the event. In addition, the predicted location of the event is transmitted to the driving control device 120. This allows a driving plan to be created taking into account the predicted location of the event, enabling efficient travel by avoiding congested areas on the road R, for example.

[0151] According to this embodiment, traffic control information including traffic control measures for mitigating an abnormal event is generated based on the type of the abnormal event predicted to occur. This makes it possible to mitigate the abnormal event, thereby enabling efficient travel using road R.

[0152] According to this embodiment, evaluation information for traffic control measures is generated based on performance information of traffic control measures that have been adopted in the past in response to abnormal events. This allows the user to select a traffic control measure for mitigating a current or future abnormal event by referring to the evaluation based on past performance. This increases the possibility of selecting an effective traffic control measure, enabling more efficient travel using road R.

[0153] According to this embodiment, the decision information indicating the traffic control means selected by the user is displayed on the screen. This allows the user to easily recognize the decision information. In addition, the decision information is transmitted to the driving control device 120. This allows the driving plan to be made taking the decision information into consideration, thereby enabling efficient travel by, for example, avoiding congested areas on the road R.

[0154] <Modification> In the embodiment, an example of generating history information of a vehicle 101 heading in one direction has been described. However, as described with reference to Fig. 1, the traveling direction of the vehicle 101 can be easily determined from the positive or negative inclination in the history information, or the direction approaching or receding from a reference position. In this modification, an example of a history generating unit that obtains history information of vehicles 101 heading in different directions (for example, vehicles 101 traveling on both the upbound and downbound lanes) will be described.

[0155] As shown in FIG. 10, the history generating unit 207 according to the modified example functionally includes a position change acquiring unit 207a, a traveling direction determining unit 207b, a first generating unit 207c, and a second generating unit 207d.

[0156] The position change acquisition unit 207a determines the change in the vehicle position over time based on the repeatedly acquired position information.

[0157] The traveling direction determination unit 207b determines whether the traveling direction of the vehicle 101 is a first direction or a second direction opposite to the first direction, based on the change over time obtained by the position change acquisition unit 207a.

[0158] The first generating unit 207c generates first history information indicating a history of the vehicle position of the vehicle 101 traveling in a first direction.

[0159] The second generating unit 207d generates second history information indicating a history of the vehicle position regarding the vehicle 101 traveling in the second direction.

[0160] Such a history generating unit 207 may be employed in place of the history generating unit 107 in the traffic monitoring device 104 according to the embodiment.

[0161] According to this modification, it is possible to grasp the vehicles 101 traveling in each of the oncoming lanes, and therefore it is possible to grasp the traffic conditions over a wide area in real time from a bird's-eye view.

[0162] Furthermore, for example, the traveling direction of the vehicle 101 traveling in the wrong direction is the same as the traveling direction of the vehicle 101 traveling in the opposite lane, so the positive and negative slopes of the lines HP1 and HP2 included in the history information shown in FIG. 1, for example, will be the same.

[0163] Therefore, in order to distinguish between a vehicle 101 traveling in the opposite lane and a vehicle 101 traveling in the wrong direction, the signal strength of the return light observed in optical fiber sensing may be used.

[0164] For example, if vibration is applied in a lane close to the optical fiber OF, the signal strength becomes large, and if vibration is applied in a lane far from the optical fiber OF, the signal strength becomes small. Therefore, by using the signal strength by optical fiber sensing, it is possible to distinguish which lane each vehicle 101 is traveling in, grasp the situation of the opposite lane, and detect vehicles traveling in the wrong direction.

[0165] The method for determining which lane each vehicle 101 is traveling in is not limited to the above-mentioned method using the signal strength by optical fiber sensing, and other methods may be adopted.

[0166] Although the embodiment and the modified examples of the present invention have been described above, the present invention is not limited to these. For example, the present invention also includes a form in which some or all of the embodiment and the modified examples described so far are appropriately combined, and a form in which the form is appropriately modified.

[0167] One or all of the above-described embodiments can be described as, but are not limited to, the following supplementary notes.

[0168] 1. A position acquisition means for acquiring position information indicating a vehicle position on a road; a history generating means for generating first history information indicating a change over time in the vehicle position based on the position information. Traffic monitoring equipment.

[0169] 2. The position acquisition means acquires the position information on the road based on optical fiber sensing using optical fibers laid on the road. 1. A traffic monitoring device as described above in paragraph 1.

[0170] 3. The vehicle further includes an abnormality detection means for detecting an abnormal event on the road based on an event-specific history pattern indicating a pattern of change over time of a vehicle position corresponding to a type of abnormal event on the road and the change over time of the vehicle position indicated by the first history information. 3. A traffic monitoring device as described in 1 or 2 above.

[0171] 4. The event-specific history pattern includes at least one of a traffic jam pattern corresponding to a traffic jam as the abnormal event, a stop pattern corresponding to a vehicle stop as the abnormal event, and an obstacle pattern corresponding to an obstacle on the road as the abnormal event. 3. A traffic monitoring device as described above in claim 3.

[0172] 5. The display control means further includes a display control means for displaying a screen; The display control means displays, when an abnormal event is detected by the abnormality detection means, a road map showing the location where the abnormal event has occurred on the screen. 5. A traffic monitoring device according to claim 3 or 4.

[0173] 6. The vehicle further includes an imaging control means for controlling an imaging means for imaging the road. The imaging control means, when an abnormal event is detected by the abnormality detection means, causes the imaging means to take an enlarged image of the location where the abnormal event occurred. 6. A traffic monitoring device according to any one of claims 3 to 5.

[0174] 7. The vehicle navigation system further includes a prediction unit that generates prediction information indicating a change over time in a predicted vehicle position on the road based on the first history information generated by the history generation unit. A traffic monitoring device according to any one of 1 to 6 above.

[0175] 8. The prediction means generates prediction information about the road using the first history information generated by the history generation means as an input, based on a trained learning model that has undergone machine learning to generate the prediction information using the first history information for training as an input. 8. A traffic monitoring device as described above in 7.

[0176] 9. The method further includes a prediction unit that generates prediction information indicating a change over time in a predicted vehicle position on the road based on a type of the abnormal event and the first history information generated by the history generation unit when an abnormal event is detected by the abnormality detection unit. 7. A traffic monitoring device according to any one of claims 3 to 6.

[0177] 10. The prediction means generates prediction information about the road using the first history information generated by the history generation means and the type of abnormal event detected by the abnormality detection means as inputs, based on a trained learning model that has undergone machine learning to generate the prediction information using the first history information for training and the type of abnormal event as inputs. 10. A traffic monitoring device as described above in claim 9.

[0178] 11. The vehicle further includes an abnormality prediction means for predicting the type of abnormal event that will occur on the road based on the event-specific history pattern and the change over time indicated by the prediction information. 11. A traffic monitoring device as described in 9 or 10 above.

[0179] 12. The display control means further includes a display control means for displaying a screen; The display control means displays, when the abnormality prediction means predicts the occurrence of an abnormal event, a road map showing the location of the predicted abnormal event on the screen. 12. A traffic monitoring device as described above in 11.

[0180] 13. The vehicle further includes a communication means for communicating with a driving control device mounted on the vehicle; When the abnormality prediction means predicts the occurrence of an abnormal event, the communication means transmits information indicating a location of the predicted abnormal event to the operation control device. 13. A traffic monitoring device as described above in claim 12.

[0181] 14. The vehicle further includes a control means for generating traffic control information including a traffic control means for mitigating an abnormal event based on the type of the abnormal event predicted to occur when the abnormal event is predicted to occur by the abnormality prediction means. 14. A traffic monitoring device according to any one of claims 11 to 13.

[0182] 15. The control means further generates an evaluation value indicating a prediction of a degree to which an abnormal event predicted to occur by the abnormal prediction means will be mitigated by the traffic control means generated by the control means, or the evaluation information indicating whether or not the abnormal event is predicted to be mitigated by the traffic control means, based on performance information including the traffic control means adopted for the past abnormal events and an evaluation value indicating a degree to which the abnormal event was mitigated by the traffic control means, or whether or not the abnormal event was mitigated by the traffic control means. 15. A traffic monitoring device as described above in claim 14.

[0183] 16. The control means further receives a traffic control means selected by a user from among the traffic control means indicated by the generated traffic control information, and generates decision information indicating the selected traffic control means. 16. A traffic monitoring device as described in 14 or 15 above.

[0184] 17. The display control means further includes a display control means for displaying a screen on the display means; The display control means displays the decision information output from the control means on the screen. 17. A traffic monitoring device as described above in claim 16.

[0185] 18. The vehicle further includes a communication means for communicating with a driving control device mounted on the vehicle. The communication means transmits the decision information output from the control means to the operation control device. 18. A traffic monitoring device as described in 16 or 17 above.

[0186] 19. The traffic control means includes at least one of the following: closing an interchange, setting up an alternate route to avoid the location where the abnormal event is predicted to occur, changing the toll fee for the road, and dispatching emergency personnel. 17. A traffic monitoring device according to any one of claims 14 to 16.

[0187] 20. The history generating means a position change acquisition means for acquiring a change in the vehicle position over time based on the position information; a travel direction determination means for determining whether the travel direction of the vehicle is a first direction or a second direction opposite to the first direction based on the obtained change over time; a first generating means for generating the first history information regarding the vehicle traveling in the first direction; and a second generating means for generating second history information indicating a change over time in the vehicle position regarding the vehicle traveling in the second direction. 20. A traffic monitoring device according to any one of claims 1 to 19.

[0188] 21. A traffic condition acquisition means for acquiring a value of a traffic parameter indicating a traffic condition of the road based on the first history information, The traffic parameters include at least one of the following: a speed of a vehicle traveling on a road; a traffic density of the road; a traffic volume, which is an amount of vehicles traveling at a predetermined point on the road per unit time; and an occupancy rate, which is a ratio of a road occupied by vehicles. A traffic monitoring device according to any one of claims 1 to 20.

[0189] 22. A traffic monitoring device according to any one of 1 to 21 above; an optical fiber laid on the road and having one end terminated with a termination treatment for suppressing reflection of an optical signal; a sensing device that inputs an optical signal into the optical fiber and observes a change in optical interference intensity, which is the intensity of light produced by interference between backscattered lights occurring in response to the input of the optical signal; The position acquisition means acquires the position information on the road based on the amount of change in the optical interference intensity observed by the sensing device. Traffic monitoring system.

[0190] 23. The computer obtaining position information indicative of a vehicle's position on a road; generating first history information indicating a change in the vehicle position over time based on the position information. Traffic monitoring method.

[0191] 24. To the computer, obtaining position information indicative of a vehicle's position on a road; generating first history information indicating a change in the vehicle position over time based on the position information. [Explanation of symbols]

[0192] 100 Traffic Monitoring System 101 vehicles 102 Sensing Device 103 Camera 104 Traffic monitoring equipment 105 Input section 106 Position acquisition part 107,207 History Generation Unit 108 First learning model memory unit 109 Traffic Condition Acquisition Department 110 Abnormality detection unit 111 Event-specific history pattern memory unit 112 Prediction Department 113 Second learning model memory unit 114 Abnormality Prediction Department 115 Control Department 116 Display section 117 Display control section 118 Imaging control section 119 Communications Department 120 Operation control device 207a Position change acquisition unit 207b Travel direction discrimination unit 207c 1st generation part 207d 2nd generation part

Claims

1. a position acquisition means for acquiring position information indicating a vehicle position, which is the position of the vehicle on a road; a history generating means for generating first history information indicating a change in the vehicle position over time based on the position information; an abnormality detection means for detecting an abnormal event on the road based on an event-specific history pattern indicating a pattern of a change over time of a vehicle position corresponding to a type of the abnormal event on the road and the change over time of the vehicle position indicated by the first history information; a prediction means for generating prediction information indicating a change over time in a predicted vehicle position on the road based on the first history information generated by the history generating means; an abnormality prediction means for predicting a type of an abnormal event occurring on the road based on the event-specific history pattern and the change over time indicated by the prediction information, The event-specific history pattern includes at least one of a congestion pattern corresponding to a congestion as the abnormal event and an obstacle pattern corresponding to an obstacle on a road as the abnormal event, The congestion pattern indicates that the speed of the vehicle is slower than a reference speed and the inter-vehicle distance is shorter than a reference distance, The obstacle pattern indicates that the vehicle is slowing down, changing lanes, and then returning to the original lane. Traffic monitoring equipment.

2. The position acquisition means acquires the position information on the road based on optical fiber sensing using optical fibers laid on the road.

2. A traffic monitoring device according to claim 1.

3. The event-specific history pattern further includes a stop pattern corresponding to a stop of the vehicle as the abnormal event. A traffic monitoring device according to claim 1 or 2.

4. Further comprising display control means for displaying a screen on the display means; The display control means displays, when an abnormal event is detected by the abnormality detection means, a road map showing the location where the abnormal event has occurred on the screen. A traffic monitoring device according to any one of claims 1 to 3.

5. The imaging device further includes imaging control means for controlling the imaging means for photographing the road, The imaging control means, when an abnormal event is detected by the abnormality detection means, causes the imaging means to take an enlarged image of the location where the abnormal event occurred. A traffic monitoring device according to any one of claims 1 to 4.

6. The prediction means generates prediction information about the road using the first history information generated by the history generation means as an input, based on a trained learning model that has been trained by machine learning to generate the prediction information using the first history information for training as an input. A traffic monitoring device according to any one of claims 1 to 5.

7. The prediction means generates the prediction information based on a type of the abnormal event and the first history information generated by the history generation means when an abnormal event is detected by the abnormality detection means. A traffic monitoring device according to any one of claims 1 to 6.

8. The prediction means generates prediction information about the road by using as input the first history information generated by the history generation means and the type of the abnormal event detected by the abnormality detection means, based on a trained learning model that has undergone machine learning to generate the prediction information by using as input the first history information for training and the type of the abnormal event. A traffic monitoring device according to claim 7.

9. Further comprising display control means for displaying a screen on the display means; The display control means displays, when the abnormality prediction means predicts the occurrence of an abnormal event, a road map showing the location of the predicted abnormal event on the screen. A traffic monitoring device according to any one of claims 1 to 8.

10. The vehicle further includes a communication means for communicating with a driving control device mounted on the vehicle, When the abnormality prediction means predicts the occurrence of an abnormal event, the communication means transmits information indicating a location of the predicted abnormal event to the operation control device.

10. A traffic monitoring device according to claim 9.

11. The system further includes a control unit that generates, when the occurrence of an abnormal event is predicted by the abnormality prediction unit, traffic control information including a traffic control unit for mitigating the abnormal event based on the type of the abnormal event predicted to occur. A traffic monitoring device according to any one of claims 1 to 10.

12. The control means further generates an evaluation value indicating a prediction of a degree to which an abnormal event predicted to occur by the abnormality prediction means will be mitigated by the traffic control means generated by the control means, or evaluation information indicating whether or not the abnormal event is predicted to be mitigated by the traffic control means, based on performance information including the traffic control means adopted for the past abnormal event and an evaluation value indicating a degree to which the abnormal event was mitigated by the traffic control means, or whether or not the abnormal event was mitigated by the traffic control means. A traffic monitoring device according to claim 11.

13. The control means further receives a traffic control means selected by a user from among the traffic control means indicated by the generated traffic control information, and generates decision information indicating the selected traffic control means. Traffic monitoring device according to claim 11 or 12.

14. Further comprising display control means for displaying a screen on the display means; The display control means displays the decision information output from the control means on the screen.

14. A traffic monitoring device according to claim 13.

15. The vehicle further includes a communication means for communicating with a driving control device mounted on the vehicle, The communication means transmits the decision information output from the control means to the operation control device. Traffic monitoring device according to claim 13 or 14.

16. The traffic control means includes at least one of the following: closing an interchange, setting an alternative route to avoid a location where the abnormal event is predicted to occur, changing a toll fee for a road, and dispatching an emergency personnel plan. A traffic monitoring device according to any one of claims 11 to 13.

17. The history generating means a position change acquisition means for acquiring a change in the vehicle position over time based on the position information; a travel direction determination means for determining whether the travel direction of the vehicle is a first direction or a second direction opposite to the first direction based on the obtained change over time; a first generating means for generating the first history information regarding the vehicle traveling in the first direction; and second generating means for generating second history information indicating a change over time in the vehicle position regarding the vehicle traveling in the second direction. A traffic monitoring device according to any one of claims 1 to 16.

18. a traffic condition acquisition means for acquiring a value of a traffic parameter indicating a traffic condition of the road based on the first history information; The traffic parameters include at least one of the following: a speed of a vehicle traveling on a road; a traffic density of the road; a traffic volume, which is an amount of vehicles traveling at a predetermined point on the road per unit time; and an occupancy rate, which is a ratio of a road occupied by vehicles. A traffic monitoring device according to any one of claims 1 to 17.

19. A traffic monitoring device according to any one of claims 1 to 18; an optical fiber laid on the road and having one end terminated with a termination treatment for suppressing reflection of an optical signal; a sensing device that inputs an optical signal into the optical fiber and observes a change in optical interference intensity, which is the intensity of light produced by interference between backscattered lights occurring in response to the input of the optical signal; The position acquisition means acquires the position information on the road based on the amount of change in the optical interference intensity observed by the sensing device. Traffic monitoring system.

20. The computer obtaining position information indicating a vehicle position, the vehicle position being a position of the vehicle on a road; generating first history information indicating a change in the vehicle position over time based on the position information; Detecting an abnormal event on a road based on an event-specific history pattern indicating a pattern of change over time of a vehicle position corresponding to a type of the abnormal event on the road and the change over time of the vehicle position indicated by the first history information; generating prediction information indicating a change over time in a predicted vehicle position for the road based on the generated first history information; predicting a type of abnormal event occurring on the road based on the event-specific history pattern and a change over time indicated by the prediction information; The event-specific history pattern includes at least one of a congestion pattern corresponding to a congestion as the abnormal event and an obstacle pattern corresponding to an obstacle on a road as the abnormal event, The congestion pattern indicates that the speed of the vehicle is slower than a reference speed and the inter-vehicle distance is shorter than a reference distance, The obstacle pattern indicates that the vehicle is slowing down, changing lanes, and then returning to the original lane. Traffic monitoring methods.

21. On the computer, obtaining position information indicating a vehicle position, the vehicle position being a position of the vehicle on a road; generating first history information indicating a change in the vehicle position over time based on the position information; Detecting an abnormal event on a road based on an event-specific history pattern indicating a pattern of change over time of a vehicle position corresponding to a type of the abnormal event on the road and the change over time of the vehicle position indicated by the first history information; generating prediction information indicating a change over time in a predicted vehicle position for the road based on the generated first history information; predicting a type of abnormal event occurring on the road based on the event-specific history pattern and the change over time indicated by the prediction information; The event-specific history pattern includes at least one of a congestion pattern corresponding to a congestion as the abnormal event and an obstacle pattern corresponding to an obstacle on a road as the abnormal event, The congestion pattern indicates that the speed of the vehicle is slower than a reference speed and the inter-vehicle distance is shorter than a reference distance, The obstacle pattern indicates that the vehicle will decelerate and change lanes, then return to the original lane.

Citation Information

Patent Citations

  • On-road object detector

    JP1998019510A

  • Route guiding device

    JP1999203594A

  • Road traffic monitor system

    JP2004524618A

  • Photographing device and congestion determination system

    JP2014078148A

  • Method for notifying that mobile entity is approaching specific area, and server computer and server computer program therefor

    JP2016095695A

Cited By

  • Traffic signal control apparatus, control method of traffic signal machine and program recording medium

    US12380801B2