Information notification system

Abstract

To provide a system capable of accurately and easily determining abnormality and notifying a traveling vehicle traveling on a track of information for giving a warning.SOLUTION: An information notification system includes a traveling vehicle 10 traveling on a track and a server 20 formed to transmit / receive information about the traveling vehicle 10. The traveling vehicle 10 includes a position measurement unit 17b for measuring a position, an inclination detection unit 17c for detecting a roll angle AR, a pitch angle AP, and a yaw angle AY, and an information transmission unit 17d for transmitting the measured position, the AR, the AP, and the AY to the server 20. The server 20 includes an inclination estimation unit 23c for estimating a roll angle ARr, a pitch angle APr, and a yaw angle AYr for comparison based on a correlation between a position L2 of the traveling vehicle 10 and a predetermined position L1 and an inclination degree T1 of the traveling vehicle 10 and an information notification unit 23d for notifying the traveling vehicle 10 of information for giving a warning based on comparison between the AR, the AP, and the AY and the ARr, the APr, the AYr for comparison.SELECTED DRAWING: Figure 12

Description

Technical Field

[0001] The present invention relates to an information notification system that is a system for notifying information to a traveling vehicle that travels on a track such as a railway line.

Background Art

[0002] Conventionally, as a traveling vehicle that travels on a track, a technique for determining an abnormality related to a railway vehicle is known. Generally, railway vehicles often have axle girder type bogies, and techniques for subjecting such bogies to abnormality determination have been developed. As a technique of this kind, in the device disclosed in Patent Document 1, the rotation angle of the axle girder of the bogie is calculated, and an abnormality is determined based on the calculated rotation angle.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

[0004] According to the device in Patent Document 1 described above, it is possible to obtain a change in the rotation angle due to an event such as a decrease in the rigidity of the axle box support system of the bogie or sagging of the rubber provided between the bogie frame and the axle box, and to detect an abnormality of the bogie. However, since this type of traveling vehicle usually has a plurality of axle girders, a plurality of rotation angle sensors provided for each axle girder are also required for accurate abnormality determination. For this reason, the device configuration for abnormality determination becomes complicated.

[0005] Furthermore, even if an abnormality of the bogie itself can be detected, it is difficult to detect an abnormality caused by factors other than the bogie, such as deformation of the railway line. Therefore, it is difficult to accurately detect an abnormality with this device. From the above, in a traveling vehicle that travels on a track, a technique that can accurately and simply determine an abnormality and notify information for prompting a warning to the traveling vehicle is desired.

[0006] Therefore, in view of the above requirements, the present invention aims to provide a system for notifying a vehicle traveling on a track of information that can accurately and easily detect anomalies and notify the vehicle of information to prompt a warning. [Means for solving the problem]

[0007] The technical means of the present invention for solving this technical problem is characterized by the following points. The information notification system of the present invention is a system for notifying information from the server to the vehicle, comprising a vehicle that travels on a track, and a server connected to the vehicle via an information communication network and configured to send and receive information of the vehicle. In the information notification system of the present invention, the vehicle comprises a position measuring unit for measuring the position of the vehicle, a tilt detection unit for detecting the degree of tilt of the vehicle, and an information transmitting unit for transmitting the measured position of the vehicle and the detected degree of tilt of the vehicle to the server. The server comprises a tilt estimation unit for estimating a comparative degree of tilt of the vehicle based on the transmitted position of the vehicle and a predetermined correlation between the position of the vehicle and the degree of tilt of the vehicle, and an information notification unit for notifying the vehicle of information to prompt a warning based on a comparison between the transmitted degree of tilt of the vehicle and the estimated comparative degree of tilt of the vehicle.

[0008] According to the above configuration, when a vehicle travels on a track such as a railway line, the vehicle's position and degree of tilt are measured and detected, and transmitted sequentially to the server. The server aggregates a large amount of position and tilt data, and this information is organized, for example, into a database. Based on the database, a correlation between position and tilt can be generated when there are no abnormalities in the track. In the tilt estimation unit, the position transmitted from the vehicle is input to this correlation, and the degree of tilt within a safe range corresponding to that position can be estimated. In the information notification unit, the estimated degree of tilt is used as a comparison value and compared with the degree of tilt transmitted from the vehicle.

[0009] In this case, the degree of tilt often changes significantly from normal due to deformation occurring at a predetermined position on the track or in the various components of the running vehicle. In this case, the comparison described above is performed to accurately and easily determine if an anomaly has occurred. When an anomaly is detected, the information notification unit sends a warning to the running vehicle from the server. Therefore, with the above configuration, information on the position and degree of tilt of the running vehicle can be utilized to accurately and easily determine if an anomaly has occurred and to notify the running vehicle of a warning.

[0010] In the information notification system of the present invention, the information transmission unit of the vehicle being driven transmits a first position and a second position measured after the measurement of the first position as the position of the vehicle being driven to be transmitted to the server, and transmits a first tilt degree corresponding to the first position and a second tilt degree corresponding to the second position as the degree of tilt of the vehicle being driven to be transmitted to the server, the server further includes a correlation generation unit that generates a learning model obtained by machine learning based on the first position and the first tilt degree as the correlation between the position of the vehicle being driven and the degree of tilt of the vehicle being driven, the tilt estimation unit of the server is configured to estimate the degree of tilt of the vehicle being driven for comparison by inputting the second position of the vehicle being driven that has been transmitted into the generated learning model, and the information notification unit of the server is configured to notify the vehicle being driven of information to prompt a warning based on a comparison between the second tilt degree of the vehicle being driven that has been transmitted and the estimated degree of tilt of the vehicle being driven for comparison.

[0011] In the information notification system of the present invention, the information notification unit of the server is configured to notify the vehicle of information prompting a warning when the difference between the second degree of tilt and the estimated degree of tilt of the comparison vehicle is greater than or equal to a predetermined value. [Effects of the Invention]

[0012] According to the present invention, information on the position and degree of tilt of a moving vehicle can be utilized to accurately and easily detect abnormalities, and information to prompt a warning to the moving vehicle can be notified. [Brief explanation of the drawing]

[0013] [Figure 1] This is an overall schematic diagram of an information notification system according to an embodiment of the present invention. [Figure 2] Figure 1 is a side view of the vehicle in motion. [Figure 3] Figure 1 is a front view of a moving vehicle, illustrating the state in which the vehicle tilts around an axis along its longitudinal direction. [Figure 4] Figure 1 is a side view of a moving vehicle, illustrating the state in which the vehicle tilts around an axis along the width direction of the vehicle. [Figure 5] Figure 1 is a plan view of a moving vehicle, illustrating the state in which the vehicle tilts around an axis along the height direction. [Figure 6] Figure 1 is a functional block diagram of the vehicle in operation. [Figure 7] This is a time chart illustrating the measurement and transmission timing of the first position, second position, first tilt degree, and second tilt degree for the vehicle shown in Figure 1. [Figure 8] Figure 1 is a functional block diagram of the server. [Figure 9] This figure shows an example of the manner of travel and position measurement when the vehicle shown in Figure 1 is traveling on a railway track. [Figure 10] Figure 1 shows an example of a database and dataset generated by the server shown. [Figure 11] This figure illustrates the process of estimating the degree of slope for comparison based on the correlation, which is a learning model obtained from the server shown in Figure 1. [Figure 12] Figure 1 is a flowchart showing the actual operation of the information notification system.

Best Mode for Carrying Out the Invention

[0014] Hereinafter, embodiments of the present invention will be described based on the drawings.

[0015] <Information Notification System> As shown in FIG. 1, an information notification system 100 according to an embodiment of the present invention is a system for notifying information for prompting a warning to a traveling vehicle 10. The information notification system 100 includes a traveling vehicle 10 and a server 20. The traveling vehicle 10 traveling on a track is connected to the server 20 via an information communication network N. Information acquired by the traveling vehicle 10 can be transmitted to the server 20 by wireless communication.

[0016] The server 20 is a cloud server or the like, and receives and stores transmission information from the traveling vehicle 10. The server 20 is located at a location different from the traveling vehicle 10. Further, the server 20 is configured to notify information for prompting a warning to the traveling vehicle 10 based on the stored information, transmission information from the traveling vehicle 10, and the like.

[0017] <Traveling Vehicle> As shown in FIGS. 2 and 6, the traveling vehicle 10 includes a vehicle body 11, a current collector 12, a traveling device 13, a display device 14, a positioning device 15, a tipping detection device 16, a control device 17, and a communication device 18. In the present embodiment, the traveling vehicle 10 travels on a track such as a line R and periodically stops at a parking lot arranged on the line R. In the present embodiment, the traveling vehicle 10 can take on and off passengers at the parking lot and is used to transport passengers. Note that the traveling vehicle 10 may be a vehicle for transporting goods, and is not limited to these as long as it travels along the track on the track. Also, one or a plurality of traveling vehicles 10 may travel on one line R. The traveling vehicle 10 may be driven by a driver, an operator, etc. getting on, or may be automatically driven. More specifically, the traveling vehicle 10 may be a train, a railway vehicle, a subway vehicle, a Shinkansen vehicle, a linear motor car, a tram, a monorail vehicle, or the like.

[0018] The vehicle body 11 includes a coupler 11a, a driver's cab 11b, and an operating tool 11c. The vehicle body 11 is a substantially rectangular parallelepiped housing, and its longitudinal direction is along the traveling direction K. An entrance / exit is provided on the side surface of the vehicle body 11 in the vehicle width direction, and passengers can board and alight from the interior space of the vehicle body 11 through the entrance / exit. A coupler 11a is provided behind the vehicle body 11 and is connected to the front of another vehicle body 11. That is, a plurality of vehicle bodies 11 are connected via the coupler 11a, and the traveling vehicle 10 is composed of a train of a plurality of vehicle bodies 11.

[0019] Note that, when viewed from the driver facing the same direction as the traveling direction K of the traveling vehicle 10, the front / back, left / right, and up / down directions correspond to the front direction / back direction, left direction / right direction, and up direction / down direction of the arrows shown in each figure, respectively.

[0020] A driver's cab 11b is provided in front of the foremost vehicle body 11. The driver can board and alight from the driver's cab 11b. A display device 14, an operating tool 11 c, etc. are provided on the console of the driver's cab 11b. A control device 17, a communication device 18, etc. are provided below the driver's cab 11b. The operating tool 11c is a lever, a switch, etc. that can be operated by the driver, and sends an instruction signal for the drive system to the control device 17. That is, in response to the operation of the operating tool 11c, the power supplied from the current collector device 12 is adjusted via the control device 17, and the drive of the traveling device 13 is controlled.

[0021] A current collector device 12 is provided on the upper part (for example, the roof) of the vehicle body 11. The current collector device 12 is provided in each vehicle body 11. The current collector device 12 is a pantograph or the like, and slides on the trolley wire when the traveling vehicle 10 travels. The power of the trolley wire is collected by the current collector device 12. Power is supplied from the current collector device 12 to the traveling device 13 via the control device 17.

[0022] A running gear 13 is provided at the bottom of the vehicle body 11. The running gear 13 comprises wheels 13a, a prime mover 13b, and a brake 13c. Multiple running gears 13 are provided on a single vehicle body 11, enabling self-propulsion using a distributed power system. The running gear 13 has an axle extending in the width direction of the vehicle, and the wheels 13a are rotatably connected to the ends of the axles. The running vehicle 10 travels on the track by guiding the wheels 13a of the running gear 13 along the track R. The prime mover 13b is an electric motor or the like, and is supplied with power to rotate the wheels 13a. The brake 13c is an air brake or the like, and is supplied with power to brake the wheels 13a. The control of the prime mover 13b and the controller 13c is performed by instructions sent to the control device 17 in response to the operation of the operating tool 11c.

[0023] The display device 14 in the driver's seat 11b consists of meters, monitors, etc., that are visible to the driver. The display device 14 displays the status of the vehicle 10 (e.g., vehicle speed) and information notified from the server 20 (e.g., warnings) according to the instructions of the control device 17. If the server 20 determines that there is an abnormality in the degree of tilt of the vehicle 10, information to prompt a warning is notified. The display device 14 displays the warning, prompting the driver to take note of the warning. The determination of an abnormality in the degree of tilt of the vehicle 10 is performed based on the measurement position of the vehicle 10 and the degree of tilt.

[0024] A positioning device 15 is installed on the upper part of the vehicle body 11 (for example, on the roof). The positioning device 15 is capable of detecting positioning information, including latitude and longitude, using a satellite positioning system such as GNSS. The positioning device 15 is configured to receive satellite signals transmitted from positioning satellites, including the position of the positioning satellite, transmission time, correction information, etc. Based on these satellite signals, the positioning device 15 determines the latitude and longitude of the vehicle 10 traveling on the railway track R.

[0025] As shown in Figures 2, 3, 4, and 5, a tilt detection device 16 is provided on the underside of the driver's seat 11b, approximately in the center in the vehicle width direction. The tilt detection device 16 is, for example, an IMU (Inertial Measurement Unit) composed of an acceleration sensor, a gyro sensor, etc.

[0026] In this embodiment, the tilt detection device 16 is configured to detect the degree of tilt of the traveling vehicle 10, specifically the roll angle AR, pitch angle AP, and yaw angle AY. Alternatively, instead of detecting all three, one or two of these angles may be detected. Axes KR, KP, and KY are defined corresponding to the roll angle AR, pitch angle AP, and yaw angle AY. Axe KR is along the longitudinal direction of the traveling vehicle 10, and the roll angle AR is the rotation angle around axis KR. Axe KP is along the vehicle width direction of the traveling vehicle 10, and the pitch angle AP is the rotation angle around axis KP. Axe KY is along the vehicle height direction of the traveling vehicle 10, and the yaw angle AY is the rotation angle around axis KY. P and axis KY are orthogonal to each other.

[0027] The state shown by the dashed line in Figure 3, a front view, is when the vehicle 10 is traveling on the reference horizontal track R. In this state, the intersection point P1 between the axle KY and the upper end of the vehicle body 11 is defined. From this state, as shown by the solid line in Figure 3, when the vehicle 10 tilts around the axle KR, the intersection point P1 moves to the intersection point P2 as it rotates. The roll angle AR is the angle formed by the curved line obtained by connecting the intersection point P1, the axle KR, and the intersection point P2. The roll angle AR becomes larger as the degree of tilting around the axle KR increases. This roll angle AR is detected by the tilt detection device 16.

[0028] The state shown by the dashed line in the side view of Figure 4 is when the vehicle 10 is traveling on the reference horizontal track R. In this state, the intersection point P3 between the axle KR and the front end of the vehicle body 11 is defined. From this state, as shown by the solid line in Figure 4, if the vehicle tilts around the axle KP, the intersection point P3 moves to the intersection point P4 as it rotates. The pitch angle AP is the angle formed by the curved line obtained by connecting the intersection point P3, the axle KP, and the intersection point P4. The pitch angle AP becomes larger as the degree of tilting around the axle KP increases. This pitch angle AP is detected by the tilt detection device 16.

[0029] The dashed line in Figure 5, a plan view, shows the state in which the vehicle 10 is traveling on the reference horizontal track R. In this state, the intersection point P5 between the axle KP and the side end of the vehicle body 11 is defined. From this state, as shown by the solid line in Figure 5, if the vehicle tilts around the axle KY, the intersection point P5 moves to intersection point P6 as it rotates. The yaw angle AY is the angle formed by the curved line obtained by connecting intersection point P5, axle KY, and intersection point P6. The yaw angle AY becomes larger the greater the degree of tilting around axle KY. This yaw angle AY is detected by the tilt detection device 16.

[0030] As shown in Figure 6, the control device 17 is composed of a CPU, electrical circuits, etc., and is electrically connected to various devices. The control device 17 includes a drive control unit 17a, a position measuring unit 17b, a tilt detection unit 17c, an information transmission unit 17d, and an information display unit 17e. The drive control unit 17a is supplied with power from the current collector 12 and receives operation signals from the operating device 11c. The drive control unit 17a adjusts the supplied power according to the operation signals from the operating device 11c, supplies the adjusted power to the running device 13, and controls the driving of the prime mover 13b and braker 13c.

[0031] The position measuring unit 17b instructs the positioning device 15 to measure the position of the vehicle 10. The tilt detection unit 17c instructs the tilt detection device 16 to detect the roll angle AR, pitch angle AP, and yaw angle AY of the vehicle 10. In this embodiment, the measurement and detection by the positioning device 15 and the tilt detection device 16 are performed simultaneously at predetermined intervals while the vehicle 10 is in motion. For example, as the vehicle 10 travels back and forth from one end of the track R to the other, the above-described measurement and detection are performed simultaneously multiple times. As a result, multiple different positions on the track R are measured in small increments. Along with this, multiple degrees of tilt (roll angle AR, pitch angle AP, and yaw angle AY) are detected to correspond to these multiple positions.

[0032] The information transmission unit 17d instructs the communication device 18 to transmit the measured position of the moving vehicle 10, as well as the detected roll angle AR, pitch angle AP, and yaw angle AY of the moving vehicle 10, to the server 20. The position of the moving vehicle 10 is transmitted to the server 20 sequentially each time it is measured as described above. The roll angle AR, pitch angle AP, and yaw angle AY of the moving vehicle 10 are transmitted to the server 20 sequentially each time they are detected as described above.

[0033] As shown in Figure 7, in this embodiment, the position measured and transmitted at predetermined intervals dt is defined as the first position L1. The position measured and transmitted after the measurement of the first position L1 is defined as Let this be the second position L2. The degree of tilt (roll angle AR, pitch angle AP, and yaw angle AY) corresponding to the first position L1 and the second position L2 shall be the first degree of tilt T1 and the second degree of tilt T2, respectively.

[0034] More specifically, as shown in Figure 7(a), the position of the latest moving vehicle 10 measured at time t becomes the second position L2, and the corresponding degree of tilt of the latest moving vehicle 10 becomes the second degree of tilt T2. At times t-dt, t-2d, etc., prior to time t, the measured position becomes the first position L1, and the corresponding degree of tilt becomes the first degree of tilt T1, etc. When time progresses by a period dt from time t, as shown in Figure 7(b), at time t+dt, the position of the latest moving vehicle 10 is measured as the second position L2, and the corresponding degree of tilt of the latest moving vehicle 10 is detected as the second degree of tilt T2. At times t, td, t-2d, etc., prior to time t+dt, the measured position becomes the first position L1, and the corresponding degree of tilt becomes the first degree of tilt T1, etc.

[0035] In other words, in this embodiment, the position of the most recent vehicle 10 and the degree of tilt of the most recent vehicle 10 become the second position L2 and the second degree of tilt T2. On the other hand, the positions of multiple past vehicles 10 and the degrees of tilt of multiple past vehicles 10 become the first position L1 and the first degree of tilt T1. The first position L1 and the first degree of tilt T1 transmitted in this manner are used by the server 20 to generate a correlation, as will be described later. On the other hand, the second position L2 and the second degree of tilt T2 are used by the server 20 to determine whether or not to notify the vehicle 10 of warning information, as will be described later.

[0036] The communication device 18 is a communication module that communicates information with the communication device 21 of the server 20. The communication device 18 may perform wireless communication using, for example, a wireless communication system of the IEEE 802.11 series, a fifth-generation communication system, a predetermined mobile phone communication network, or a data communication network. In response to instructions from the information transmission unit 17d, the communication device 18 transmits information on the first position L1, the second position L2, the first tilt degree T1, and the second tilt degree T2 to the server 20 as described above. The communication device 18 also receives information to prompt a warning that is notified from the server 20 to the moving vehicle 10. The information display unit 17e displays a warning on the display device 14 so that it can be seen by the driver, according to the information received by the communication device 18.

[0037] <server> As shown in Figure 8, the server 20 includes a communication device 21, a storage device 22, and an arithmetic unit 23. The communication device 21, storage device 22, and arithmetic unit 23 are electrically connected to each other via a bus, interface, etc. The communication device 21 is a communication module that communicates information with the communication device 18 of the vehicle 10. The communication standard for this information communication is the same as that of the communication device 18. The communication device 21 receives a first position L1, a second position L2, a first tilt degree T1, and a second tilt degree T2 transmitted from the vehicle 10. The communication device 21 also transmits information to the vehicle 10 to prompt a warning.

[0038] The storage device 22 is a storage device, such as a flash memory SSD or a magnetic disk HDD. The storage device 22 is capable of storing various information received from the moving vehicle 10 via the communication device 21. The storage device 22 is also capable of storing various correlations generated by the correlation generation unit 23b. Furthermore, the storage device 22 also stores programs, parameters, etc., used for calculation processing by the arithmetic unit 23. The stored information can be updated and read by the arithmetic unit 23 according to calculations.

[0039] The arithmetic unit 23 consists of a CPU, electrical circuits, etc., and performs information transmission and reception with the communication device 21, storage in the memory device 22, and various calculations. The arithmetic unit 23 includes an information storage unit 23a, a correlation generation unit 23b, a tilt estimation unit 23c, and an information notification unit 23d.

[0040] The information storage unit 23a stores the position and tilt degree information of the moving vehicle 10, which is transmitted from the information transmission unit 17d of the moving vehicle 10 and received by the communication device 21, in the storage device 22. The stored information is organized into a database DB, which will be described later (see Figure 10). Each time new information is transmitted from the moving vehicle 10, that information is sequentially added to the database DB. The information storage unit 23a also stores the correlations generated by the correlation generation unit 23b in the storage device 22. The stored correlations are the correlation between the first position L1 and the first tilt degree T1, and these are sequentially stored in the storage device 22 each time they are updated by the correlation generation unit 23b.

[0041] The correlation generation unit 23b generates a correlation between the position of the moving vehicle 10 and the degree of tilt of the moving vehicle 10. A learning model is generated from this correlation. This learning model is obtained by machine learning based on the first position L1 and the first degree of tilt T1 stored in the storage device 22.

[0042] The tilt estimation unit 23c estimates the degree of tilt of the comparison vehicle 10 based on the position of the vehicle 10 and a predetermined correlation between the position of the vehicle 10 and the degree of tilt of the vehicle 10. More specifically, the degree of tilt of the comparison vehicle 10 is estimated when the second position L2 of the vehicle 10 is input to the learning model generated as a correlation by the correlation generation unit 23b.

[0043] The information notification unit 23d notifies the train 10 of information to prompt a warning based on a comparison between the second degree of tilt T2 of the transmitted train 10 and the degree of tilt of a comparison train 10 estimated by the tilt estimation unit 23c. For example, the "information to prompt a warning" may be a message that the driver can visually recognize and detect when displayed on the display device 14, such as "Abnormality on the track," or a warning sound may be emitted at the same time.

[0044] <Warning notifications using correlation> Server 20 generates a correlation between the first position L1 and the first degree of tilt T1 of the moving vehicle 10. In this embodiment, a learning model is generated from this correlation. Using the generated learning model and the second position L2 of the moving vehicle 10, information to prompt a warning is notified to the moving vehicle 10. The generation of the correlation and the notification of warnings using the correlation will be described in detail below.

[0045] As shown in Figure 9, on the track R on which the train 10 travels, one end is defined as the starting point PD and the other end as the ending point PA. The train 10 travels from the starting point PD towards the ending point PA. Upon reaching the ending point PA, the train 10 travels towards the starting point PD. In other words, the train 10 periodically travels back and forth between the starting point PD and the ending point PA. The direction of travel K also switches as the train 10 travels back and forth on the track R.

[0046] The first position L1 of the train 10 being measured is the latitude, longitude, etc., of the train 10 as it travels along track R. As the train 10 travels back and forth along track R, the first position L1 of the train 10 is measured at predetermined intervals dt (see Figure 7). Therefore, positions L11, L12, L13, L14, ..., Ln-1, Ln are measured from the starting point PD to the ending point PA of track R. As the train 10 operates on a daily basis and travels back and forth along track R many times, many first tilt degrees T1 (roll angle AR, pitch angle AP, and yaw angle AY) corresponding to each of the above positions are detected. Positions L11, L12, L13, L14, ..., Ln-1, Ln, and the corresponding roll angle AR, pitch angle AP, and yaw angle AY are transmitted in large quantities from the moving vehicle 10 to the server 20. This information is aggregated into a massive amount of big data on the server 20.

[0047] As shown in Figure 10, the information aggregated in the server 20 is organized into a database DB by the information storage unit 23a and stored in the storage device 22. The database DB consists of a first position L1 and a first tilt degree T1 (roll angle AR, pitch angle AP, and yaw angle AY) detected when the first position L1 is measured. That is, the roll angle AR, pitch angle AP, and yaw angle AY are linked to their respective positions L11, L12, L13, L14, L15, etc., and the database DB is configured accordingly. Each time the first position L1 and the corresponding first tilt degree T1 are transmitted to the server 20, the information in the database DB is accumulated and updated as a whole. Correlations are generated based on the database DB constructed in this way.

[0048] As shown in Figure 11, in this embodiment, based on the database DB, the correlation generation unit 23b generates three correlations: CR1 between the first position L1 and the roll angle AR, CR2 between the first position L1 and the pitch angle AP, and CR3 between the first position L1 and the yaw angle AY. Correlations CR1, CR2, and CR3 may be generated, for example, as a learning model obtained through machine learning. Various methods may be used for machine learning, such as supervised learning and unsupervised learning. When using supervised learning, a linear polynomial may be used as the model for predicting the target value.

[0049] More specifically, information may be read from the database DB of the storage device 22, a dataset may be generated in the correlation generation unit 23b, and machine learning may be performed using this dataset as training data. In this case, as shown in Figure 10, for example, datasets DS1, in which the first position L1 is associated with the roll angle AR, dataset DS2, in which the first position L1 is associated with the pitch angle AP, and dataset DS3, in which the first position L1 is associated with the yaw angle AY, are generated. As learning is repeated using datasets DS1, DS2, and DS3, correlations CR1, CR2, and CR3 corresponding to datasets DS1, DS2, and DS3 are generated, respectively.

[0050] The degree of tilt T1 (roll angle AR, pitch angle AP, and yaw angle AY) transmitted from the train 10 to the server 20 normally remains within a safe range at any position L1, provided there are no abnormalities such as deformation in the track R. Therefore, the learned models CR1, CR2, and CR3, which are learned as described above, can output roll angle AR, pitch angle AP, and yaw angle AY within a safe range. In other words, when arbitrary position information is input to the learned model, the degree of tilt within a safe range corresponding to that position is output. By comparing the outputted degree of tilt within a safe range with the actually detected degree of tilt, it is possible to determine whether or not there is an abnormality. Using this, the server 20 sends a warning notification to the train 10 using the correlates CR1, CR2, and CR3.

[0051] As shown in Figure 11(a), the tilt estimation unit 23c outputs the roll angle ARr when the second position L2 of the moving vehicle 10 is input to the learned model correlation CR1. As shown in Figure 11(b), the tilt estimation unit 23c outputs the pitch angle APr when the second position L2 of the moving vehicle 10 is input to the learned model correlation CR2. As shown in Figure 11(c), the tilt estimation unit 23c outputs the yaw angle AYr when the second position L2 of the moving vehicle 10 is input to the learned model correlation CR3. The output roll angle ARr, pitch angle APr, and yaw angle AYr correspond to the roll angle AR, pitch angle AP, and yaw angle AY "within a safe range" at the second position L2, and therefore the roll angle A for "comparison" is also output. These can be used as R, pitch angle AP, and yaw angle AY. In this way, the tilt estimation unit 23c estimates the comparative roll angle ARr, pitch angle APr, and yaw angle AYr.

[0052] This second position L2 is the most recent of the positions of the traveling vehicle 10 that are measured and transmitted, and is measured and transmitted after the first position L1 has been measured and transmitted (see Figure 7). Therefore, the correlations CR1, CR2, and CR3 based on the first position L1 are predetermined before the measurement and transmission of the second position L2. On the other hand, the second tilt T2 is also detected and transmitted simultaneously with the measurement and transmission of the second position L2 (see Figure 7). Therefore, the actually detected second tilt degree T2 (roll angle AR, pitch angle AP, and yaw angle AY) can be compared with the comparison roll angle ARr, pitch angle APr, and yaw angle AYr estimated as described above.

[0053] In this embodiment, the information notification unit 23d notifies the vehicle 10 of a warning if the difference between the actually detected second tilt degree T2 and the estimated comparative tilt degree is greater than or equal to a predetermined value. More specifically, for example, if any one of the following three conditions is met, the vehicle 10 may be notified of a warning if an abnormality is detected. In this way, individual judgments are made for tilts at the three angles: roll angle AR, pitch angle AP, and yaw angle AY. This makes it possible to determine abnormalities with greater accuracy.

[0054] 1. When the absolute value of the difference between the comparative roll angle ARr and the actually detected roll angle AR is greater than or equal to a predetermined value dAR. 2. When the absolute value of the difference between the comparative pitch angle APr and the actually detected pitch angle AP is greater than or equal to a predetermined value dAP. 3. When the absolute value of the difference between the yaw angle AYr used for comparison and the actually detected yaw angle AY is greater than or equal to a predetermined value dAY.

[0055] Here, the predetermined values ​​dAR, dAP, and dAY are the maximum displacements at each angle that can ensure safety. The predetermined values ​​dAR, dAP, and dAY may each be constant, or they may be set differently depending on the difference in the position of the vehicle 10, the difference in gradient, the difference in curve curvature, the difference in vehicle speed, etc. In addition, the absolute values ​​of the difference between the roll angle AR, pitch angle AP, and yaw angle AY are used for the determination, but instead, their respective ratios may be used.

[0056] <Actual Operation> The actual operation of the information notification system 100, which includes the vehicle 10 configured as described above and the server 20, will be explained with reference to the series of flowcharts shown in Figure 12. Here, it is assumed that the vehicle 10 is traveling on track R. Furthermore, it is assumed that the server 20 has already generated correlations CR1, CR2, and CR3 based on past first position L1 and first tilt degree T1.

[0057] First, in step ST1, the position measuring unit 17b measures the position of the vehicle 10, and the tilt detection unit 17c detects the degree of tilt of the vehicle 10. The degree of tilt is the roll angle AR, pitch angle AP, and yaw angle AY of the vehicle 10. The position measurement and tilt degree detection are performed after a period dt from the timing of the previous measurement and detection. Here, the position measured and the tilt degree detected are the latest second position L2 and second tilt degree T2 (see Figure 7).

[0058] Next, in step ST2, the information transmission unit 17d performs the measurement and detection in step ST1. The second position L2 and second tilt degree T2 (roll angle AR, pitch angle AP, and yaw angle AY) are transmitted from the vehicle 10 to the server 20.

[0059] Next, in step ST3, the tilt estimation unit 23c inputs the second position L2 transmitted in step ST2 to correlations CR1, CR2, and CR3, respectively. These correlations CR1, CR2, and CR3 have already been generated based on the first position L1 and first tilt degree T1, which were transmitted prior to the timing at which the second position L2 and second tilt degree T2 were transmitted.

[0060] Next, in step ST4, the tilt estimation unit 23c estimates the degree of tilt for comparison based on the input to the correlation of the second position L2 in step ST3. The learned model correlation CR1 outputs the roll angle ARr for comparison. The learned model correlation CR2 outputs the pitch angle APr for comparison. The learned model correlation CR3 outputs the yaw angle AYr for comparison (see Figure 11).

[0061] Next, in step ST5, the information notification unit 23d determines whether the absolute value of the difference between the comparison roll angle ARr estimated in step ST4 and the roll angle AR actually detected in step ST1 is greater than or equal to a predetermined value dAR. If the absolute value of the difference is less than the predetermined value dAR at this point, the determination in step ST5 is "No".

[0062] Next, in step ST6, the information notification unit 23d determines whether the absolute value of the difference between the comparison pitch angle APr estimated in step ST4 and the pitch angle AP actually detected in step ST1 is greater than or equal to a predetermined value dAP. If the absolute value of the difference is less than the predetermined value dAP at this point, the determination in step ST6 is "No".

[0063] Next, in step ST7, the information notification unit 23d determines whether the absolute value of the difference between the comparison yaw angle AYr estimated in step ST4 and the yaw angle AY actually detected in step ST1 is greater than or equal to a predetermined value dAY. If, at this point, the absolute value of the difference is less than the predetermined value dAY, the determination in step ST7 is "No".

[0064] Next, in step ST8, the information storage unit 23a adds the second position L2 and second tilt degree T2 transmitted in step ST2 to the database DB as a new first position L1 and a new first tilt degree T1 (roll angle AR, pitch angle AP, and yaw angle AY) (see Figure 10). As a result, the database DB is updated to reflect the newly added information.

[0065] Next, in step ST9, the correlation generation unit 23b generates datasets DS1, DS2, and DS3 from the database DB updated in step ST8, and generates the learned models correlations CR1, CR2, and CR3 using machine learning. These correlations CR1, CR2, and CR3 are newly updated from the correlations CR1, CR2, and CR3 in step ST3, and are used when transmitting the second position L2 and second tilt degree T2 next time.

[0066] Then, the process returns to step ST1, and as long as "No" is determined in all of steps ST5 to ST7, the processing of steps ST1 to ST9 is repeatedly executed. In this case, the information notification unit 23d considers there to be no abnormality and does not send a warning notification to the vehicle 10. In addition, the first position L1 and the first tilt degree T1 are accumulated in the database DB, and the correlations CR1, CR2, and CR3 are updated each time through machine learning.

[0067] On the other hand, if "Yes" is determined in any one of steps ST5 to ST7 while steps ST1 to ST9 are being executed repeatedly, the process proceeds to step ST10, and information The notification unit 23d notifies the moving vehicle 10 of information to prompt a warning.

[0068] Next, in step ST11, the information display unit 17e displays a warning on the display device 14 so that it can be seen by the driver, in accordance with the information notified in step ST10.

[0069] Next, in step ST12, the driver who has seen the warning display controls the running gear 13. For example, the operating device 11c is operated in the deceleration direction, and in response to this operation, the drive control unit 17a controls the deceleration of the prime mover 13b and the brake 13c. Then, the process returns to step ST1, and the subsequent processes are repeatedly executed.

[0070] <Effects of the Embodiment> As described above, according to the information notification system 100 of the embodiment of the present invention, when a vehicle 10 travels on a track such as a railway track R, the position and degree of tilt of the vehicle 10 are measured and detected and transmitted sequentially to the server 20. The server 20 aggregates information on numerous positions and degrees of tilt, and this information is organized, for example, as a database DB. The correlation generation unit can generate correlations CR1, CR2, and CR3 of position L1 and degree of tilt T1 based on the database DB, assuming there is no abnormality in the railway track R. The tilt estimation unit inputs the position L2 transmitted from the vehicle 10 into these correlations CR1, CR2, and CR3, and estimates the degree of tilt within a safe range corresponding to that position. The information notification unit uses the estimated degree of tilt as a comparison value and compares it with the degree of tilt T2 transmitted from the vehicle 10.

[0071] In this case, the degree of tilt often changes significantly from normal due to deformation occurring at a predetermined position on the track, such as the R section of the rail, or deformation occurring in the various components of the running vehicle 10. In this case, the comparison described above is performed to accurately and easily determine if an anomaly has occurred. When an anomaly is detected, the information notification unit sends a warning to the running vehicle 10 from the server 20. Therefore, the information notification system 100 utilizes information on the position and degree of tilt of the running vehicle 10 to accurately and easily determine if an anomaly has occurred and to notify the running vehicle 10 of a warning.

[0072] Furthermore, in the above embodiment, the information transmission unit transmits the first position L1, the second position L2, the first tilt degree T1, and the second tilt degree T2, respectively. The second position L2 is a position measured after the measurement of the first position L1, and the first and second tilt degrees T1 and T2 are values ​​corresponding to the first and second positions L1 and L2. The correlation generation unit generates learning models (correlation CR1, CR2, CR3) obtained by machine learning based on the datasets DS1, DS2, and DS3 of the first position L1 and the first tilt degree T1. Therefore, a large number of first position L1 and first tilt degrees acquired in the past can be effectively utilized as information indicating that there are no abnormalities in the track R. In addition, for abnormality detection, the latest second position L2 can be input, and the estimated comparative tilt degree and the latest second tilt degree T2 can be used. Therefore, a large amount of information necessary for machine learning can be efficiently aggregated, and a learning model can be automatically obtained through machine learning, allowing for the early establishment of correlations for anomaly detection.

[0073] Furthermore, in the above embodiment, the information notification unit notifies the running vehicle 10 of a warning if the difference between the second tilt degree T2 and the comparative tilt degree exceeds a predetermined value. This allows the predetermined value to be set to the maximum displacement of the tilt degree that can ensure safety. The predetermined value may be a constant value, or it may be set differently depending on the position of the running vehicle 10, the gradient, the curvature of the curve, the vehicle speed, etc. Therefore, the predetermined value that serves as the judgment criterion can be set easily and appropriately according to the circumstances of the track R and the running vehicle 10.

[0074] [Differentiation] The information notification system 100 in the above embodiment was applied to a vehicle 10 traveling on a single track R, but it may be applied to multiple vehicles 10 traveling on multiple different tracks instead. In this case, each vehicle 10 traveling on multiple tracks may transmit its position and degree of tilt to a single server 20. In this case, it is preferable that the server 20 generates a correlation for each of the multiple tracks and performs an anomaly detection for each of the multiple tracks.

[0075] Furthermore, in the information notification system 100 of the above embodiment, three rotation angles, roll angle AR, pitch angle AP, and yaw angle AY, are used as the degree of tilt of the moving vehicle 10. However, one or two of these three angles may be used instead. That is, one or two of the above three rotation angles may be used when the server 20 determines whether or not to notify information to prompt a warning.

[0076] The embodiments disclosed herein should be considered in all respects to be illustrative and not restrictive. The scope of the present invention is indicated by the claims rather than by the foregoing description, and all modifications within the meaning and scope equivalent to the claims are intended. [Explanation of symbols]

[0077] 10…Traveling vehicle, 15…Positioning device, 16…Tilt detection device, 17…Control device, 17b…Position measurement unit, 17c…Tilt detection unit, 17d…Information transmission unit, 20…Server, 23…Calculation unit, 23b…Correlation generation unit, 23c…Tilt estimation unit, 23d…Information notification unit, 100…Information notification system, AR…Roll angle, AP…Pitch angle, AY…Yaw angle, ARr…Roll angle for comparison, APr…Pitch angle for comparison, AYr…Yaw angle for comparison, CR1,CR2,CR3…Correlation, DB…Database, DS1,DS2,DS3…Database, L1…First position, L2…Second position, N…Information communication network, R…Train line, T1…First degree of tilt, T2…Second degree of tilt

Claims

1. Vehicles that travel on tracks, A server connected to the aforementioned vehicle via an information and communication network and configured to send and receive information about the vehicle, An information notification system for notifying information from the server to the moving vehicle, comprising: The aforementioned vehicle is A position measuring unit for measuring the position of the aforementioned moving vehicle, A tilt detection unit for detecting the degree of tilt of the vehicle being driven, An information transmission unit transmits to the server the first position and a second position measured after the measurement of the first position as the measured position of the moving vehicle, and the first degree of tilt corresponding to the first position and the second degree of tilt corresponding to the second position as the detected degree of tilt of the moving vehicle. Equipped with, The aforementioned server, An information storage unit that stores the first position and the first degree of tilt in a storage device, A correlation generation unit generates a learning model obtained by machine learning based on the first position and the first degree of tilt, which is the correlation between the position of the vehicle and the degree of tilt of the vehicle. A tilt estimation unit that estimates the degree of tilt of a comparison vehicle by inputting the second position of the transmitted vehicle into the generated learning model, An information notification unit notifies the vehicle of information to prompt a warning based on a comparison between the second degree of tilt of the transmitted vehicle and the estimated degree of tilt of the comparison vehicle. Equipped with, If the information storage unit determines that the second degree of tilt is normal by the information notification unit, it updates the database of the storage device by adding the second position and the second degree of tilt as a new first position and a new first degree of tilt, The correlation generation unit generates a dataset from the updated database and updates the learning model by machine learning using the dataset as training data. The first degree of tilt and the second degree of tilt each have a roll angle, a pitch angle, and a yaw angle, respectively. The aforementioned information notification unit, An information notification system that, when any one of the following three conditions is met, indicates an abnormality and notifies the vehicle in motion of information to prompt a warning; 1. If the absolute value of the difference between the roll angle used for comparison and the roll angle actually detected is greater than or equal to a predetermined value, 2. If the absolute value of the difference between the pitch angle used for comparison and the pitch angle actually detected is greater than or equal to a predetermined value, 3. When the absolute value of the difference between the yaw angle used for comparison and the yaw angle actually detected is greater than or equal to a predetermined value.

2. The information notification system according to claim 1, wherein the degree of tilt of the comparison vehicle is within a safe range when there is no abnormality in the track.

Images