Lane identification system, lane identification device, and lane identification method

Abstract

A lane identification system according to the present disclosure comprises: an optical fiber (10) that is embedded in a road (R); a measurement unit (21B) that measures, on the basis of an optical signal received from the optical fiber (10), the vibration characteristics of vibrations generated on the road (R); and an identification unit (23B) that derives vibration intensity when a vehicle passed through an observation point, on the basis of vibration characteristics, which are among the vibration characteristics, when the vehicle passed through the observation point on the road (R), and identifies the lane in which the vehicle that passed through the observation point is traveling, on the basis of the derived vibration intensity and a vehicle section of the vehicle that passed through the observation point.

Description

Lane identification system, lane identification device, and lane identification method

[0001] The present disclosure relates to a lane identification system, a lane identification device, and a lane identification method.

[0002] Optical fiber sensing, which uses optical fiber buried in the road as a line sensor, enables a sensing device connected to the optical fiber to measure the amplitude of vibrations of vehicles traveling on the road throughout the entire section where the optical fiber is buried.The sensing device can also visualize the trajectory of the vehicle by plotting the measured vehicle vibration amplitude as a graph of distance and time from the sensing device.

[0003] Meanwhile, in recent years, there has been progress in the study of using self-driving cars on expressways. To make this a reality, it is necessary to know whether there are vehicles in the driving lane before merging points such as interchanges and junctions on expressways, so that self-driving cars can safely enter the driving lane without slowing down.

[0004] An example of an autonomous vehicle merging onto an expressway will now be described with reference to Fig. 1. The example of Fig. 1 shows an autonomous vehicle AC merging onto road R, which is an expressway with two lanes on each side (a driving lane and an overtaking lane). In the example of Fig. 1, an optical fiber 10 for optical fiber sensing is buried along the shoulder of road R.

[0005] When an autonomous vehicle AC merges onto road R, it needs to determine whether there are any vehicles in the merging lane approximately 2 to 10 seconds before the time of merging in order to safely and smoothly enter the driving lane.

[0006] Therefore, when autonomous vehicle AC merges onto road R, the area that autonomous vehicle AC needs to grasp is the merging point and the area just before it (hereinafter referred to as the attention area). Optical fiber sensing, which uses optical fiber 10 as a line sensor, can grasp whether or not a vehicle is traveling in the attention area. However, it has been difficult for optical fiber sensing to grasp whether a vehicle traveling in the attention area is traveling in the driving lane or the overtaking lane.

[0007] 1, a fixed-point monitoring system such as a camera is installed just before the junction of road R. However, the range that can be monitored by the camera is limited to the area centered on the junction, and it is not possible to monitor the entire area of ​​interest. Furthermore, there are often no fixed-point monitoring systems installed at the junction of road R.

[0008] For this reason, recently, techniques have been proposed for identifying the lane in which a vehicle is traveling using optical fiber sensing (for example, Patent Documents 1 and 2). For example, the technique described in Patent Document 1 identifies the lane in which a vehicle is traveling by utilizing the fact that when vibrations are applied in a lane close to the optical fiber, the signal strength acquired by the optical fiber sensing increases, and when vibrations are applied in a lane far from the optical fiber, the signal strength acquired by the optical fiber sensing decreases.

[0009] Furthermore, the technology described in Patent Document 2 identifies the lane in which a vehicle is traveling by utilizing the fact that optical fiber sensing can acquire observation information that can identify the lane in which a vehicle is traveling when the road pavement differs for each lane.

[0010] International Publication No. WO 2022 / 024208 International Publication No. WO 2022 / 185922

[0011] As described above, the technology described in Patent Document 1 identifies the lane a vehicle is traveling in by utilizing the fact that signal strength varies depending on the distance between the optical fiber and each lane. However, since signal strength also varies depending on the size of the vehicle, there is a problem in that identifying the lane a vehicle is traveling in by using only signal strength results in poor identification accuracy.

[0012] Furthermore, the technology described in Patent Document 2 identifies the lane a vehicle is traveling in by utilizing the ability to acquire observation information that can identify the lane a vehicle is traveling in when the road pavement differs for each lane. Therefore, the technology described in Patent Document 2 has a problem in that when the road pavement is the same for each lane, even if it can identify the lane a vehicle is traveling in, the identification accuracy is low.

[0013] In view of the above-mentioned problems, an object of the present disclosure is to provide a lane identification system, a lane identification device, and a lane identification method that are capable of more accurately identifying the lane in which a vehicle is traveling.

[0014] A lane identification system according to one embodiment comprises: an optical fiber buried in a road; a measurement unit that measures vibration characteristics of vibrations generated on the road based on an optical signal received from the optical fiber; and an identification unit that derives vibration intensity when a vehicle passes an observation point on the road based on vibration characteristics among the vibration characteristics when the vehicle passes the observation point, and identifies the lane in which a vehicle that has passed the observation point is traveling based on the derived vibration intensity and the vehicle classification of the vehicle that has passed the observation point.

[0015] A lane identification device according to one embodiment comprises: a measurement unit that measures vibration characteristics of vibrations occurring on a road based on an optical signal received from an optical fiber buried in the road; and an identification unit that derives vibration intensity when a vehicle passes an observation point on the road based on vibration characteristics among the vibration characteristics when the vehicle passes the observation point, and identifies the lane in which a vehicle that has passed the observation point is traveling based on the derived vibration intensity and the vehicle classification of the vehicle that has passed the observation point.

[0016] A lane identification method according to one embodiment is a lane identification method executed by a lane identification device, and includes: a measurement step of measuring vibration characteristics of vibrations generated on the road based on an optical signal received from an optical fiber buried in the road; and an identification step of deriving vibration intensity when a vehicle passes an observation point on the road based on vibration characteristics among the vibration characteristics when the vehicle passes the observation point, and identifying the lane in which the vehicle that passed the observation point is traveling based on the derived vibration intensity and the vehicle classification of the vehicle that passed the observation point.

[0017] According to the above-described aspects, it is possible to provide a lane identification system, a lane identification device, and a lane identification method that are capable of identifying the lane in which a vehicle is traveling with higher accuracy.

[0018] 1 is a diagram showing an example of an autonomous vehicle merging into a driving lane on a highway. FIG. 2 is a diagram showing a schematic configuration example of a lane identification system according to embodiment 1. FIG. 3 is a diagram showing an example of vibration measurement data generated by a learning unit and a recognition unit according to embodiment 1. FIG. 4 is a diagram showing an example of vibration intensity derived by a learning unit and a recognition unit according to embodiment 1. FIG. 5 is a diagram showing an example of statistical data representing, for each vehicle classification of a vehicle, the relationship between vibration intensity when a vehicle passes an observation point on a road and the lane in which the vehicle is traveling after passing the observation point on the road. FIG. 6 is a diagram showing an example of waveform length derived by a learning unit and a recognition unit according to embodiment 1. FIG. 7 is a diagram showing an example of the number of peaks derived by a learning unit and a recognition unit according to embodiment 1. FIG. 8 is a diagram showing a schematic operation example of a lane identification system according to embodiment 1. FIG. 9 is a diagram showing a schematic configuration example of a lane identification system according to embodiment 2. FIG. 10 is a diagram showing an example of statistical data representing, for each vehicle classification of a vehicle, the relationship between vibration intensity and the traveling speed of the vehicle when the vehicle passes an observation point on a road and the lane in which the vehicle is traveling after passing the observation point on the road. FIG. 11 is a diagram showing another example of vibration measurement data generated by a learning unit and a recognition unit according to embodiment 2. Fig. 1 is a diagram showing a schematic operation example of a lane identification system according to embodiment 2. Fig. 2 is a diagram showing a schematic configuration example of a lane identification system according to embodiment 3. Fig. 3 is a flow chart showing an example of a schematic operation flow of the lane identification system according to embodiment 3. Fig. 4 is a block diagram showing a schematic hardware configuration example of a computer that realizes a lane identification device according to each embodiment.

[0019] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. Note that the following description and drawings have been omitted and simplified as appropriate for clarity of explanation. In addition, in each of the following drawings, the same elements are given the same reference numerals, and duplicate explanations are omitted as necessary. Furthermore, specific numerical values ​​etc. shown below are merely examples to facilitate understanding of the present disclosure, and are not limited thereto.

[0020] First Embodiment First, a schematic configuration example of a lane identification system according to the first embodiment will be described with reference to Fig. 2. As shown in Fig. 2, the lane identification system according to the first embodiment includes an optical fiber 10 and a lane identification device 20.

[0021] The optical fiber 10 is buried in the road R. Specifically, in the first embodiment, the road R is a road with two lanes on each side (a driving lane and an overtaking lane), and the optical fiber 10 is buried along the road R and beneath the shoulder of the road R. However, the present invention is not limited to this, and the road R may be a road with three or more lanes, and the optical fiber 10 may be buried beneath a central reservation.

[0022] The lane identification device 20 is realized by a sensing device such as a Distributed Fiber Optic Sensing (DFOS) device, etc. The lane identification device 20 includes a measurement unit 21, a learning unit 22, and an identification unit 23.

[0023] The optical fiber 10 is connected to the measuring unit 21. The measuring unit 21 transmits pulsed light to the optical fiber 10. The measuring unit 21 also receives backscattered light, which is generated as the pulsed light is transmitted through the optical fiber 10, from the optical fiber 10 as an optical signal.

[0024] Here, when vibrations occur on the road R, the vibrations are transmitted to the optical fiber 10, and the characteristics (e.g., wavelength) of the optical signal transmitted through the optical fiber 10 change. Therefore, the measurement unit 21 can detect vibrations occurring on the road R based on the optical signal received from the optical fiber 10. Furthermore, the measurement unit 21 can measure the amplitude of vibrations occurring on the road R based on the degree of change in the characteristics of the optical signal received from the optical fiber 10.

[0025] In this way, the optical fiber 10 functions as a sensor when the measuring unit 21 detects vibrations. Furthermore, in the optical fiber 10, the sensors are distributed linearly along the optical fiber 10, so the optical fiber 10 functions as a line sensor.

[0026] Furthermore, based on the time difference between the time when pulsed light is transmitted to the optical fiber 10 and the time when the optical signal is received from the optical fiber 10, the measurement unit 21 can identify the position where the optical signal was generated, i.e., the position where the vibration on the road R detected based on the optical signal occurred (the distance of the optical fiber 10 from the lane identification device 20).

[0027] In this way, the measuring unit 21 can measure the vibration characteristics of the vibration generated on the road R, such as the amplitude of the vibration and the position where the vibration occurs (the distance of the optical fiber 10 from the lane identification device 20).

[0028] In advance, in a learning phase, the learning unit 22 generates measurement data indicating time-series vibration characteristics of vibrations occurring at an observation point, which is an arbitrary point on the road R, based on the measurement results of the measurement unit 21. Fig. 3 shows an example of measurement data of vibrations occurring at an observation point on the road R. In Fig. 3, the horizontal axis represents time, and the vertical axis represents vibration amplitude.

[0029] Furthermore, vibrations are generated when a vehicle travels on the road R. The waveform of the vibrations generated by the vehicle travelling has a specific waveform pattern in terms of the strength of the vibration, the vibration position, the transition of the fluctuation in the vibration frequency, and the like.

[0030] Therefore, if there is a waveform having a waveform pattern specific to vehicle travel among the waveforms of the measurement data of vibrations generated at an observation point on road R, the learning unit 22 detects that a vehicle corresponding to that waveform has passed the observation point and extracts that waveform. For example, in the example of Fig. 3, there are three waveforms corresponding to three vehicles, respectively. Therefore, the learning unit 22 detects the three vehicles and extracts the three waveforms corresponding to the three vehicles, respectively.

[0031] Next, the learning unit 22 derives the vibration intensity of the waveform as a feature based on the waveform corresponding to a vehicle that has passed an observation point on road R. Fig. 4 shows an example of the vibration intensity derived based on the waveform corresponding to a vehicle. As shown in Fig. 4, the vibration intensity corresponds to the difference between the maximum peak value and the minimum peak value of the waveform.

[0032] The learning unit 22 then generates a feature quantity model that models the relationship between the vibration intensity when a vehicle passes an observation point on the road R and the lane in which the vehicle is traveling after passing the observation point, for each vehicle classification of the vehicle. Specifically, the feature quantity model is a model that represents the boundary of the vibration intensity of each lane for each vehicle classification of the vehicle. Note that the vehicle classification indicates the size (weight) of the vehicle, such as small, normal, or large.

[0033] Figure 5 shows an example of statistical data representing the relationship between the vibration intensity when a vehicle passes an observation point on road R and the lane in which the vehicle is traveling, for each vehicle classification. Figure 5 shows an example in which there are two vehicle classification classes: small vehicles and large vehicles. In Figure 5, the horizontal axis represents vibration intensity, and the vertical axis represents the number of vehicles.

[0034] 5, vibration intensity is characterized by being dependent on the lane in which the vehicle is traveling and the vehicle classification. Also, the closer the lane is to the optical fiber 10, the greater the vibration intensity. Therefore, the learning unit 22 utilizes these characteristics to perform modeling, thereby enabling the lane to be identified from the vibration intensity and vehicle classification.

[0035] In the learning phase, it is already known which lane a vehicle is traveling in and which vehicle's vehicle category is traveling in when the vehicle passes the observation point on road R. For example, if learning is performed at a location that can be monitored by a camera (not shown), the learning unit 22 can identify the lane a vehicle is traveling in and the vehicle category of the vehicle based on the camera data.

[0036] Alternatively, if learning is performed under conditions where the lane the vehicle is traveling in can be known in advance, the learning unit 22 can identify the lane. For example, a driving test in which the vehicle travels in a predetermined lane satisfies the above condition because the lane can be known in advance.

[0037] Alternatively, the learning unit 22 may estimate the vehicle classification of the vehicle based on the measurement results of the measurement unit 21. For example, the learning unit 22 extracts a waveform corresponding to a vehicle that has passed an observation point on road R from the measurement data used to derive the vibration intensity, and then derives the waveform length of the waveform as a feature based on the extracted waveform. FIG. 6 shows an example of the waveform length derived based on the waveform corresponding to a vehicle. As shown in FIG. 6, the waveform length corresponds to the vibration duration. Here, the larger the vehicle, the longer the vehicle length and therefore the longer the vibration duration. Therefore, the learning unit 22 may estimate the vehicle classification based on the waveform length.

[0038] Alternatively, the learning unit 22 extracts from the measurement data a waveform corresponding to a vehicle that has passed an observation point on road R, and then derives the number of peaks in the waveform as a feature based on the extracted waveform. FIG. 7 shows an example of the number of peaks derived based on a waveform corresponding to a vehicle. In the example of FIG. 7, the area surrounded by a dashed circle corresponds to the peak. A peak is considered to occur when an axle of the vehicle passes an observation point on road R. Here, the larger the vehicle, the more axles it has, and therefore the more peaks it has. Therefore, the learning unit 22 may estimate the vehicle classification based on the number of peaks.

[0039] In the operation phase, the identification unit 23 generates measurement data indicating the time-series vibration characteristics of vibrations occurring at observation points on the road R, based on the measurement results of the measurement unit 21. An example of this measurement data is the same as that shown in FIG.

[0040] Next, the identification unit 23 detects a vehicle that has passed the observation point based on the waveform of the measurement data of vibrations that occurred at the observation point on the road R, and if a vehicle that has passed the observation point is detected, it extracts the waveform corresponding to the vehicle that has passed the observation point.

[0041] Next, the identification unit 23 derives the vibration intensity of the waveform as a feature based on the waveform corresponding to the vehicle that passed the observation point on road R. Then, the identification unit 23 identifies the lane in which the vehicle that passed the observation point is traveling based on the derived vibration intensity, the vehicle classification of the vehicle that passed the observation point, and the feature model generated by the learning unit 22.

[0042] The identification unit 23 then outputs the identified lane as an identification result. The identification result may be output to a management system that manages the road R, a terminal in a management room that manages the road R, or the like.

[0043] In the operation phase, the identification unit 23 may estimate the vehicle classification of the vehicle based on the measurement results of the measurement unit 21. The method of estimating the vehicle classification in the identification unit 23 may be the same as that used by the learning unit 22 described above.

[0044] Next, a schematic example of the operation of the lane identification system according to the first embodiment will be described with reference to Fig. 8. First, the operation of the learning phase will be described. First, the learning unit 22 generates measurement data indicating the time-series vibration characteristics of vibrations occurring at observation points on the road R based on the measurement results of the measurement unit 21 (step X11).

[0045] Next, the learning unit 22 detects a vehicle that has passed the observation point on the road R based on the waveform of the measurement data of the vibration generated at the observation point (step X12). If a vehicle that has passed the observation point is detected, the measurement unit 21 extracts a waveform corresponding to the vehicle that has passed the observation point (step X13).

[0046] Next, the learning unit 22 derives the vibration intensity of the waveform as a feature quantity based on the waveform corresponding to the vehicle that passed the observation point on the road R (step X14).

[0047] Then, the learning unit 22 generates a feature model that models the relationship between the vibration intensity when the vehicle passes an observation point on the road R and the lane in which the vehicle that passed the observation point is traveling, for each vehicle classification of the vehicle (step X15).

[0048] Next, the operation of the operation phase will be described. First, the identification unit 23 generates measurement data indicating the time-series vibration characteristics of vibrations occurring at observation points on the road R based on the measurement results of the measurement unit 21 (step Y11).

[0049] Next, the measurement unit 21 detects a vehicle that has passed the observation point on the road R based on the waveform of the measurement data of the vibration generated at the observation point (step Y12). If a vehicle that has passed the observation point is detected, the measurement unit 21 extracts a waveform corresponding to the vehicle that has passed the observation point (step Y13).

[0050] Next, the identification unit 23 derives the vibration intensity of the waveform as a feature based on the waveform corresponding to the vehicle that passed the observation point on the road R (step Y14).

[0051] Next, the identification unit 23 identifies the lane in which the vehicle that passed the observation point on road R is traveling based on the derived waveform length, the vehicle classification of the vehicle that passed the observation point, and the feature model generated by the learning unit 22 (step Y15).

[0052] Thereafter, the identification unit 23 outputs the identified lane for the vehicle that has passed the observation point on the road R as the identification result (step Y16).

[0053] As described above, according to the first embodiment, the measurement unit 21 measures the vibration characteristics of vibrations occurring on the road R based on the optical signal received from the optical fiber 10. In the learning phase, the learning unit 22 derives vibration intensity based on the vibration characteristics measured by the measurement unit 21 when a vehicle passes an observation point on the road R, and generates, for each vehicle classification of the vehicle, a feature quantity model that models the relationship between the vibration intensity when the vehicle passes an observation point on the road R and the lane in which the vehicle is traveling. In addition, in the operation phase, the identification unit 23 derives vibration intensity based on the vibration characteristics measured by the measurement unit 21 when the vehicle passes an observation point, and identifies the lane in which the vehicle is traveling based on the derived vibration intensity, the vehicle classification of the vehicle, and the feature quantity model. In this way, the lane in which the vehicle is traveling is identified using the vehicle classification of the vehicle in addition to the vibration intensity when the vehicle passes an observation point. Therefore, the lane in which the vehicle is traveling can be identified more accurately than in the related art.

[0054] <Embodiment 2> In the above-described embodiment 1, the lane in which a vehicle is traveling is identified by utilizing the vibration intensity when the vehicle passes an observation point on road R and the vehicle classification of the vehicle.

[0055] In contrast to this, in the present embodiment 2, in addition to the vibration intensity when a vehicle passes an observation point on road R and the vehicle classification of the vehicle, the vehicle's traveling speed is also used to identify the lane in which the vehicle is traveling.

[0056] First, a schematic configuration example of a lane identification system according to the second embodiment will be described with reference to Fig. 9. As shown in Fig. 9, the lane identification system according to the second embodiment is configured such that the lane identification device 20 in the lane identification system according to the first embodiment described above is replaced with a lane identification device 20A.

[0057] Furthermore, compared to the lane identification device 20 according to the first embodiment described above, the lane identification device 20A has a configuration in which the learning unit 22 and the identification unit 23 are replaced with a learning unit 22A and an identification unit 23A.

[0058] In advance, during the learning phase, the learning unit 22A generates measurement data (measurement data such as that shown in Figure 3) that indicates the time-series vibration characteristics of vibrations occurring at observation points on the road R based on the measurement results of the measurement unit 21.

[0059] Next, the learning unit 22A detects a vehicle that has passed the observation point based on the waveform of the measurement data of vibrations that occurred at the observation point on the road R, and if a vehicle that has passed the observation point is detected, it extracts the waveform corresponding to the vehicle that has passed the observation point.

[0060] Next, based on the waveform corresponding to a vehicle that has passed an observation point on road R, learning unit 22A derives the vibration intensity of the waveform as a feature amount.

[0061] Then, the learning unit 22A generates, for each vehicle classification of the vehicle, a feature quantity model that models the relationship between the vibration intensity when the vehicle passes an observation point on the road R, the traveling speed of the vehicle that passed the observation point, and the lane in which the vehicle that passed the observation point is traveling. Specifically, the feature quantity model is a model that represents the boundaries of the vibration intensity and traveling speed of each lane for each vehicle classification of the vehicle.

[0062] Fig. 10 shows an example of statistical data representing the relationship between the vibration intensity when a vehicle passes an observation point on road R, the traveling speed of the vehicle passing the observation point, and the lane in which the vehicle passing the observation point is traveling, for each vehicle classification. Fig. 10 shows an example in which there are two vehicle classification classes: small vehicles and large vehicles. In Fig. 10, the horizontal axis represents vibration intensity, and the vertical axis represents the traveling speed of the vehicle.

[0063] As shown in Figure 10, vibration intensity is characterized by being dependent on the lane in which the vehicle is traveling and the vehicle classification. In addition, the traveling speed is characterized by being dependent on the lane in which the vehicle is traveling. In addition, the closer the lane is to the optical fiber 10, the greater the vibration intensity is. Therefore, the learning unit 22A utilizes these characteristics to perform modeling, thereby enabling lane identification from vibration intensity, vehicle classification, and traveling speed.

[0064] In the learning phase, the lane in which a vehicle that has passed an observation point on road R is traveling, the vehicle classification of the vehicle, and the traveling speed of the vehicle are known. For example, if learning is performed at a location that can be monitored by a camera (not shown), the learning unit 22A can identify the lane in which the vehicle is traveling and the vehicle classification of the vehicle based on the camera data.

[0065] Alternatively, if learning is performed under conditions where the lane the vehicle is traveling in can be known in advance, the learning unit 22A can identify the lane. For example, a driving test in which the vehicle travels in a predetermined lane satisfies the above condition because the lane can be known in advance.

[0066] Alternatively, the learning unit 22A may estimate the vehicle classification and traveling speed of the vehicle based on the measurement results of the measurement unit 21. For example, the learning unit 22A generates measurement data as shown in Fig. 11 based on the measurement results of the measurement unit 21. In Fig. 11, the horizontal axis indicates the distance of the optical fiber 10 from the lane identification device 20, and the vertical axis indicates the elapsed time since vibration occurred. Furthermore, the data becomes older as it moves in the positive direction on the vertical axis.

[0067] In the measurement data shown in Figure 11, one vehicle traveling on road R is represented by a single diagonal line. The absolute value of the slope of the line represents the vehicle's traveling speed, with a smaller absolute value of the slope indicating a faster vehicle traveling speed. The positive and negative slopes of the line indicate the vehicle's traveling direction. The horizontal spacing of the lines represents the inter-vehicle distance, with a shorter spacing indicating a shorter inter-vehicle distance.

[0068] Therefore, the learning unit 22A may identify a vehicle based on the time at which the vehicle passed the observation point in the measurement data shown in Figure 11, and estimate the vehicle's traveling speed based on the absolute value of the slope of the line corresponding to that vehicle. Note that the method for estimating the vehicle classification in the learning unit 22A may be the same as that in the first embodiment described above.

[0069] In the operation phase, the identification unit 23A generates measurement data (measurement data such as that shown in Figure 3) that indicates the time-series vibration characteristics of vibrations occurring at observation points on the road R based on the measurement results of the measurement unit 21.

[0070] Next, the identification unit 23A detects a vehicle that has passed the observation point based on the waveform of the measurement data of vibrations that occurred at the observation point on the road R, and if a vehicle that has passed the observation point is detected, it extracts the waveform corresponding to the vehicle that has passed the observation point.

[0071] Next, the identification unit 23A derives the vibration intensity of the waveform as a feature based on the waveform corresponding to the vehicle that passed the observation point on road R. Then, the identification unit 23A identifies the lane in which the vehicle that passed the observation point is traveling based on the derived vibration intensity, the vehicle classification and traveling speed of the vehicle that passed the observation point, and the feature model generated by the learning unit 22A.

[0072] The identification unit 23A then outputs the identified lane as an identification result. The identification result may be output to the same destination as in the first embodiment.

[0073] In the operation phase, the identification unit 23A may estimate the vehicle classification and traveling speed of the vehicle based on the measurement results of the measurement unit 21. The method of estimating the vehicle classification and traveling speed in the identification unit 23A may be the same as that of the learning unit 22A described above.

[0074] Next, a schematic example of the operation of the lane identification system according to the second embodiment will be described with reference to Fig. 12. First, the operation in the learning phase will be described. First, the learning unit 22A performs the processes of steps X21 to X24 similar to steps X11 to X14 in Fig. 8 of the first embodiment described above.

[0075] Then, the learning unit 22A generates a feature model for each vehicle classification of the vehicle that models the relationship between the vibration intensity when the vehicle passes an observation point on the road R, the traveling speed of the vehicle that passed the observation point, and the lane in which the vehicle that passed the observation point is traveling (step X25).

[0076] Next, the operation in the operation phase will be described. First, the identification unit 23A performs the processes of steps Y21 to Y24, which are the same as steps Y11 to Y14 in FIG. 8 of the first embodiment.

[0077] Next, the identification unit 23A identifies the lane in which the vehicle that passed the observation point is traveling based on the derived vibration intensity, the vehicle classification and traveling speed of the vehicle that passed the observation point on road R, and the feature model generated by the learning unit 22A (step Y25).

[0078] Thereafter, the identification unit 23A outputs the identified lane for the vehicle that has passed the observation point on the road R as the identification result (step Y26).

[0079] As described above, according to the second embodiment, the measurement unit 21 measures the vibration characteristics of vibrations occurring on the road R, as in the first embodiment. In the learning phase, the learning unit 22A derives vibration intensity based on the vibration characteristics measured by the measurement unit 21 when the vehicle passes an observation point on the road R, and generates, for each vehicle classification of the vehicle, a feature quantity model that models the relationship between the vibration intensity when the vehicle passes an observation point on the road R, the vehicle's traveling speed, and the lane in which the vehicle is traveling. In addition, in the operation phase, the identification unit 23A derives vibration intensity based on the vibration characteristics measured by the measurement unit 21 when the vehicle passes an observation point, and identifies the lane in which the vehicle is traveling based on the derived vibration intensity, the vehicle classification and traveling speed of the vehicle, and the feature quantity model. In this way, the lane in which the vehicle is traveling is identified using the vehicle classification and traveling speed of the vehicle in addition to the vibration intensity when the vehicle passes an observation point. Therefore, compared to the related art and the first embodiment described above, the lane in which the vehicle is traveling can be identified with higher accuracy.

[0080] <Third Embodiment> This third embodiment corresponds to an embodiment that is a broader concept than the first and second embodiments described above. First, with reference to Fig. 13, a schematic configuration example of a lane identification system according to this third embodiment will be described. As shown in Fig. 13, the lane identification system according to this third embodiment includes an optical fiber 10 and a lane identification device 20B. The lane identification device 20B includes a measurement unit 21B and an identification unit 23B.

[0081] The measurement unit 21B measures the vibration characteristics of vibrations occurring on the road R based on an optical signal received from the optical fiber 10 buried in the road R. The identification unit 23B derives the vibration intensity when the vehicle passes the observation point based on the vibration characteristics measured by the measurement unit 21B when the vehicle passes the observation point on the road R. The identification unit 23B also identifies the lane in which the vehicle that passed the observation point is traveling based on the derived vibration intensity and the vehicle classification of the vehicle that passed the observation point.

[0082] Next, an example of a schematic operation flow of the lane identification system according to the third embodiment will be described with reference to Fig. 14. As shown in Fig. 14, first, the measurement unit 21B measures the vibration characteristics of vibrations occurring on the road R based on an optical signal received from the optical fiber 10 buried in the road R (step S11).

[0083] Next, the identification unit 23B derives the vibration intensity when the vehicle passes the observation point on road R based on the vibration characteristics measured by the measurement unit 21B when the vehicle passes the observation point (step S12).

[0084] Thereafter, the identification unit 23B identifies the lane in which the vehicle that has passed the observation point is traveling, based on the derived vibration intensity and the vehicle classification of the vehicle that has passed the observation point (step S13).

[0085] As described above, according to the third embodiment, the measurement unit 21B measures the vibration characteristics of vibrations occurring on the road R based on an optical signal received from the optical fiber 10 embedded in the road R. The identification unit 23B derives the vibration intensity when the vehicle passes an observation point on the road R based on the vibration characteristics when the vehicle passes an observation point on the road R, among the vibration characteristics measured by the measurement unit 21B. Furthermore, the identification unit 23B identifies the lane in which the vehicle that passed the observation point is traveling, based on the derived vibration intensity and the vehicle classification of the vehicle that passed the observation point. In this way, the lane in which the vehicle is traveling is identified using the vehicle classification of the vehicle in addition to the vibration intensity when the vehicle passed the observation point. Therefore, the lane in which the vehicle is traveling can be identified more accurately than in the related art.

[0086] In addition, the identification unit 23B may estimate the vehicle classification of a vehicle that has passed an observation point on road R based on the vibration characteristics when the vehicle passes the observation point, and identify the lane in which the vehicle that has passed the observation point is traveling based on the derived vibration intensity and the estimated vehicle classification.

[0087] Furthermore, lane identification device 20B may further include a learning unit that generates, in advance, a feature quantity model that models the relationship between the vibration intensity when a vehicle passes an observation point on road R and the lane in which the vehicle that passed the observation point is traveling, for each vehicle classification of the vehicle. In this case, identification unit 23B may identify the lane in which the vehicle that passed the observation point is traveling, based on the derived vibration intensity, the vehicle classification of the vehicle that passed the observation point, and the feature quantity model.

[0088] In addition, the identification unit 23B may identify the lane in which a vehicle that has passed the observation point is traveling based on the derived vibration intensity and the vehicle classification and traveling speed of the vehicle that has passed the observation point on road R.

[0089] In addition, the identification unit 23B may estimate the vehicle classification and traveling speed of a vehicle that has passed the observation point based on the vibration characteristics of the vibrations that have occurred on the road R, and identify the lane in which the vehicle that has passed the observation point is traveling based on the derived vibration intensity and the estimated vehicle classification and traveling speed.

[0090] Furthermore, lane identification device 20B may further include a learning unit that generates, in advance, a feature quantity model that models the relationship between the vibration intensity when a vehicle passes an observation point on road R, the traveling speed of the vehicle that passed the observation point, and the lane in which the vehicle that passed the observation point is traveling, for each vehicle classification of the vehicle. In this case, identification unit 23B may identify the lane in which the vehicle that passed the observation point is traveling, based on the derived vibration intensity, the vehicle classification and traveling speed of the vehicle that passed the observation point, and the feature quantity model.

[0091] <Other Embodiments> In the above-described first embodiment, the learning unit 22 and the identification unit 23 are provided inside the lane identification device 20, but this is not limiting. The learning unit 22 and the identification unit 23 may be provided in a device separate from the lane identification device 20, or may be provided on the cloud. The same applies to the learning unit 22A according to the above-described second embodiment and the identification units 23A and 23B according to the above-described second and third embodiments.

[0092] <Hardware Configuration of Lane Identification Device According to Each Embodiment> Next, with reference to FIG. 15, a schematic example of the hardware configuration of a computer 90 that realizes the lane identification devices 20, 20A, and 20B according to the above-described first to third embodiments will be described.

[0093] 15, a computer 90 includes a processor 91, a memory 92, a storage 93, an input / output interface (input / output I / F) 94, and a communication interface (communication I / F) 95. The processor 91, the memory 92, the storage 93, the input / output interface 94, and the communication interface 95 are connected by a data transmission path for transmitting and receiving data to and from each other.

[0094] The processor 91 is an arithmetic processing device such as a central processing unit (CPU) or a graphics processing unit (GPU). The memory 92 is a memory such as a random access memory (RAM) or a read only memory (ROM). The storage 93 is a storage device such as a hard disk drive (HDD), a solid state drive (SSD), or a memory card. The storage 93 may also be a memory such as a RAM or a ROM.

[0095] The storage 93 stores programs that realize the functions of the components provided in the lane identification devices 20, 20A, and 20B. The processor 91 executes each of these programs to realize the functions of the components provided in the lane identification devices 20, 20A, and 20B. When executing each of the above programs, the processor 91 may read these programs onto the memory 92 before executing them, or may execute them without reading them onto the memory 92. The memory 92 and the storage 93 also serve to store information and data held by the components provided in the lane identification devices 20, 20A, and 20B.

[0096] The above-described program can be stored on various types of non-transitory computer-readable media and supplied to a computer (including computer 90). Non-transitory computer-readable media include various types of tangible storage media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), compact disc-ROMs (CD-ROMs), CD-Recordables (CD-Rs), CD-Rewritables (CD-R / Ws), and semiconductor memories (e.g., mask ROMs, programmable ROMs (PROMs), erasable PROMs (EPROMs), flash ROMs, and RAMs). The program can also be supplied to a computer by various types of transitory computer-readable media. Examples of transitory computer-readable media include electrical signals, optical signals, and electromagnetic waves. The temporary computer-readable medium can supply the program to the computer via a wired communication path such as an electric wire or an optical fiber, or via a wireless communication path.

[0097] The input / output interface 94 is connected to a display device 941, an input device 942, a sound output device 943, etc. The display device 941 is a device that displays a screen corresponding to drawing data processed by the processor 91, such as an LCD (Liquid Crystal Display), a CRT (Cathode Ray Tube) display, or a monitor. The input device 942 is a device that accepts operational inputs from an operator, such as a keyboard, a mouse, or a touch sensor. The display device 941 and the input device 942 may be integrated and realized as a touch panel. The sound output device 943 is a device that outputs sound corresponding to the sound data processed by the processor 91, such as a speaker.

[0098] The communication interface 95 transmits and receives data to and from an external device. For example, the communication interface 95 communicates with the external device via a wired communication path or a wireless communication path.

[0099] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure.

[0100] Furthermore, some or all of the above-described embodiments can be described as, but are not limited to, the following supplementary notes. (Supplementary Note 1) A lane identification system comprising: an optical fiber buried in a road; a measurement unit that measures vibration characteristics of vibrations generated on the road based on an optical signal received from the optical fiber; and an identification unit that derives vibration intensity when a vehicle passes an observation point on the road based on vibration characteristics when the vehicle passes the observation point among the vibration characteristics, and identifies a lane in which a vehicle that passed the observation point is traveling based on the derived vibration intensity and a vehicle classification of the vehicle that passed the observation point. (Supplementary Note 2) The lane identification system according to Supplementary Note 1, wherein the identification unit estimates a vehicle classification of a vehicle that passed the observation point based on the vibration characteristics when the vehicle passed the observation point, and identifies a lane in which a vehicle that passed the observation point is traveling based on the derived vibration intensity and the estimated vehicle classification. (Supplementary Note 3) The lane identification system according to Supplementary Note 1 or 2, further comprising a learning unit that generates in advance, for each vehicle classification of a vehicle, a feature quantity model that models the relationship between vibration intensity when the vehicle passes the observation point and the lane in which the vehicle that passed the observation point is traveling, and the identification unit identifies the lane in which the vehicle that passed the observation point is traveling based on the derived vibration intensity, the vehicle classification of the vehicle that passed the observation point, and the feature quantity model. (Supplementary Note 4) The lane identification system according to Supplementary Note 1, wherein the identification unit identifies the lane in which the vehicle that passed the observation point is traveling based on the derived vibration intensity, and the vehicle classification and traveling speed of the vehicle that passed the observation point. (Supplementary Note 5) The lane identification system according to Supplementary Note 4, wherein the identification unit estimates the vehicle classification and traveling speed of the vehicle that passed the observation point based on the vibration characteristics, and identifies the lane in which the vehicle that passed the observation point is traveling based on the derived vibration intensity, the estimated vehicle classification and traveling speed.(Supplementary Note 6) The lane identification system according to Supplementary Note 4 or 5, further comprising a learning unit that generates in advance, for each vehicle classification of a vehicle, a feature quantity model that models the relationship between the vibration intensity when a vehicle passes the observation point and the traveling speed of the vehicle that passed the observation point, and the lane in which the vehicle that passed the observation point is traveling, and the identification unit identifies the lane in which the vehicle that passed the observation point is traveling based on the derived vibration intensity, the vehicle classification and traveling speed of the vehicle that passed the observation point, and the feature quantity model. (Supplementary Note 7) A lane identification device comprising: a measurement unit that measures vibration characteristics of vibrations generated on the road based on an optical signal received from an optical fiber buried in the road, and an identification unit that derives vibration intensity when the vehicle passes the observation point based on vibration characteristics when the vehicle passes an observation point on the road among the vibration characteristics, and identifies the lane in which the vehicle that passed the observation point is traveling based on the derived vibration intensity and the vehicle classification of the vehicle that passed the observation point. (Supplementary Note 8) The lane identification device according to Supplementary Note 7, wherein the identification unit estimates a vehicle classification of a vehicle that has passed the observation point based on vibration characteristics when the vehicle has passed the observation point, and identifies a lane in which the vehicle that has passed the observation point is traveling based on the derived vibration intensity and the estimated vehicle classification. (Supplementary Note 9) The lane identification device according to Supplementary Note 7 or 8, further comprising a learning unit that generates in advance, for each vehicle classification of a vehicle, a feature quantity model that models the relationship between the vibration intensity when the vehicle has passed the observation point and the lane in which the vehicle that has passed the observation point is traveling, and the identification unit identifies the lane in which the vehicle that has passed the observation point is traveling based on the derived vibration intensity, the vehicle classification of the vehicle that has passed the observation point, and the feature quantity model. (Supplementary Note 10) The lane identification device according to Supplementary Note 7, wherein the identification unit identifies a lane in which the vehicle that has passed the observation point is traveling based on the derived vibration intensity, and the vehicle classification and traveling speed of the vehicle that has passed the observation point.(Supplementary Note 11) The lane identification device according to Supplementary Note 10, wherein the identification unit estimates a vehicle classification and a traveling speed of a vehicle that has passed the observation point based on the vibration characteristics, and identifies a lane in which the vehicle that has passed the observation point is traveling based on the derived vibration intensity and the estimated vehicle classification and traveling speed. (Supplementary Note 12) The lane identification device according to Supplementary Note 10 or 11, further comprising a learning unit that generates in advance, for each vehicle classification of a vehicle, a feature quantity model that models the relationship between the vibration intensity when the vehicle passes the observation point, the traveling speed of the vehicle that has passed the observation point, and the lane in which the vehicle that has passed the observation point is traveling, and the identification unit identifies the lane in which the vehicle that has passed the observation point is traveling based on the derived vibration intensity, the vehicle classification and traveling speed of the vehicle that has passed the observation point, and the feature quantity model. (Supplementary Note 13) A lane identification method executed by a lane identification device, comprising: a measurement step of measuring vibration characteristics of vibrations generated on the road based on an optical signal received from an optical fiber buried in the road; and an identification step of deriving vibration intensity when a vehicle passes an observation point on the road based on vibration characteristics when the vehicle passes the observation point among the vibration characteristics, and identifying a lane in which a vehicle that passed the observation point is traveling based on the derived vibration intensity and a vehicle classification of the vehicle that passed the observation point. (Supplementary Note 14) The lane identification method according to Supplementary Note 13, wherein the identification step estimates a vehicle classification of a vehicle that passed the observation point based on the vibration characteristics when the vehicle passed the observation point, and identifies a lane in which a vehicle that passed the observation point is traveling based on the derived vibration intensity and the estimated vehicle classification. (Supplementary Note 15) A lane identification method according to Supplementary Note 13 or 14, further comprising a learning step of generating in advance, for each vehicle classification of a vehicle, a feature model that models the relationship between the vibration intensity when the vehicle passes the observation point and the lane in which the vehicle that passed the observation point is traveling, and in the identification step, identifying the lane in which the vehicle that passed the observation point is traveling based on the derived vibration intensity, the vehicle classification of the vehicle that passed the observation point, and the feature model.(Supplementary Note 16) The lane identification method according to Supplementary Note 13, wherein in the identification step, a lane in which a vehicle that has passed the observation point is traveling is identified based on the derived vibration intensity and the vehicle classification and traveling speed of the vehicle that has passed the observation point. (Supplementary Note 17) The lane identification method according to Supplementary Note 16, wherein in the identification step, a vehicle classification and traveling speed of the vehicle that has passed the observation point are estimated based on the vibration characteristics, and a lane in which a vehicle that has passed the observation point is traveling is identified based on the derived vibration intensity and the estimated vehicle classification and traveling speed. (Supplementary Note 18) A lane identification method according to Supplementary Note 16 or 17, further comprising a learning step of generating in advance, for each vehicle classification of a vehicle, a feature model that models the relationship between the vibration intensity when the vehicle passes the observation point, the traveling speed of the vehicle that passed the observation point, and the lane in which the vehicle that passed the observation point is traveling, and in the identification step, identifying the lane in which the vehicle that passed the observation point is traveling based on the derived vibration intensity, the vehicle classification and traveling speed of the vehicle that passed the observation point, and the feature model.

[0101] 10 Optical fiber 20, 20A, 20B Lane identification device 21, 21B Measurement unit 22, 22A Learning unit 23, 23A, 23B Identification unit 30 Camera 90 Computer 91 Processor 92 Memory 93 Storage 94 Input / output interface 941 Display device 942 Input device 943 Sound output device 95 Communication interface

Claims

1. Optical fiber buried in the road, a measuring unit that measures vibration characteristics of vibrations generated on the road based on the optical signal received from the optical fiber; an identification unit that derives a vibration intensity when a vehicle passes an observation point on the road based on the vibration characteristics when the vehicle passes the observation point among the vibration characteristics, and identifies the lane in which the vehicle that passed the observation point is traveling based on the derived vibration intensity and the vehicle classification of the vehicle that passed the observation point. Lane identification system.

2. The identification unit estimating a vehicle classification of a vehicle that has passed the observation point based on vibration characteristics when the vehicle has passed the observation point; identifying a lane in which a vehicle that has passed the observation point is traveling based on the derived vibration intensity and the estimated vehicle classification; The lane identification system of claim 1 .

3. a learning unit that generates, in advance, a feature quantity model that models the relationship between the vibration intensity when a vehicle passes the observation point and the lane in which the vehicle that has passed the observation point is traveling, for each vehicle classification of the vehicle; the identification unit identifies a lane in which a vehicle that has passed the observation point is traveling, based on the derived vibration intensity, a vehicle classification of the vehicle that has passed the observation point, and the feature model.

3. A lane identification system according to claim 1 or 2.

4. a measuring unit that measures vibration characteristics of vibrations generated on the road based on an optical signal received from an optical fiber buried in the road; an identification unit that derives a vibration intensity when a vehicle passes an observation point on the road based on the vibration characteristics when the vehicle passes the observation point among the vibration characteristics, and identifies the lane in which the vehicle that passed the observation point is traveling based on the derived vibration intensity and the vehicle classification of the vehicle that passed the observation point. Lane identification device.

5. The identification unit estimating a vehicle classification of a vehicle that has passed the observation point based on vibration characteristics when the vehicle has passed the observation point; identifying a lane in which a vehicle that has passed the observation point is traveling based on the derived vibration intensity and the estimated vehicle classification; The lane identification device according to claim 4.

6. a learning unit that generates, in advance, a feature quantity model that models the relationship between the vibration intensity when a vehicle passes the observation point and the lane in which the vehicle that has passed the observation point is traveling, for each vehicle classification of the vehicle; the identification unit identifies a lane in which a vehicle that has passed the observation point is traveling, based on the derived vibration intensity, a vehicle classification of the vehicle that has passed the observation point, and the feature model.

6. The lane identification device according to claim 4 or 5.

7. the identification unit identifies a lane in which a vehicle that has passed the observation point is traveling, based on the derived vibration intensity, and the vehicle classification and traveling speed of the vehicle that has passed the observation point. The lane identification device according to claim 4.

8. The identification unit estimating a vehicle classification and a traveling speed of a vehicle that has passed through the observation point based on the vibration characteristics; identifying a lane in which a vehicle that has passed the observation point is traveling based on the derived vibration intensity and the estimated vehicle classification and traveling speed; The lane identification device according to claim 7.

9. a learning unit that generates in advance, for each vehicle classification of a vehicle, a feature quantity model that models the relationship between the vibration intensity when the vehicle passes the observation point, the traveling speed of the vehicle that has passed the observation point, and the lane in which the vehicle that has passed the observation point is traveling; the identification unit identifies a lane in which a vehicle that has passed the observation point is traveling, based on the derived vibration intensity, the vehicle classification and traveling speed of the vehicle that has passed the observation point, and the feature model.

9. The lane identification device according to claim 7 or 8.

10. A lane identification method executed by a lane identification device, comprising: a measuring step of measuring vibration characteristics of vibrations generated on the road based on an optical signal received from an optical fiber buried in the road; and an identification step of deriving a vibration intensity when a vehicle passes an observation point on the road based on the vibration characteristics when the vehicle passes the observation point among the vibration characteristics, and identifying a lane in which the vehicle that passed the observation point is traveling based on the derived vibration intensity and the vehicle classification of the vehicle that passed the observation point. Lane identification method.