Vehicle type inference system

Abstract

The present disclosure relates to a vehicle type inference system that acquires a signal measured by an optical fiber vibration measuring device connected to an optical fiber having multiple crossing sections laid across a road or a railway at a known interval shorter than the interval between front and rear wheels provided to a vehicle subjected to inference, calculates the vibration waveform of the acquired signal to detect the passing time at which the front and rear wheels of the vehicle passed the multiple crossing sections, and infers the vehicle type on the basis of the passing time of the front and rear wheels of the vehicle.

Description

Vehicle type estimation system

[0001] The present disclosure relates to traffic monitoring technology using optical fiber sensing.

[0002] A method for estimating traffic conditions on a road using a distributed acoustic sensing (DAS) device has been proposed (see, for example, Non-Patent Document 2). DAS is a technology for measuring the longitudinal vibration distribution of an optical fiber serving as a sensing probe (see, for example, Non-Patent Document 1).

[0003] In Non-Patent Document 2, an optical cable is laid along a road underground, and the vibration distribution transmitted to the optical cable is measured using DAS. In Non-Patent Documents 2 and 3, by using artificial intelligence, it is possible to derive the running trajectory of a vehicle from the vibration distribution.

[0004] “Distributed AcousticNibration Sensing (DAS / DVS)”, AP SENSINGhttps: / / www.apsensing.com / technology / distributed-acoustic-sensing-das-dvs“NEC, NEXCO Central Japan Incorporate High-Precision Traffic Monitoring System Utilizing Optical Fiber Sensing Technology”, NEC Press Release, May 24, 2022http: / / jpn.nec.com / press / 202205 / 20220524_01.html#top“Road Monitoring Using Optical Fiber Sensing”, #Deep Learning - Qiitahttps: / / qiita.com / d_ike / items / 08492b3ba5d3c83adccf

[0005] In monitoring traffic flow on a road or railway track, the vehicle type serves as an indicator of road characteristics. However, there are no examples of discriminating vehicle types using vibration distribution. Therefore, an object of the present disclosure is to enable the estimation of the type of vehicle passing on a road based on the vibration waveform obtained by DAS.

[0006] The vehicle speed measurement system of the present disclosure includes a plurality of transverse sections laid across a road or a railway track in an optical cable laid on the road or the railway track. For example, at a point on the road or the track where a vehicle is to be detected, two or more parallel optical cables laid at a known interval shorter than the interval between the front and rear of the wheels provided on the target vehicle are used to cross the road or the track.

[0007] Specifically, the vehicle type estimation system of the present disclosure includes an optical cable, an optical fiber vibration measuring device that measures vibration of the optical cable, and a signal processing device of the present disclosure that acquires a signal from the optical fiber vibration measuring device.

[0008] The signal processing circuit of the present disclosure executes the vehicle type estimation method of the present disclosure. In the vehicle type estimation method of the present disclosure, a signal processing device acquires a signal measured by an optical fiber vibration measuring device connected to an optical cable having a plurality of transverse sections laid across a road or a railway track at a known interval shorter than the interval between the front and rear of the wheels provided on the vehicle to be estimated, and calculates the vibration waveform of the acquired signal, thereby detecting the passing time when the front and rear wheels of the vehicle pass through the plurality of transverse sections, and estimating the type of the vehicle based on the passing time of the front and rear wheels of the vehicle.

[0009] In the present disclosure, the optical cables of the plurality of transverse sections may be laid in parallel. Further, each of the plurality of transverse sections may be provided with a mechanism for increasing the vibration generated when a vehicle passes.

[0010] In addition, the above disclosures can be combined as much as possible.

[0011] According to the present disclosure, it is possible to estimate the type of a vehicle passing on a road or a railway track based on the vibration waveform obtained by DAS.

[0012] An example of a system using DAS is shown. An example of the length of the wheelbase and the type of the vehicle is shown. A configuration example of the vehicle type estimation system of the present embodiment is shown. A configuration example of the vehicle type estimation system of the present embodiment is shown. An example of the vibration waveform of the transverse section is shown. An example of the vibration energy is shown. It is an explanatory diagram of the laying mode of the optical cable.

[0013] Embodiments of this disclosure will be described in detail below with reference to the drawings. However, this disclosure is not limited to the embodiments shown below. These examples are illustrative, and this disclosure can be implemented in various modified and improved forms based on the knowledge of those skilled in the art. In this specification and in the drawings, components with the same reference numerals refer to the same components.

[0014] Figure 1 shows an example of a system using a DAS. This system comprises an optical fiber vibration measuring instrument (DAS) 11 and a signal processing device 12. In this disclosure, as a preliminary step, an optical cable 13 is laid on the road. The optical cable 13 houses an optical fiber (sensing probe) that connects to the DAS 11. When estimating the traffic conditions on the road, the optical cable 13 is connected to the DAS 11, and the vibration of the optical cable 13 is measured using the DAS 11.

[0015] DAS11 can employ any device capable of measuring the vibration distribution along the longitudinal direction of the optical fiber. For example, a coherent-optical time-domain reflectometer (C-OTDR) can be used.

[0016] The signal processing device 12 is a device that analyzes the vibration distribution in the longitudinal direction of the optical cable 13 using the signal obtained from the DAS 11, and performs arbitrary calculation processing. For example, the signal processing device 12 detects and extracts the vibration of the vehicle 100 from the vibration distribution in the longitudinal direction of the optical cable 13.

[0017] As shown in Figure 2, the distance between the front and rear wheels of a vehicle 100 differs depending on the type of vehicle 100, such as a large vehicle, a passenger car, or a small car. For example, the wheelbase W is 4.5 to 7.5 m for large vehicles, 2.5 to 3 m for passenger cars, and 2.0 to 2.5 m for small cars. Therefore, in this disclosure, an optical cable 13 having multiple crossing sections laid to cross a road or railway at a known distance shorter than the distance between the front and rear wheels of the vehicle 100 to be estimated is connected to the DAS 11, and the vehicle type estimation method of this disclosure is executed.

[0018] In the vehicle type estimation method of the present disclosure, the signal processing device 12 acquires the signal measured by the DAS 11, calculates the vibration waveform of the signal acquired from the DAS 11, thereby detecting the passing time when the front and rear wheels of the vehicle pass through the plurality of transverse sections, and estimates the type of the vehicle 100 based on the passing times of the front and rear wheels of the vehicle 100. Details will be described below.

[0019] (First Embodiment) FIG. 3 shows a configuration example of the vehicle type estimation system of the present embodiment. In this figure, the state of the road as seen from above is shown. The optical cable 13 connected to the DAS 11 is laid in the longitudinal direction across the road to be estimated, and includes two transverse sections S Ｎ１ and S Ｎ２ . The transverse sections S Ｎ１ and S Ｎ２ are laid across the lanes of the road, and a transverse underground optical cable that does not change its laid state over a long period under an environment where the vibration of the vehicle 100 is received can be adopted.

[0020] The two transverse sections S Ｎ１ and S Ｎ２ are parallel, and the interval therebetween is D Ｎ . The interval D Ｎ can be a known value shorter than the wheelbase length W of the vehicle 100 to be estimated. Also, the optical pulse transmission time interval of the DAS 11 is made shorter than the time required for the vehicle 100 to travel the interval D Ｎ . Under this condition, the system of the present embodiment measures the vibration of the optical cable 13 using the DAS 11.

[0021] The signal processing device 12 acquires the signal measured by the DAS 11, and uses the acquired signal to calculate the vibration waveforms of the two transverse sections S Ｎ１ and S Ｎ２ . The interval D Ｎ１ between the transverse sections S Ｎ２ and S ＮSince the vibration waveform is shorter than the wheelbase length W of the vehicle 100, the signal processing device 12 obtains a total of four vibration waveforms: two from the front wheels and two from the rear wheels before the vehicle 100 passes. The signal processing device 12 uses these four vibration waveforms to calculate the wheelbase length W of the vehicle 100. This allows the signal processing device 12 to estimate the type of vehicle 100.

[0022] As explained above, the system of this disclosure can estimate the type of vehicle 100 using only measurement data from DAS 11. Therefore, the system of this disclosure has a function to measure the vibration distribution in the longitudinal direction of the optical fiber, and of the measured vibration distribution, the cross section S Ｎ１ and S Ｎ２ It has a function to calculate vibration waveforms, and with only a simple analysis, it becomes possible to estimate the type of passing vehicle, which is an indicator of road characteristics.

[0023] In this embodiment, there are two crossing sections S at two points. Ｎ１ and S Ｎ２ An example is shown that includes this configuration, but the crossing section may have three or more points. Furthermore, although this embodiment shows an example of traffic conditions for a vehicle 100 traveling on a road, by reinterpreting the distance between the front and rear wheels as the distance between two or more wheels, it can also be applied to the conditions of a train traveling along a railway track. The same applies to the following embodiments.

[0024] (Second Embodiment) Figure 4 shows an example of the configuration of the vehicle type estimation system of this embodiment. In this embodiment, there are two cross sections L in the middle of the optical cable 13 ＡＢ and L ＣＤ An example of the arrangement is shown. In DAS11, the time change of vibration intensity can be measured for each section. The signal processing device 12 measures section L ＡＢ , L ＢＣ , L ＣＤ Separate into intervals L ＡＢ and L ＣＤ The vibration waveform is calculated.

[0025] Figure 5 shows the cross section L. ＡＢ and L ＣＤ An example of the vibration waveform is shown. In this embodiment, the transverse section L ＡＢ and LＣＤ The vibration waveform is calculated for each, and the vibration waveform F is calculated accordingly. ＡＦ and its transit time t ＡＦ , vibration waveform F ＡＲ and its transit time t ＡＲ , vibration waveform F ＣＦ and its transit time t ＣＦ , vibration waveform F ＣＲ and its transit time t ＣＲ , can be obtained individually.

[0026] (Procedure S11) The signal processing device 12 calculates the vibration waveform generated by the passage of the vehicle 100. In this embodiment, interval D Ｎ Since it is shorter than the wheelbase of the target vehicle 100, the signal processing device 12 calculates the vibration waveform F generated by the passage of the front wheels of the vehicle 100. ＡＦ and F ＣＦ The signal processing device 12 processes the data at the time t ＡＦ and t ＣＦ The data is recorded on the recording medium. The signal processing device 12 then processes the data at the time t ＡＦ and t ＣＦ Time interval Δt Ｆ Calculate.

[0027] (Procedure S12) The front wheels of vehicle 100 cross two sections L ＡＢ and L ＣＤ After passing through, the rear wheels of vehicle 100 cross two crossing sections L ＡＢ and L ＣＤ It passes through. Therefore, the passage time t ＡＦ and t ＣＦ The signal processing device 12 then calculates the vibration waveform F generated by the passage of the rear wheels of the vehicle 100. ＡＲ and F ＣＲ The signal processing device 12 processes the data at the time t ＡＲ and t ＣＲ The data is recorded on the recording medium. The signal processing device 12 then processes the data at the time t ＡＦ and t ＡＲ Time interval Δt ＦＲ Calculate the time interval Δt. ＦＲ The passage time t ＣＦ and t ＣＲ You may also use [this].

[0028] (Procedure S13) The signal processing device 12 processes the time interval Δt ＦＲ , interval D Ｎ , time interval Δt Ｆ Therefore, the wheelbase length W of vehicle 100 is calculated using the following equation: (Equation 1) W = Δt ＦＲ ×L ＢＣ / Δt Ｆ (1)

[0029] Here, the passage time t ＡＦ The vibration waveform F ＡＦ This could be the time when the maximum amplitude of the vibration waveform F is obtained, or the time when the vibration waveform F is obtained. ＡＦ The vibration intensity in section L ＡＢ It may also be the time when the threshold is exceeded. Other transit times t ＡＲ ,t ＣＦ ,t ＣＲ The same applies to this matter.

[0030] Furthermore, as shown in Figure 3, a single optical cable 13 is laid in a single continuous line to create two parallel cross-sectional sections L. ＡＢ and L ＣＤ Alternatively, a configuration may be formed, or vibrations may be measured using two different optical cables 13 and two DAS units 11.

[0031] (Third Embodiment) By analyzing the vibration distribution, a variety of information can be obtained. Therefore, in this embodiment, an example will be described in which the signal processing device 12 analyzes the vibration distribution of the crossing section, constructs a vehicle type estimation model using the variety of information obtained from the analysis, and estimates the type of vehicle 100 using the vehicle type estimation model.

[0032] (Wheelbase length W) In this embodiment, the wheelbase length W and the histogram statistics of the vehicle type 100 are obtained and used as training data.

[0033] (Vibration Energy E) Since the vibration energy E during vehicle passage depends on the vehicle weight and speed, the vibration energy E during vehicle passage can be a parameter for estimating the type of vehicle 100. The vibration energy E is obtained by squaring the time change of the vibration intensity. For example, the signal processing device 12 uses section L ＡＢ Waveform F ＡＦ and F ＡＲSquaring this, we obtain the time waveform F of the vibration energy E as shown in Figure 6. ＡＢ This can be obtained. The peak value of the vibrational energy E can be used as an explanatory variable.

[0034] The signal processing device 12 processes the time interval Δt Ｆ and known interval D Ｎ Therefore, the speed V of vehicle 100 can be calculated using the following equation: (Mathematics 2) V = D Ｎ / Δt Ｆ (2)

[0035] The signal processing device 12 obtains statistical data by creating scatter plots of vibration energy E and vehicle speed V for each type of vehicle 100, and uses them as training data.

[0036] (During learning) The signal processing device 12 uses this training data to perform learning and constructs a vehicle type estimation model with wheelbase length W, velocity V, and vibration energy E as explanatory variables. For this machine learning model, any model with high estimation accuracy from among representative classification models such as k-nearest neighbors, support vector machines, decision trees, logistic regression, and neural networks can be used.

[0037] (During estimation) The signal processing device 12 uses the signals obtained from the DAS 11 to calculate the wheelbase W, vehicle speed V, and vibration energy E, and applies them to the vehicle type estimation model. This allows the signal processing device 12 to estimate the type of vehicle 100 (large vehicle, passenger car, or small vehicle).

[0038] This embodiment shows an example in which the signal processing device 12 itself constructs a vehicle type estimation model through learning, but the disclosure is not limited thereto. For example, the signal processing device 12 may acquire the vehicle type estimation model obtained as a result of learning from an external source and use it to estimate the type of vehicle 100. Thus, the learning of the vehicle type estimation model itself may be performed by another computer.

[0039] (Fourth Embodiment) The laying of the optical cable 13 in this disclosure can be carried out in any manner that allows for the measurement of road vibrations. For example, as shown in Figure 1, it may be underground under the road, or it may be realized using road surface wiring optical fiber cable design technology.

[0040] In particular, this disclosure describes laying the optical cable 13 across a road. Therefore, as shown in Figure 7(a), a groove 14 may be provided on the road surface, and the optical cable 13 may be housed in the groove 14. Alternatively, as shown in Figure 7(b), a laying layer 15 may be provided on the road surface, and the optical cable 13 may be housed in a cutout in the laying layer 15. The laying layer 15 can use any means to fix the optical cable 13 to the road, for example, resin can be used.

[0041] The grooves 14 and the laying layer 15 may be equipped with a mechanism that generates specific vibrations without hindering the movement of the vehicle 100. For example, vibrations may be reflected within the grooves 14, or irregularities may be provided on the surface of the laying layer 15.

[0042] Furthermore, the system may be equipped with a mechanism to amplify the vibrations generated when the vehicle 100 passes. For example, a speed bump (a structure that raises a part of the road) may be placed directly in front of the optical cable 13. This makes the vibration waveform when the vehicle 100 passes clearer, and also makes it possible to measure the vibrations of slow-moving vehicles and small vehicles such as motorcycles.

[0043] Furthermore, the grooves 14 and the laying layer 15 do not need to be present along the entire length of the optical cable 13. When laid along the longitudinal direction of a road, they can be buried underground, and the grooves 14 or laying layer 15 can be provided only in the sections that cross the road.

[0044] (Other Embodiments) The signal processing device 12 can also be implemented by a computer and a program, and the program can be recorded on a recording medium or provided via a network. The program of this disclosure is a program that causes a computer to implement each of the functions provided in the signal processing device 12 according to this disclosure, and is a program that causes a computer to execute each of the procedures provided in the method executed by the signal processing device 12 according to this disclosure.

[0045] 11: DAS 12: Signal processing unit 13: Optical cable 14: Groove 15: Laying layer 100: Vehicle

Claims

1. A signal processing device that acquires signals measured by an optical fiber vibration measuring instrument to which an optical cable has multiple cross sections laid across a road or railway at known intervals shorter than the front-to-rear distance between the wheels of the vehicle to be estimated is connected, calculates the vibration waveform of the acquired signals to detect the passage time of the front and rear wheels of the vehicle through the multiple cross sections, and estimates the type of vehicle based on the passage times of the front and rear wheels of the vehicle.

2. A vehicle type estimation system comprising: an optical cable having multiple crossing sections laid across a road or railway at known intervals shorter than the distance between the front and rear wheels of the vehicle to be estimated; an optical fiber vibration measuring instrument for measuring the vibration of the optical cable; and a signal processing device according to claim 1 for acquiring a signal from the optical fiber vibration measuring instrument.

3. The vehicle type estimation system according to claim 2, wherein the optical cables in the plurality of crossing sections are laid in parallel, and each of the plurality of crossing sections is provided with a mechanism for amplifying vibrations generated when vehicles pass over it.

4. A method for estimating the type of vehicle, comprising: a signal processing device acquiring a signal measured by an optical fiber vibration measuring instrument to which an optical cable having multiple cross sections laid across a road or railway at a known interval shorter than the front-to-rear distance of the wheels of the vehicle to be estimated is connected; the signal processing device calculating the vibration waveform of the acquired signal to detect the passage time of the front and rear wheels of the vehicle through the multiple cross sections; and the signal processing device estimating the type of vehicle based on the passage times of the front and rear wheels of the vehicle.

Images