Vehicle tracking method, vehicle tracking system, and program

Abstract

To provide a vehicle tracking method, a vehicle tracking system, and a program.SOLUTION: A vehicle tracking method includes: receiving dispersion type optical fiber sensing (DFOS) data; specifying a plurality of first hit points inside the DFOS data, each of which corresponds to a position of a corresponding first vehicle at a detection point; determining an initial seed point from among the plurality of specified first hit points; determining whether or not any of the plurality of specified first hit points is an outlier hit point; defining a first cluster by clustering the plurality of specified first hit points from which an arbitrary outlier hit point is excluded; and estimating first vehicle parameters of the first vehicle based on the first cluster.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] This application relates to a distributed fiber optic sensing (DFOS) system and a method of using the same.

Background Art

[0002] Optical fibers are present along a number of roadways. Distributed acoustic sensors (DAS) attached to these optical fibers can detect vibrations where the optical fibers are located. In some cases, these vibrations are the result of passing vehicles. DAS can collect data regarding the number of vehicles, the lane position of the vehicles, and the vehicle speed.

[0003] DAS generates waterfall data based on time and distance to determine traffic parameters. The ability of DAS to detect individual vehicles is related to the amount of noise in the signals detected by DAS.

Summary of the Invention

Problems to be Solved by the Invention

[0004] Improvement in the accuracy of vehicle tracking is desired.

Means for Solving the Problems

[0005] The vehicle tracking method according to the first embodiment of the present disclosure includes receiving distributed fiber optic sensing (DFOS) data, identifying a plurality of first hit points in the DFOS data, each corresponding to the position of a corresponding first vehicle at the detection time, determining an initial seed point from among the identified plurality of first hit points, determining whether any of the identified plurality of first hit points is an outlier hit point, clustering the identified plurality of first hit points with any outlier hit points excluded to define a first cluster, and estimating a first vehicle parameter of the first vehicle based on the first cluster.

[0006] The vehicle tracking system according to the second embodiment of the present disclosure includes a non-transitory computer-readable medium configured to store instructions and a processor connected to the non-transitory computer-readable medium. The processor is configured to execute the instructions for receiving distributed fiber optic sensing (DFOS) data, identifying a plurality of first hit points in the DFOS data, each corresponding to the position of a corresponding first vehicle at the detection time, determining an initial seed point from among the identified plurality of first hit points, determining whether any of the identified plurality of first hit points is an outlier hit point, clustering the identified plurality of first hit points with any outlier hit points excluded to define a first cluster, and estimating a first vehicle parameter of the first vehicle based on the first cluster.

[0007] A program according to a third embodiment of the present disclosure causes a processor to receive distributed fiber optic sensing (DFOS) data, identify a plurality of first hit points in the DFOS data, each corresponding to a position of a corresponding first vehicle at a detection time, determine an initial seed point from among the identified plurality of first hit points, determine whether any of the identified plurality of first hit points is an outlier hit point, cluster the identified plurality of first hit points with any outlier hit points excluded to define a first cluster, and estimate a first vehicle parameter of the first vehicle based on the first cluster.

Brief Description of the Drawings

[0008] Aspects of the present disclosure are best understood from the following detailed description when read in conjunction with the accompanying drawings. Note that various features are not drawn to scale in accordance with industry standard practice. In fact, the dimensions of various features may be arbitrarily increased or decreased for clarity of explanation.

Figure 1A

Figure 1B

Figure 2

Figure 3

Figure 4

Figure 5A

Figure 5B

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

[0009] The following disclosure provides many different embodiments or examples for implementing different features of the provided subject matter. Hereinafter, to simplify the present disclosure, specific examples of components, values, operations, materials, arrangements, etc. are described. Of course, these are merely examples and are not intended to be limiting. Other components, values, operations, materials, arrangements, etc. are conceivable. For example, the formation of a second feature covering or on top of a first feature in the following description may include embodiments in which the first and second features are formed in direct contact, and may also include embodiments in which additional features are formed between the first and second features so that the first and second features are not in direct contact. Further, the present disclosure may repeat reference numerals and / or letters in various examples. This repetition is for the purpose of simplification and clarity and does not itself define the relationship between the various embodiments and / or configurations discussed.

[0010] Furthermore, spatially relative terms such as "beneath," "below," "lower," "above," "upper," etc. may be used herein for ease of explanation to describe the relationship of one element or feature to another element or feature as illustrated in the figures. Spatially relative terms are intended to encompass different orientations of the device in use or operation in addition to the orientation shown in the figures. The device may be oriented in other directions (rotated 90 degrees or other orientations), and the spatially relative descriptors used herein may be interpreted accordingly as well.

[0011] Utilizing data from optical fibers along a roadway is useful for determining traffic volume, traffic speed, accidents, and other events along the roadway. To enhance the usefulness of traffic information obtained based on data from optical fibers, determination of vehicle parameters such as speed, acceleration, and lane position provides information useful for identifying traffic patterns and navigation information. The quality of the collected distributed fiber optic sensing (DFOS) data is determined by many factors. Since DFOS data is based at least in part on vibrations, the size of the vehicle, traffic volume, and the type of roadway affect the quality of the data. For example, a single vehicle traveling along a roadway directly on the ground is more likely to provide a higher quality signal than a large truck traveling across a busy bridge. Noise removal of DFOS data serves to analyze the DFOS data for use in applications such as traffic analysis, autonomous driving, and navigation instructions.

[0012] In addition, the accuracy of traffic information assists in navigating vehicles traveling along a lane. By providing more accurate traffic data to the driver, the navigation system and / or navigation application becomes more useful to the driver. High-precision navigation is also useful for the autonomous driving or driver assistance functions of a vehicle. By accurately determining where traffic congestion or an accident has occurred, an autonomous vehicle or driver assistance system can guide the vehicle along a more efficient route.

[0013] This application describes a method and system for improving vehicle tracking along a lane using DFOS data. Using the detection positions, also called hit points, of objects along the lane with DFOS data makes it possible to determine the position of a vehicle along the lane at various points in time. Using threshold processing techniques helps to remove incorrect data, also called outlier hit points, from the DFOS data used to track the vehicle. As a result, the accuracy and precision of vehicle tracking are improved compared to other methods. The clustering of hit points also helps to provide vehicle data such as speed to further improve the usefulness of DFOS data for both vehicle tracking and traffic congestion analysis.

[0014] Furthermore, the improvement in the accuracy and precision of vehicle tracking described in this application helps to facilitate the tracking of multiple vehicles along the same lane even when the DFOS data of multiple vehicles includes intersections. By utilizing threshold processing and clustering, this application provides a method for accurately tracking multiple vehicles by using vehicle data such as speed to accurately determine the respective paths of each vehicle even when there are multiple vehicles passing through an intersection of DFOS data. An intersection of DFOS data is a location where the hit points of each of multiple vehicles cross each other on a distance-versus-time plot. An intersection of DFOS data does not indicate an intersection within a lane.

[0015] FIG. 1A is a schematic diagram of a distributed acoustic sensor (DAS) system 100A along a lane 130A according to some embodiments. The DAS system 100A includes a traffic monitoring device 111 that communicates with a DAS 112. The DAS system 100A further includes an optical fiber 121 connected to the DAS 112. The optical fiber 121 is along the lane 130A. The lane 130A includes three lanes. A number of vehicles are on the lane 130A. Some vehicles 140 on the lane 130A are larger than other vehicles 150 on the lane 130A. Although the description refers to the optical fiber 121, those skilled in the art will understand that in some embodiments the optical fiber 121 includes a multi-fiber bundle.

[0016] When vehicles 140 and 150 pass along the lane 130A, the vehicles generate vibrations. These vibrations cause a change in the way light propagates along the optical fiber 121. The DAS 112 is connected to the optical fiber 121, sends an optical signal into the optical fiber 121, and detects the return light from the optical fiber 121. The resulting data is called waterfall data. The waterfall data provides information regarding the number of vehicles, the direction of travel by the vehicles, the vehicle speeds, and the lane positions of the vehicles on the lane 130A.

[0017] The lane 130A of FIG. 1 is on solid ground. The solid ground does not vibrate with an amplitude high enough to obscure the detection of vehicles 140 and 150 traveling along the lane 130A. As a result, the DAS 112 can accurately detect vehicles 140, 150 traveling along the lane 130A. In some embodiments, the lane 130A includes at least one bridge such as the lane 130B of FIG. 1B.

[0018] Unlike solid ground, a bridge exhibits different vibration characteristics such as attenuation. The vibration characteristics of the bridge are affected by the length of the bridge, the building materials of the bridge, wind, and other factors. These differences in the vibration characteristics of the bridge can be utilized to determine where the bridge is located along the optical fiber 121.

[0019] FIG. 1A also includes exemplary measured DAS data. This exemplary measured DAS data is provided to assist in understanding the waterfall data collected by DAS 112.

[0020] FIG. 1B is a schematic diagram of DAS system 100B along lane 130B according to some embodiments. Similar to DAS system 100A of FIG. 1A, DAS system 100B includes DAS 112 and optical fiber 121. In contrast to FIG. 1A, lane 130B of FIG. 1B includes first bridge 160A and second bridge 160B. Further, FIG. 1B includes additional fiber portion 170.

[0021] Considering the first vehicle 150A and the second vehicle 150B is helpful for understanding the use of fixed reference points such as the first bridge 160A and the second bridge 160B when determining the position corresponding to traffic information with high precision. The distance between the first vehicle 150A and the second vehicle 150B along the lane 130B is significantly different from the length of the optical fiber 121 between the position of the first vehicle 150A and the position of the second vehicle 150B. This difference is due to the presence of the additional fiber portion 170 and the fact that the optical fiber 121 is not exactly parallel to the lane 130B. Determining the position of the first bridge 160A along the optical fiber 121 is helpful for determining the high-precision position of the first vehicle 150A along the lane 130B. The position of the first bridge 160A along the lane 130B is known based on the published geographical data. By determining the position of the first bridge 160A with respect to the optical fiber 121, the length of the optical fiber 121 from the DAS 112 to the end of the first bridge 160A closest to the first vehicle 150A is determined based on the waterfall data. Next, based on the waterfall data, the length of the optical fiber 121 between the end of the first bridge 160A and the first vehicle 150A is determined. By limiting the distance of the optical fiber 121 from the fixed reference point of the first bridge 160A to the first vehicle 150A, the error in the length of the lane 130 from the end of the first bridge 160A to the DAS 112 is excluded from the positioning. As a result, the position of the first vehicle 150A along the lane 130B can be determined with higher precision using the fixed reference point of the first bridge 160A.

[0022] Similarly, the position of the second vehicle 150B is more accurately determined by using the fixed reference point of the second bridge 160B. The waterfall data from the DAS 112 can be used to determine the length of the optical fiber 121 between the DAS 112 and the end of the second bridge 160B closest to the second vehicle 150B. Next, only the length of the optical fiber 121 between the second vehicle 150B and the second bridge 160B is used to determine the position of the second vehicle 150B along the lane 130B. By using the length of this short optical fiber 121, the length of the optical fiber 121 between the second vehicle 150B and the DAS 112, including the additional fiber portion 170, is excluded from the positioning. As a result, the position of the second vehicle 150B is determined with higher accuracy by using the fixed reference point of the second bridge 160B.

[0023] FIG. 2 is a flowchart of a method for analyzing data from distributed fiber optic sensing (DFOS) data according to some embodiments. The method 200 can be used in the DAS system 100A (FIG. 1A), the DAS system 100B (FIG. 1B), or another suitable system that provides DFOS data.

[0024] In operation 205, raw DFOS data is received. In some embodiments, the raw DFOS data includes waterfall data such as, for example, the waterfall data detected by the DAS 112 (FIG. 1). The waterfall data includes information regarding time and position along the optical fiber where vibration data was detected. Further details of the waterfall data are discussed with respect to FIG. 3 below. In some embodiments, the waterfall data includes data along a lane that includes both a rigid ground and at least one bridge.

[0025] In operation 210, the raw DFOS data is localized using the structural information. The structural information includes information obtained from an external source with respect to fixed reference points along the roadway. For example, in some embodiments, the structural information includes the location of a bridge, such as bridge 160A (FIG. 1B), the location of additional fiber, such as additional fiber 170 (FIG. 1B), or other suitable reference points along the roadway. Using the structural information to localize the DFOS data helps improve the accuracy and precision of the detected vehicle's position, as well as the accurate determination of vehicle parameters such as speed and acceleration. The localization of the DFOS data also helps improve the accuracy and precision of navigation instructions, autonomous driving functions, traffic monitoring, or other suitable applications of the DFOS data.

[0026] In operation 215, the raw DFOS data is pre - processed to enhance the received data. Pre - processing the data includes normalizing the vibration amplitude of the data at each position along the roadway, such as roadway 130B (FIG. 1B), over a predetermined duration. Normalizing the vibration amplitude helps account for variations in the sensitivity of the optical fiber. Variations in the sensitivity of the optical fiber are due to several causes including, but not limited to, non - uniform roadway surfaces, inconsistent installation of the optical fiber, and optical fiber mismatches. Normalizing the vibration amplitude also helps account for variations in traffic volume. For example, as the number of vehicles on the roadway increases, the magnitude of the vibrations detected by the DAS increases. By normalizing the vibration amplitude based on a predetermined duration, the impact of large vibrations detected during high traffic volumes on time periods of low traffic volume is reduced, and more accurate data for estimating traffic flow characteristics is generated.

[0027] In some embodiments, pre - processing the data also includes limiting the maximum vibration amplitude at each position along the optical fiber over a predetermined duration. Limiting the maximum vibration amplitude helps prevent vibrations from large vehicles, such as trucks or construction vehicles, from obscuring vibrations generated by smaller vehicles, such as passenger cars.

[0028] Next, the preprocessed data is filtered by a band-pass filter based on the estimated frequency range from operation 210. Filtering the data excludes from the data portions of the lanes that did not exhibit the vibration damping characteristics of the bridge. FIG. 4 below provides an example of the filtered DFOS data.

[0029] In operation 220, an initial seed is determined for each vehicle based on the hit point data. The initial seed for each vehicle is determined based on the hit point identified at the first time the vehicle is detected, e.g., at time (t)=0. In some embodiments, the first time is the time the vehicle first enters the detection area of a DAS system, e.g., DAS system 100A (FIG. 1A), DAS system 100B (FIG. 1B), or another suitable DAS system. In some embodiments, the first time is a time selected by the user of the DAS system. Hit points are identified based on the detected vibrations within the DFOS data. The detected vibrations are determined based on the width of the lines within the DFOS data. In some cases, the vibrations occur more widely, which indicates a higher vibration intensity. Wider lines have a higher risk of resulting in multiple hit points being identified for the same vehicle. Additionally, in some embodiments, additional vibrations that are not caused by vehicles traveling along the lane generate the detected vibrations within the DFOS data. As a result of these additional vibrations, false hit points may be identified where there are no vehicles.

[0030] In some embodiments, a trained neural network (NN) is utilized to identify hit points within the DFOS data. In some embodiments, the hit points are identified or verified by a user of the DAS system. The identification of the initial hit points assists in tracking a vehicle through at least a portion of the detection area of the DAS system to determine vehicle parameters for use in autonomous driving, navigation instructions, traffic monitoring, or other suitable applications. Additional details regarding the detection of the initial hit points are provided with respect to FIG. 5 below, according to some embodiments.

[0031] In operation 225, subsequent hit points are identified for an individual vehicle. The subsequent hit points are vehicle positions determined using DFOS data at a later time, for example at times t = 1, t = 2, etc. In some embodiments, the interval between the first time and the subsequent times is uniform over each time. In some embodiments, there is an irregular interval between the first time and the subsequent times. In some embodiments, the interval is predetermined. In some embodiments, the interval is set based on the speed limit along the lane. In some embodiments, the interval is set based on the expected traffic volume along the lane, for example, a shorter interval for areas of heavier traffic congestion. In some embodiments, the subsequent hit points are identified in the same manner as the initial hit points. In some embodiments, in addition to detecting vibrations based on the DFOS data, the subsequent hit points are identified based on the number of initial hit points that can be used to predict the number of vehicles being tracked through the detection area of the DAS system. In some embodiments, in addition to detecting vibrations based on the DFOS data, the subsequent hit points are identified based on the expected position of the vehicle. In some embodiments, the trained NN can be used to identify the subsequent hit points. Additional details regarding the detection of the initial hit points are provided with respect to FIG. 5 below, according to some embodiments.

[0032] In operation 230, the outlier hit points are removed. The outlier hit points are hit points within the DFOS data that cannot reasonably correspond to the vehicle being tracked. For example, a hit point that is 100 meters (m) away from a hit point detected 0.5 seconds (s) ago is considered an outlier hit point because the vehicle would have to travel at 200 meters per second (m / s) which is 720 kilometers per hour (km / h) or 447 miles per hour (mph). Such a speed is not reasonable even if it is assumed that the vehicle can travel at such a speed. In some embodiments, determining whether a potential outlier hit point can reasonably correspond to the vehicle being tracked is based on the speed limit along the lane being monitored by the DAS. In some embodiments, determining whether a potential outlier hit point can reasonably correspond to the vehicle being tracked is based on the previously determined speed of the vehicle. For example, in some embodiments, if the difference between the speed required for the vehicle to reach a potential outlier hit point and the previously determined speed of the vehicle is greater than 50% of the previously determined speed of the vehicle, the hit point is determined to be an outlier hit point. In some embodiments, determining whether a potential outlier hit point can reasonably correspond to the rail vehicle is based on the direction of travel of the vehicle. For example, if a potential outlier hit point indicates that the vehicle has changed direction rapidly multiple times, the potential outlier hit point is less likely to reasonably correspond to the vehicle being tracked. In some embodiments, a trained neural network (NN) is utilized to identify outlier hit points within the DFOS data. In some embodiments, the NN is trained using traffic congestion and / or vehicle tracking data along the same lane or a similar lane. By using traffic congestion and / or vehicle tracking data as training data, the NN can determine how frequently the vehicle travels along the lane in order to determine whether a potential outlier hit point can reasonably correspond to the vehicle being tracked.

[0033] In some cases, the outlier hit points are the result of the DAS detecting vibrations along the lane other than strong winds, construction, traffic accidents, or the movement of the vehicle being tracked. Other such occurrences result in the outlier hit points.

[0034] Once a hit point is identified as an outlier hit point, the hit point is no longer considered during vehicle tracking or traffic congestion analysis using DFOS data. In some embodiments, the outlier hit points are removed from the DFOS data. In some embodiments, the outlier hit points are utilized to train the NN to better identify outlier hit points during future analysis of DFOS data.

[0035] In operation 235, a determination is made as to whether a sufficient number of hit points have been identified to perform the clustering operation. By clustering the hit points, it becomes possible to determine vehicle parameters such as the speed along the lane. As the number of hit points within a cluster increases, the accuracy of the determined vehicle parameters improves. However, the more hit points there are within a cluster, the longer it takes to collect more hit points. As a result, based on the use of DFOS data, the determination of the sufficient number of hit points can be adjusted. In some embodiments, the number of hit points sufficient for clustering is a fixed value. In some embodiments, the number of hit points sufficient for clustering varies based on the vehicle or lane situation. For example, in situations where near real-time data such as autonomous driving is desired, in some embodiments, a small number of hit points, such as two or three hit points, are determined to be sufficient for clustering.

[0036] In some embodiments, the determination of the number of hit points sufficient for clustering is based on the detected traffic congestion. As traffic congestion increases, the vehicle speed decreases. As a result, the number of hit points considered sufficient for clustering increases during severe traffic congestion. For example, in some embodiments where DFOS data is utilized in vehicle navigation such as a vehicle global positioning system (GPS), by increasing the number of hit points sufficient for clustering, more data can be collected to provide more accurate results without increasing the risk that the vehicle will pass a location such as an exit ramp or turn due to the vehicle's decreased speed.

[0037] In some embodiments, the determination of the number of hit points sufficient for clustering is based on the previously determined speed of the vehicle. As the vehicle speed increases, the number of hit points considered sufficient for clustering decreases. For example, in some embodiments where DFOS data is utilized in vehicle navigation, decreasing the number of hit points sufficient for clustering reduces the risk that the vehicle will pass a location due to the vehicle's increased speed.

[0038] In some embodiments, the determination of the number of hit points sufficient for clustering is based on a threshold determined by an operator of the DAS. In some embodiments, the threshold corresponds to hit points corresponding to 10 seconds (s). In some embodiments, the threshold is based on empirical data of driving along the monitored lane.

[0039] In response to a determination that an insufficient number of hit points has been identified for clustering, method 200 returns to operation 225 and additional hit points are identified. In response to a determination that a sufficient number of hit points for clustering has been identified, method 200 proceeds to operation 240.

[0040] In operation 240, clustering is performed on the identified hit points. In some embodiments, the hit points are clustered using K-means clustering, and the number of identified hit points determined in operation 235 is used as the value of K. In some embodiments, the clustering is performed using density-based spatial clustering of applications with noise (DBSCAN) clustering, Gaussian mixture model clustering, balanced iterative reducing and clustering using hierarchies (BIRCH) clustering, affinity propagation clustering, mean shift clustering, ordering points to identify the clustering structure (OPTICS) clustering, agglomerative hierarchical clustering, or another suitable type of clustering.

[0041] In operation 245, the next hit point is estimated based on the clustered hit points. The predicted position of the next hit point is estimated based on vehicle parameters such as the speed determined based on the clustered hit points. For example, in some embodiments where DFOS data is captured at regular time intervals, the speed of the vehicle and the regular time intervals can be used to estimate the position where a hit point is expected to be found. In some embodiments, any detected hit point other than the outlier hit points is determined to be a match between the estimated hit point and the detected hit point. In response to the detection of a hit point at the estimated position, the hit point can be used to update the vehicle parameters determined based on the clustered hit points from operation 240. In response to the inability to detect a hit point at the estimated position, subsequent hit points are estimated and operation 245 is repeated to attempt to detect a hit point at the estimated subsequent hit points. The subsequent hit points are hit points after a detection period that is at least one after the next hit point first determined in operation 245.

[0042] In some embodiments, operation 245 is repeatedly iterated until a hit point is detected at a position where the estimated hit point is expected. In some embodiments, a maximum number of iterations is allowed before proceeding to operation 250. That is, if no match is found between the hit point detected within the maximum number of iterations and the estimated hit point, method 200 proceeds to operation 250. In some embodiments, the maximum number of iterations is set by an operator of the DAS. In some embodiments, the maximum number of iterations is based on the speed of the vehicle. As the speed number of the vehicle increases, the maximum number of iterations decreases. In some embodiments, the maximum number of iterations is determined based on the user of the DFOS data. For example, in some embodiments, autonomous driving has a lower maximum number of iterations than traffic congestion analysis. In some embodiments, the maximum number of iterations ranges from 2 to 10. Estimating the hit point helps improve the accuracy of vehicle tracking by reducing the risk of relying on hit point data that is incorrect but not so far off as to be identified as an outlier hit point.

[0043] In operation 250, the initial seed is updated to a new hit point. In some embodiments, the initial seed corresponds to the estimated hit point determined in operation 245. In some embodiments, the initial seed corresponds to the most recent hit point within the clustered hit points from operation 240. In some embodiments, the initial seed corresponds to the first detected hit point after the clustered hit points from operation 240. By using the most recent hit point within the clustered hit points, the accuracy of method 200 in vehicle tracking or traffic congestion analysis is improved. By using a hit point after the clustered hit points, the processing speed for tracking the vehicle or performing traffic congestion analysis is improved.

[0044] Following operation 250, method 200 returns to operation 225 and further vehicle tracking is performed.

[0045] In some embodiments, method 200 includes additional operations. For example, in some embodiments, method 200 includes generating instructions for controlling an autonomous vehicle based on estimated vehicle parameters. In some embodiments, at least one operation of method 200 is omitted. For example, in some embodiments, operation 245 is omitted and the estimation of hit points is not performed. In some embodiments, the order of operations of method 200 is adjusted. For example, in some embodiments, operation 215 is performed before operation 210.

[0046] Using method 200, DFOS data can be used to enhance the accuracy and precision of tracking a vehicle along a lane as compared to other techniques that cannot remove outlier hit points and cluster hit points. The improved determination of vehicle tracking helps improve the accuracy of traffic monitoring, navigation instructions, autonomous driving instructions, and other applications.

[0047] FIG. 3 is a schematic diagram of DAS system 100B with waterfall data 300 collected by a DAS system according to some embodiments. DAS system 100B is the same as DAS system 100B of FIG. 1B. Similar to FIG. 1B, the lane (not shown) in FIG. 3 includes two bridges as shown by waterfall data 300. Waterfall data 300 is preprocessed waterfall data.

[0048] Waterfall data 300 includes regions 302, 304, 306, 308, and 310. Regions 302, 306, and 310 include distinguishable lines indicative of vibrations created by vehicles crossing the lane. Regions 304 and 308 represent the bridges. Compared to regions 302, 306, and 310, regions 304 and 308 do not include distinguishable lines because the damped vibrations of the bridges obscure the detected vibrations of the vehicles crossing the bridges.

[0049] FIG. 4 is a diagram of filtered DFOS data 400 according to some embodiments. The filtered DFOS data 400 includes a region 410 that exhibits a higher vibration intensity within the estimated frequency range. Region 410 is likely to be a bridge along the lane. The filtered DFOS data 400 also includes a region 420 that exhibits a lower vibration intensity within the estimated frequency range. Region 420 is likely to be a non-bridge portion of the lane.

[0050] FIG. 5A is a diagram of a plot 500A of detected vehicle positions based on DFOS data according to some embodiments. Plot 500A includes hit points 505, missing data points 510, and seed points 515. Plot 500A includes this data in a distance-versus-time graph. The distance axis is a measure of how far the detected vibration is from a DAS such as DAS 112 (FIG. 1). Thus, the data in plot 500A shows that the vehicle is traveling towards the DAS as the distance decreases over time. Hit points 505 indicate that a vehicle is present at a distance from the DAS at a particular time. Missing data points 510 indicate the absence of expected hit points based on an analysis to track the vehicle. Seed points 515 indicate hit points 505 that were determined as initial seeds, for example using operation 220 (FIG. 2). The following description includes non-limiting examples of some implementations of method 200 (FIG. 2).

[0051] The hit point 505 is detected from the DFOS data, for example, using operations 205, 210, and 215 (Figure 2). Using the hit point 505, the seed point 515 is determined, for example, using operation 220 (Figure 2). Once the seed point 515 is determined, subsequent hit points 520 and 525 are identified, for example, using operation 225 (Figure 2). The subsequent hit points 520 and 525 are the points 505 of the hit points detected at a time after the seed point 515. In plot 500A, only the subsequent hit points 520 and 525 are labeled. However, those skilled in the art will recognize that plot 500A includes more subsequent hit points than the immediate subsequent hit points 520 and 525.

[0052] In the region 530 of plot 500A, a missing data point 535a is at a position where subsequent hit points are expected. Simultaneously with the missing data point 535a, a first hit point 535b and a second hit point 535c are detected. To determine whether either the first hit point 535b or the second hit point 535c is due to a vehicle associated with the seed point 515, a determination is made as to whether either the first hit point 535b or the second hit point 535c is an outlier hit point, for example, using operation 230 (Figure 2).

[0053] To determine whether the first hit point 535b is an outlier hit point, the position of the first hit point 535b is compared to the position of the missing data point 535a. That is, the positions of the expected subsequent hit points corresponding to the missing data point 535a and the first hit point 535b are analyzed to determine whether a reasonable movement of the vehicle can account for the difference between these two positions. The first hit point 535b is separated from the missing data point 535a by a distance D1. In some embodiments, the distance D1 is compared to a threshold value or is compared using another option discussed above with respect to operation 230 (FIG. 2). In the example of plot 500A, the first hit point 535b indicates that the vehicle would have to change directions rapidly twice in order to be detected by both a subsequent hit point 525 that occurs before the first hit point 535b and a third hit point 535d that occurs after the first hit point 535b. The likelihood of two rapid direction changes in such a short period of time is extremely low, and thus the first hit point 535b is determined to be an outlier hit point. As a result, even if the distance D1 is within the threshold value, the rapid direction change indicates that the plausibility of the first hit point 535b is low enough to indicate that the first hit point 535b is an outlier hit point. Accordingly, the first hit point 535b is not considered for the analysis of plot 500A with respect to the vehicle being tracked.

[0054] To determine whether the second hit point 535c is an outlier hit point, the position of the second hit point 535c is compared to the position of the missing data point 535a. That is, the position of the expected subsequent hit point corresponding to the missing data point 535a and the position of the second hit point 535c are analyzed to determine whether the reasonable movement of the vehicle can account for the difference between these two positions. The second hit point 535c is separated from the missing data point 535a by a distance D2. In some embodiments, the distance D2 is compared to a threshold value or using another option discussed above with respect to operation 230 (Figure 2). A large distance D2 indicates that it is highly unlikely that the second hit point 535c corresponds to the vehicle being tracked. For example, the change in speed of the vehicle to move from the subsequent hit point 525 to the second hit point 535c is very large, so the second hit point 535c cannot reasonably correspond to the same vehicle that was tracked from the seed point 515 through the subsequent hit points 520 and 525. Further, in the example of plot 500A, the second hit point 535c indicates that the vehicle would have to change direction rapidly twice in order to be detected by both the subsequent hit point 525 that occurs before the second hit point 535c and the third hit point 535d that occurs after the second hit point 535c. The likelihood of two rapid direction changes in such a short time is extremely low, so the second hit point 535c is further determined to be an outlier hit point. Accordingly, the second hit point 535c is not considered for the analysis of plot 500A with respect to the vehicle being tracked.

[0055] One of ordinary skill in the art will understand that the analysis described above for plot 500A is merely an example for assisting in the understanding of some embodiments of method 200 (Figure 2) or other methods of analyzing DFOS data, and that the analysis of DFOS data in this description is not limited to only the above example of analysis.

[0056] FIG. 5B is a diagram of a plot 500B of detected vehicle positions based on DFOS data, according to some embodiments. Plot 500B is similar to plot 500A. Some of the hit points and missing data points labeled in plot 500A are not labeled in plot 500B for clarity of the drawing. As explained above, since all the hit points detected simultaneously with the missing data points in region 530 were determined to be outlier hit points, the detected hit points were not identified as reasonable candidates to replace the missing data points in region 530.

[0057] In some embodiments where the identified subsequent hit points, such as subsequent hit points 520 and 525 (FIG. 5A), are not sufficient to perform clustering, the missing data points in region 530 are skipped. For example, if operation 235 (FIG. 2) indicates an insufficient number of hit points for clustering, method 200 returns to operation 225 (FIG. 2). However, since the missing data in region 530 is between subsequent hit point 540 and another subsequent hit point 545, in some embodiments, the missing data is skipped in the repeated execution of operation 225 (FIG. 2). Thus, the determination of whether a sufficient number of hit points are available for clustering, such as operation 235 (FIG. 2), is performed without considering the missing data points in region 530.

[0058] The missing data points in region 530 include the missing data points in a single time measurement. However, an analysis similar to that described above with respect to plot 500B and plot 500A (FIG. 5A) is applicable to region 550 that includes multiple consecutive missing data points. The multiple consecutive missing data points are two or more consecutive periods where no hit points were detected at the expected positions. In response to the determination that there are no suitable candidates to replace the missing data points in region 550, the identification of subsequent hit points skips from subsequent hit point 560 to subsequent hit point 565.

[0059] One of ordinary skill in the art will understand that the analysis of Plot 500B described above is merely an example to assist in understanding some embodiments of Method 200 (FIG. 2) or other methods of analyzing DFOS data, and that the analysis of DFOS data in this description is not limited to only the above analysis examples.

[0060] FIG. 6 is a diagram of a plot 600 of clustered vehicle positions based on DFOS data according to some embodiments. Plot 600 is similar to Plot 500A (FIG. 5A) and Plot 500B (FIG. 5B). Some of the hit points and missing data points labeled in Plot 500A or Plot 500B are not labeled in Plot 600 for clarity of the drawing. Plot 600 includes a plurality of clustered hit points 605 to form clusters 610. A vehicle parameter line 615 is determined based on the clustered hit points 605. The slope of the vehicle parameter line 615 can be used to determine the speed of the vehicle over the duration of the clustered hit points 605.

[0061] In some embodiments, Plot 600 shows DFOS data after operation 240 (FIG. 2). In some embodiments, Plot 600 shows DFOS data analyzed using a method other than Method 200 (FIG. 2). Clusters 610 do not include cluster hit points 605 for each detection time along the y-axis of Plot 600. For example, in some embodiments, clusters 610 do not include cluster hit points 605 where missing data 535a (FIG. 5A) was within Plot 500A. When a sufficient number of cluster hit points 605 are identified, clusters 610 are generated. For example, when operation 235 (FIG. 2) is satisfied, in some embodiments, clusters 610 are formed.

[0062] The vehicle parameter line 615 is determined based on the linear regression of the cluster hit points 605. In some embodiments, the vehicle parameter line 615 is determined based on an algorithm other than linear regression. In some embodiments, the vehicle parameter line 615 is determined using a trained NN. The vehicle parameter line 615 indicates the speed of the vehicle during the clustered hit points 605.

[0063] One of ordinary skill in the art will understand that the above-described analysis of plot 600 is merely illustrative to aid in the understanding of some embodiments of method 200 (FIG. 2) or other methods of analyzing DFOS data, and that the analysis of DFOS data in this description is not limited to only the above analysis examples.

[0064] FIG. 7 is a diagram of plots 700A and 700B for tracking a vehicle position along a lane according to some embodiments. Plot 700A is similar to plot 500A (FIG. 5A), but the DFOS data input to plot 700A is different compared to plot 500A. Plot 700B is similar to plot 600 (FIG. 6), but has a larger cluster 720 compared to plot 600.

[0065] Plot 700A includes a plurality of hit points 705 and a plurality of missing data points 710. The hit points 705 indicate that the detected vibration indicates that a vehicle is present at a distance from the DAS at a particular time. The missing data points 710 indicate the absence of predicted hit points based on the analysis for tracking the vehicle. The hit points 705 are detected from the DFOS data using, for example, operations 205, 210, and 215 (FIG. 2).

[0066] Plot 700B includes a plurality of clustered hit points 715 to form cluster 720. A vehicle parameter line 730 is determined based on the clustered hit points 715. The slope of the vehicle parameter line 730 can be used to determine the speed of the vehicle over the duration of the clustered hit points 715.

[0067] Plot 700B is generated by analyzing the DFOS data within plot 700A to track the vehicle. Hit points 705 are analyzed to determine which hit points accurately reflect the movement of the vehicle and which hit points are outlier hit points. Hit points 705 are clustered into clustered hit points 705 when a sufficient number of hit points 715 are identified. The analysis of cluster 720 can be used to determine the speed of the vehicle being tracked using the slope of vehicle parameter line 730. In some embodiments, plot 700B is generated by performing method 200 (Figure 2) on the DFOS data of plot 700A. In some embodiments, tracking methods other than method 200 (Figure 2) are utilized to generate plot 700B based on the DFOS data within plot 700A.

[0068] One of ordinary skill in the art will understand that the above-described analysis of plots 700A and 700B is merely illustrative to aid in the understanding of some embodiments of method 200 (Figure 2) or other methods of analyzing DFOS data, and that the analysis of DFOS data in this description is not limited to only the above examples of analysis.

[0069] FIG. 8 is a diagram of plots 800A and 800B for tracking multiple vehicles along a lane according to some embodiments. Similar to plots 700A and 700B (FIG. 7), plots 800A and 800B include hit points. In contrast to plots 700A and 700B (FIG. 7), plots 800A and 800B include hit points of multiple vehicles. Plots 800A and 800B do not include missing data points or outlier hit points for clarity of the drawing. However, one of ordinary skill in the art will recognize that it is within the scope of this description for plots of multiple vehicles to include missing data points and outlier hit points.

[0070] Plot 800A includes a first cluster 810 of hit points and a second cluster 820 of hit points. The first cluster 810 corresponds to hit points of a first vehicle. The second cluster 820 corresponds to hit points of a second vehicle. One of ordinary skill in the art will understand that it is within the scope of this description to track more than two vehicles using DFOS data. In some embodiments, the first cluster 810 and the second cluster 820 are determined using method 200 (FIG. 2). In some embodiments, at least one of the first cluster 810 or the second cluster 820 is determined using a method other than method 200 (FIG. 2). The first cluster 810 intersects the second cluster 820 at intersection 830. Intersection 830 is the location where hit points of multiple vehicles intersect in plot 800A and does not correspond to an intersection of the lane. In some embodiments, intersection 830 is generated by one vehicle overtaking another, by a vehicle changing lanes, or by another suitable situation. When the hit points for each vehicle are close, the difficulty of separately tracking each of the first and second vehicles after intersection 830 increases.

[0071] Utilizing a vehicle tracking method such as method 200 (FIG. 2) helps improve the tracking of multiple vehicles passing through intersection 830. Determining a first cluster 810 and determining a vehicle parameter line 815 of the first cluster 810 provides the speed of the first vehicle over the duration of the first cluster 810. Similarly, determining a second cluster 820 and determining a vehicle parameter line 825 of the second cluster 820 provides the speed of the second vehicle over the duration of the second cluster 820.

[0072] When the first cluster 810 and the second cluster 820 are determined on the first side of intersection 830, a seed point is identified on the second side of intersection 830, opposite the first side of intersection 830. One seed point is determined for each of the vehicles being tracked in plots 800A and 800B. A vehicle tracking method such as method 200 (FIG. 2) is used to determine clusters using the newly identified seed points on the second side of intersection 830. Plot 800B includes a third cluster 810' determined using one seed point on the second side of intersection 830. Plot 800B further includes a fourth cluster 820' determined using another seed point on the second side of intersection 830. A vehicle parameter line 815' is determined using the third cluster 810'. A vehicle parameter line 825' is determined using the fourth cluster 820'.

[0073] To track each of a plurality of vehicles passing through intersection 830, vehicle parameter line 815’ is compared with each of vehicle parameter line 815 and vehicle parameter line 825. Vehicle parameter line 815’ is determined to correspond to the vehicle associated with the vehicle parameter line having the gradient most similar to vehicle parameter line 815’. Similarly, vehicle parameter line 825’ is compared with each of vehicle parameter line 815 and vehicle parameter line 825 to determine which vehicle corresponds to vehicle parameter line 825’. In the case of plot 800B, vehicle parameter line 815’ has the gradient most similar to vehicle parameter line 815. Thus, vehicle parameter line 815’ is determined to correspond to the first vehicle. Similarly, vehicle parameter line 825’ has the gradient most similar to vehicle parameter line 825. Vehicle parameter line 825’ is determined to correspond to the second vehicle. In some embodiments, when a vehicle is determined to correspond to vehicle parameter lines on each side of intersection 830, all vehicle parameter lines corresponding to that vehicle are excluded from further comparison related to intersection 830 to reduce the processing load.

[0074] By applying a tracking method such as method 200 (FIG. 2) to the hit points on both sides of the intersection, multiple vehicles can be tracked with high precision and accuracy using DFOS data.

[0075] FIG. 9 is a flowchart of a method 900 for analyzing distributed fiber optic sensing (DFOS) data according to some embodiments. Method 900 can be used in combination with method 200 (FIG. 2). Method 900 can also be used independently of method 200 (FIG. 2). Method 900 can be used to implement functions as discussed above with respect to plot 500A (FIG. 5A), plot 500B (FIG. 5B), plot 600 (FIG. 6), plots 700A and 700B (FIG. 7), and plots 800A and 800B (FIG. 8). Method 900 can also be used to implement functions other than those discussed above with respect to various plots.

[0076] Method 900 is similar to method 200 (FIG. 2), and similar operations have the same reference numbers as method 200. Compared with method 200 (FIG. 2), method 900 includes operation 910 in which the tracked vehicle speed and vehicle position are updated. The update of the vehicle position is based on the latest detected hit point that is not an outlier hit point. In some embodiments, the latest detected hit point corresponds to the detected hit point that coincides with the estimated next hit point from operation 245. The update of the vehicle speed is based on the analysis of the clustered hit points as described above with respect to plot 600 (FIG. 6).

[0077] In some embodiments, method 900 includes additional operations. For example, in some embodiments, method 900 includes generating instructions for controlling an autonomous vehicle based on the estimated vehicle parameters. In some embodiments, at least one operation of method 900 is omitted. For example, in some embodiments, operation 245 is omitted and the estimation of the hit point is not performed. In some embodiments, the order of the operations of method 900 is adjusted. For example, in some embodiments, operation 215 is performed before operation 210.

[0078] Using method 900, DFOS data can be used to improve the accuracy and precision of tracking a vehicle along a lane compared to other techniques that cannot remove outlier hit points and cluster hit points. The improved determination of vehicle tracking helps to improve the accuracy of traffic monitoring, navigation instructions, autonomous driving instructions, and other applications.

[0079] Figure 10 is a block diagram of a system 1000 for analyzing DFOS data according to some embodiments. System 1000 includes a hardware processor 1002 and a non-transitory computer-readable storage medium 1004 encoded with, i.e., storing, a set of executable instructions, i.e., computer program code 1006. The computer-readable storage medium 1004 is also encoded with instructions 1007 for interfacing with external devices. Processor 1002 is electrically coupled to computer-readable storage medium 1004 via bus 1008. Processor 1002 is also electrically coupled to I / O interface 1010 by bus 1008. Network interface 1012 is also electrically connected to processor 1002 via bus 1008. Network interface 1012 is connected to network 1014 such that processor 1002 and computer-readable storage medium 1004 can be connected to external elements via network 1014. Processor 1002 is configured to execute the encoded computer program code 1006 within computer-readable storage medium 1004 to enable system 100 to be used to perform some or all of the operations described with respect to DAS system 100A (FIG. 1A), DAS system 100B (FIG. 1B), method 200 (FIG. 2), method 900 (FIG. 9), or another suitable system for analyzing DFOS data.

[0080] In some embodiments, processor 1002 is a central processing unit (CPU), a multiprocessor, a distributed processing system, an application specific integrated circuit (ASIC), and / or a suitable processing unit.

[0081] In some embodiments, the computer-readable storage medium 1004 is an electronic, magnetic, optical, electromagnetic, infrared, and / or semiconductor system (or apparatus or device). For example, the computer-readable storage medium 1004 includes semiconductor or solid state memory, magnetic tape, removable computer diskettes, random access memory (RAM), read-only memory (ROM), rigid magnetic disks, and / or optical disks. In some embodiments that use optical disks, the computer-readable storage medium 1004 includes compact disk read-only memory (CD-ROM), compact disk read / write (CD-R / W), and / or digital video disk (DVD).

[0082] In some embodiments, the storage medium 1004 stores computer program code 1006 configured to cause the system 1000 to perform some or all of the operations described with respect to the DAS system 100A (FIG. 1A), the DAS system 100B (FIG. 1B), the method 200 (FIG. 2), the method 900 (FIG. 9), or another suitable system for analyzing DFOS data. In some embodiments, the storage medium 1004 also stores the information necessary to perform some or all of the operations described with respect to the DAS system 100A (FIG. 1A), the DAS system 100B (FIG. 1B), the method 200 (FIG. 2), the method 900 (FIG. 9), or another suitable system for analyzing DFOS data, as well as the information generated during the performance of some or all of the operations described with respect to the DAS system 100A (FIG. 1A), the DAS system 100B (FIG. 1B), the method 200 (FIG. 2), the method 900 (FIG. 9), or another suitable system for analyzing DFOS data, such as sensor data parameters 1016, threshold parameters 1018, vehicle parameters 1020, hit point parameters 1022, and / or a set of executable instructions for performing some or all of the operations described with respect to the DAS system 100A (FIG. 1A), the DAS system 100B (FIG. 1B), the method 200 (FIG. 2), the method 900 (FIG. 9), or another suitable system for analyzing DFOS data.

[0083] In some embodiments, the memory medium 1004 stores instructions 1007 for interfacing with an external device. The instructions 1007 enable the processor 1002 to generate instructions readable by the external device to effectively perform some or all of the operations described with respect to the DAS system 100A (FIG. 1A), the DAS system 100B (FIG. 1B), the method 200 (FIG. 2), the method 900 (FIG. 9), or another suitable system for analyzing DFOS data.

[0084] The system 1000 includes an I / O interface 1010. The I / O interface 1010 is coupled to an external circuit. In some embodiments, the I / O interface 1010 includes a keyboard, a keypad, a mouse, a trackball, a trackpad, and / or cursor direction keys for communicating information and commands to the processor 1002.

[0085] The system 1000 also includes a network interface 1012 coupled to the processor 1002. The network interface 1012 enables the system 1000 to communicate with a network 1014 to which one or more other computer systems are connected. The network interface 1012 includes a wireless network interface such as BLUETOOTH®, WIFI, WIMAX, GPRS, or WCDMA®, or a wired network interface such as ETHERNET, USB, IEEE-1394. In some embodiments, some or all of the operations described with respect to the DAS system 100A (FIG. 1A), the DAS system 100B (FIG. 1B), the method 200 (FIG. 2), the method 900 (FIG. 9), or another suitable system for analyzing DFOS data are performed by two or more systems 1000, and information such as sensor data, bridge position, frequency range, and extra fiber portions is exchanged between different systems 1000 via the network 1014.

[0086] System 1000 is configured to receive information related to DFOS data via I / O interface 1010 or network interface 1012. The DFOS data is transferred to processor 1002 via bus 1008 for frequency range estimation and preprocessing and / or filtering. The frequency range is stored in computer-readable medium 1004 as frequency range parameter 1020. In some embodiments, the estimated frequency range parameter 1020 is received via I / O 1010 or network interface 1012. The preprocessed DFOS data is then stored in computer-readable medium 1004 as sensor data parameter 1016. Processor 1002 retrieves sensor data parameter 1016 from computer-readable medium 1004 and searches for hit points of hit point parameter 1022. Processor 1002 performs iterative noise removal of hit point parameter 1022, such as by using threshold parameter 1018. Processor 1002 performs analysis of the good clusters remaining after iterative noise removal to determine vehicle parameters of vehicle parameter 1020. In some embodiments, system 1000 can be used to implement a trained NN that can effectively perform some or all of the operations described with respect to DAS system 100A (FIG. 1A), DAS system 100B (FIG. 1B), method 200 (FIG. 2), method 900 (FIG. 9), or another suitable process for analyzing DFOS data.

[0087] Some or all of the above embodiments may also be described as follows, but are not limited thereto.

[0088] (Appendix 1) The vehicle tracking method includes receiving distributed fiber optic sensing (DFOS) data. The method further includes identifying a plurality of first hit points in the DFOS data, each corresponding to the position of a corresponding first vehicle at a detection time. The method further includes determining an initial seed point from among the identified plurality of first hit points. The method further includes determining whether any of the identified plurality of first hit points is an outlier hit point. The method further includes clustering the identified plurality of first hit points with any outlier hit points excluded to define a first cluster. The method further includes estimating a first vehicle parameter of the first vehicle based on the first cluster.

[0089] (Appendix 2) Determining whether any of the identified plurality of first hit points is an outlier hit point includes analyzing the problematic first hit point to determine whether the first vehicle could reasonably have traveled from the confirmed first hit point to the problematic first hit point, and determining that the problematic first hit point is the outlier hit point in response to a determination that the first vehicle could not reasonably have traveled from the confirmed first hit point to the problematic first hit point, the vehicle tracking method according to Appendix 1.

[0090] (Appendix 3) Analyzing the problematic first hit point includes analyzing the problematic first hit point based on the determined speed of the first vehicle, the vehicle tracking method according to Appendix 2.

[0091] (Appendix 4) Analyzing the first hit point that is the problem includes analyzing the first hit point that is the problem based on parameters of a lane on which the first vehicle travels, the vehicle tracking method according to appended note 2.

[0092] (Appended note 5) Further including determining whether a sufficient number of the plurality of first hit points have been identified for clustering, and clustering the identified plurality of first hit points is performed in response to the determination that a sufficient number of the plurality of first hit points have been identified, the vehicle tracking method according to appended note 1.

[0093] (Appended note 6) Further including identifying at least one subsequent first hit point from the DFOS data in response to the determination that an insufficient number of first hit points have been identified, the vehicle tracking method according to appended note 5.

[0094] (Appended note 7) Further including identifying a plurality of second hit points in the DFOS data, each corresponding to a position of a corresponding second vehicle, estimating second vehicle parameters of the second vehicle based on the identified plurality of second hit points, and plotting the identified plurality of first hit points and the identified plurality of second hit points, wherein the plotted identified plurality of first hit points intersect the plotted identified plurality of second hit points at an intersection, the vehicle tracking method according to appended note 1.

[0095] (Appended note 8) Further including tracking each of the first vehicle and the second vehicle crossing the intersection based on the first vehicle parameters and the second vehicle parameters, the vehicle tracking method according to appended note 7.

[0096] (Appended note 9) A vehicle tracking system comprises a non-transitory computer-readable medium configured to store instructions, and a processor connected to the non-transitory computer-readable medium. The processor is configured to execute the instructions to receive distributed fiber optic sensing (DFOS) data, identify a plurality of first hit points within the DFOS data, each corresponding to a position of a corresponding first vehicle at a detection time, determine an initial seed point from among the identified plurality of first hit points, determine whether any of the identified plurality of first hit points is an outlier hit point, cluster the identified plurality of first hit points with any outlier hit points excluded to define a first cluster, and estimate a first vehicle parameter of the first vehicle based on the first cluster.

[0097] (Appendix 10) The vehicle tracking system according to Appendix 9, wherein the processor is further configured to execute the instructions to analyze the problematic first hit point to determine whether the first vehicle could have reasonably traveled from the verified first hit point to the problematic first hit point, and in response to a determination that the first vehicle could not have reasonably traveled from the verified first hit point to the problematic first hit point, determine that the problematic first hit point is the outlier hit point, thereby determining whether any of the identified plurality of first hit points is the outlier hit point.

[0098] (Appendix 11) The vehicle tracking system according to Appendix 10, wherein the processor is further configured to execute the instructions to analyze the problematic first hit point based on the determined speed of the first vehicle.

[0099] (Appendix 12) The vehicle tracking system according to appended claim 10, wherein the processor is configured to execute the instructions for analyzing a first hit point that is a problem based on parameters of a lane on which the first vehicle travels.

[0100] (Appended claim 13) The vehicle tracking system according to appended claim 9, wherein the processor is configured to execute the instructions for determining whether a sufficient number of the plurality of first hit points have been identified for clustering, and clustering the identified plurality of first hit points is executed in response to a determination that the sufficient number of the plurality of first hit points have been identified.

[0101] (Appended claim 14) The vehicle tracking system according to appended claim 13, wherein the processor is configured to execute the instructions for identifying at least one subsequent first hit point from the DFOS data in response to a determination that an insufficient number of first hit points have been identified.

[0102] (Appended claim 15) The vehicle tracking system according to appended claim 9, wherein the processor is further configured to execute the instructions for identifying a plurality of second hit points in the DFOS data, each corresponding to a position of a corresponding second vehicle, estimating second vehicle parameters of the second vehicle based on the identified plurality of second hit points, and plotting the identified plurality of first hit points and the identified plurality of second hit points, and the plotted identified plurality of first hit points intersect at an intersection with the plotted identified plurality of second hit points.

[0103] (Appended claim 16) The vehicle tracking system according to appended claim 15, wherein the processor is further configured to execute the instructions for tracking each of the first vehicle and the second vehicle crossing the intersection based on the first vehicle parameter and the second vehicle parameter.

[0104] (Appended claim 17) The program causes the processor to receive distributed fiber optic sensing (DFOS) data, identify a plurality of first hit points in the DFOS data, each corresponding to the position of a corresponding first vehicle at the detection time, determine an initial seed point from among the identified plurality of first hit points, determine whether any of the identified plurality of first hit points is an outlier hit point, cluster the identified plurality of first hit points with any outlier hit points excluded to define a first cluster, and estimate a first vehicle parameter of the first vehicle based on the first cluster.

[0105] (Appended claim 18) The program according to appended claim 17, wherein the processor is caused to determine whether any of the identified first hit points is the outlier hit point by analyzing the problematic first hit point to determine whether the first vehicle could reasonably have traveled from the confirmed first hit point to the problematic first hit point, and determining that the problematic first hit point is the outlier hit point in response to a determination that the first vehicle could not reasonably have traveled from the confirmed first hit point to the problematic first hit point.

[0106] (Appended claim 19) Causing the processor to determine whether a sufficient number of the plurality of first hit points have been identified for clustering, and clustering the identified plurality of first hit points, the program according to Appendix 17, which is executed in response to the determination that the sufficient number of the plurality of first hit points have been identified.

[0107] (Appendix 20) Identifying a plurality of second hit points in the DFOS data, each corresponding to the position of a corresponding second vehicle; estimating second vehicle parameters of the second vehicle based on the identified plurality of second hit points; plotting the identified plurality of first hit points and the identified plurality of second hit points, wherein the plotted identified first hit points intersect the plotted identified second hit points at intersections; and causing the processor to track each of the first vehicle and the second vehicle crossing the intersection based on the first vehicle parameters and the second vehicle parameters, the program according to Appendix 17.

[0108] The foregoing outlines the features of several embodiments so that those skilled in the art can better understand aspects of the present disclosure. Those skilled in the art should understand that they can readily use the present disclosure as a basis for designing or modifying other processes and structures for performing the same purposes and / or achieving the same advantages as the embodiments introduced herein. Those skilled in the art should also understand that such equivalent configurations do not depart from the spirit and scope of the present disclosure, and various changes, substitutions, and alterations may be made herein without departing from the spirit and scope of the present disclosure.

[0109] This application claims priority to U.S. Patent Application No. 18 / 537,669, filed on December 12, 2023, and incorporates the entire disclosure thereof herein.

Description of Reference Numerals

[0110] 100A, 100B Distributed Acoustic Sensor (DAS) systems, DAS systems 111 Traffic monitoring device 121 Optical fiber 130, 130A, 130B Road 140, 150, 150A, 150B Vehicle 160A, 160B Bridge 170 Fiber 300 Waterfall data 302, 304, 306, 308, 310 Region 400 DFOS data 410, 420 Region 500A, 500B Plot 505 Hit point 510 Missing data point 515 Seed point 520, 525 Subsequent hit points 530 Region 535a Missing data point 535b Hit point 535c Hit point 535d Hit point 540 Subsequent hit point 545 Another subsequent hit point 550 Region 560 Subsequent hit point 565 Subsequent hit point 600 Plot 605 Cluster hit point 610 Cluster 615 Vehicle parameter line 700A Plot 700B Plot 705 Hit point 710 Missing data point 715 Clustered hit point 720 Cluster 730 Vehicle parameter line 800A Plot 800B Plot 810 Cluster 810’ Cluster 815 Vehicle Parameter Line 815’ Vehicle Parameter Line 820 Cluster 820’ Cluster 825 Vehicle Parameter Line 825’ Vehicle Parameter Line 830 Intersection 1000 System 1002 Hardware Processor 1004 Non - Transitory Computer - Readable Memory Medium 1006 Computer Program Code 1007 Instruction 1008 Bus 1010 I / O Interface 1012 Network Interface 1014 Network 1016 Sensor Data Parameter 1018 Threshold Parameter 1020 Vehicle Parameter, Frequency Range Parameter, Estimated Frequency Range Parameter 1022 Hit Point Parameter D1 Distance D2 Distance

Claims

1. receiving distributed fiber optic sensing (DFOS) data; identifying a plurality of first hit points within the DFOS data, each of which corresponds to a position of a corresponding first vehicle at a time of detection; determining an initial seed point from among the identified first plurality of hit points; determining whether any of the identified first plurality of hit points is an outlier hit point; clustering the identified first hit points, excluding any outlier hit points, to define first clusters; estimating a first vehicle parameter of the first vehicle based on the first cluster.

2. Determining whether any of the identified plurality of first hit points is an outlier hit point includes: analyzing the first problematic hit point to determine whether the first vehicle could have reasonably traveled from the identified first hit point to the problematic first hit point; and determining that the first vehicle could not reasonably have traveled from the confirmed first hit point to the problematic first hit point, the method comprising: determining that the first vehicle could not reasonably have traveled from the confirmed first hit point to the problematic first hit point.

3. 3. The vehicle tracking method of claim 2, wherein analyzing the first hit point in question comprises analyzing the first hit point in question based on a determined speed of the first vehicle.

4. 3. The vehicle tracking method of claim 2, wherein analyzing the first hit point in question includes analyzing the first hit point in question based on a parameter of a roadway along which the first vehicle is traveling.

5. 2. The vehicle tracking method of claim 1, further comprising: determining whether a sufficient number of the plurality of first hit points have been identified for clustering, and wherein clustering the identified plurality of first hit points is performed in response to determining that a sufficient number of the plurality of first hit points have been identified.

6. The vehicle tracking method of claim 5 , further comprising, in response to determining that an insufficient number of first hit points have been identified, identifying at least one subsequent first hit point from the DFOS data.

7. identifying a plurality of second hit points within the DFOS data, each of which corresponds to a location of a corresponding second vehicle; estimating a second vehicle parameter of the second vehicle based on the identified second plurality of hit points; plotting the identified first plurality of hit points and the identified second plurality of hit points; The vehicle tracking method of claim 1 , wherein the plotted identified first plurality of hit points intersect with the plotted identified second plurality of hit points at an intersection.

8. The method of claim 7 , further comprising: tracking each of the first and second vehicles crossing the intersection based on the first and second vehicle parameters.

9. a non-transitory computer-readable medium configured to store instructions; a processor coupled to the non-transitory computer-readable medium; The processor, receiving distributed fiber optic sensing (DFOS) data; identifying a plurality of first hit points within the DFOS data, each of which corresponds to a position of a corresponding first vehicle at a time of detection; determining an initial seed point from among the identified first plurality of hit points; determining whether any of the identified first plurality of hit points is an outlier hit point; clustering the identified first hit points, excluding any outlier hit points, to define first clusters; and estimating a first vehicle parameter of the first vehicle based on the first cluster.

10. receiving distributed fiber optic sensing (DFOS) data; identifying a plurality of first hit points within the DFOS data, each of which corresponds to a position of a corresponding first vehicle at a time of detection; determining an initial seed point from among the identified first plurality of hit points; determining whether any of the identified first plurality of hit points is an outlier hit point; clustering the identified first hit points, excluding any outlier hit points, to define first clusters; estimating a first vehicle parameter of the first vehicle based on the first cluster; A program that causes a processor to perform the following:

Images