Apparatus and method for predicting traffic information

Abstract

The present invention relates to an apparatus and method for predicting traffic information. The traffic information prediction apparatus can include a data calculation device for deriving inter-vehicle distance, headway, and vehicle density using a plurality of sensors installed on a vehicle, and a prediction device for deriving driving speed data corresponding to the vehicle density and predicting traffic information.

Description

[0001] Cross-references to related applications

[0002] This application claims priority and benefit to Korean Patent Application No. 10-2020-0139673, filed on October 26, 2020, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This invention relates to an apparatus and method for predicting traffic information, and more specifically, to an apparatus and method for predicting traffic information based on density estimation technology. Background Technology

[0004] The traffic information currently provided corresponds to information based on speed predictions from past patterns. That is, based on the assumption that similar speeds will occur within the same time interval of the day, the current traffic information (e.g., speed information) is derived using existing speed patterns.

[0005] Using traffic information from the same day and time period of the past week, for example, using the speeds from 9:00 to 9:05 a.m. on Monday, February 3 and Monday, February 10, to predict the speeds from 9:00 to 9:05 a.m. on Monday, March 2.

[0006] However, when speeds are based on past patterns, anomalies may emerge at specific points in time. For example, variables such as weather and seasons may change, and traffic volume may vary over time. This can lead to inappropriate data being used for speed prediction. In other words, the assumption that speeds will remain similar within the same time interval may increase the probability of errors in predicting traffic information.

[0007] In one example, a study is also conducted on whether changes in the operational quantities of vehicle detectors (hereinafter referred to as detectors) affect traffic in order to predict traffic information. In this case, congestion time can be macroscopically predicted based on the GPS occurrence time, but there are limitations in microscopic predictions due to the limited number of detector samples, such as speed prediction for each time interval of the link unit (the road to be predicted).

[0008] Therefore, when using past pattern speeds, it may be necessary to use speeds in similar traffic conditions rather than simply speeds over the same time interval. In one example, density, corresponding to vehicle density, is considered the most objective and effective measure for determining traffic conditions in traffic engineering.

[0009] Research related to density estimation has the following advantages: taking images of limited road sections to calculate the average number of vehicles in the corresponding sections, and identifying the total number of vehicles on the actual road through image taking, but there are limitations in ensuring data when constant density data are required, such as traffic prediction.

[0010] Therefore, it can be necessary to ensure additional variables available for speed prediction and to develop a density-based travel speed prediction technique. SUMMARY

[0011] An aspect of the present application provides an apparatus and method for predicting traffic information, which can objectively determine traffic conditions through inter-vehicle distance data between vehicles and generate an additional variable (density) that can be used for speed prediction.

[0012] Another aspect of the present application provides an apparatus and method for predicting traffic information, which can make reliable density estimation based on statistical techniques.

[0013] Another aspect of the present application provides an apparatus and method for predicting traffic information, which can derive travel speed under the same traffic conditions by providing mode speed based on density.

[0014] Another aspect of the present application provides an apparatus and method for predicting traffic information, which can determine continuity of traffic conditions based on current density states and develop a prediction model for future traffic information.

[0015] The technical problems to be solved by the present application are not limited to the above problems, and any other technical problems not mentioned herein will be clearly understood by those skilled in the art from the following description.

[0016] According to one aspect of the present application, an apparatus for predicting traffic information includes a data calculation device that derives inter-vehicle distance, headway, and vehicle density using a plurality of sensors installed on a vehicle, and a prediction device that derives travel speed data corresponding to the vehicle density and predicts traffic information.

[0017] In one embodiment, the data calculation device can include an inter-vehicle distance acquisition device that acquires respective inter-vehicle distances to a preceding vehicle and a following vehicle from front and rear sensors of a vehicle, a headway calculation device that calculates headway based on the inter-vehicle distance and the length of the preceding or following vehicle acquired by a vehicle camera, and a density calculation device that calculates vehicle density based on the headway.

[0018] In one embodiment, the headway calculation device can calculate two headway data of three detectors from one detector.

[0019] In one embodiment, the headway calculation device can calculate the headway between the vehicle and the front vehicle by adding the vehicle-to-front-vehicle headway to the vehicle length of the front vehicle, and calculate the headway between the vehicle and the rear vehicle by adding the vehicle-to-rear-vehicle headway to the vehicle length of the vehicle.

[0020] In one embodiment, the density calculation device can estimate the average headway of the population based on N headway samples, and calculate the vehicle density of the specific road section.

[0021] In one embodiment, the density calculation device can calculate the vehicle density of the specific road section based on the following mathematical formula.

[0022] [mathematical formula 1]

[0023]

[0024] where K is the vehicle density of the specific road section, L is the length of the specific road section, a is the minimum value of the headway μ of the population, and b is the maximum value of the headway μ of the population.

[0025] In one embodiment, the prediction device can include a speed derivation device that adds density data on the vehicle density to speed data corresponding to a specific time interval, constructs density-speed pattern data representing at least one speed data corresponding to the vehicle density, and derives a representative speed representing at least one speed data for each vehicle density.

[0026] In one embodiment, the prediction device can include a speed prediction device that predicts the vehicle speed at the current time point based on the density-speed pattern data, the pre-constructed time-density pattern, and the pre-constructed time-speed pattern.

[0027] In one embodiment, the speed prediction device can derive a first vehicle speed corresponding to a first real-time density of a first time point that does not conform to the time-density pattern based on the density-speed pattern data when the real-time density measured in real time does not conform to the time-density pattern, derive a second time point corresponding to the first vehicle speed from the time-speed pattern, and shift the time-speed pattern by calculating the difference between the second time point and the first time point.

[0028] In one embodiment, the speed prediction device can include a deep learning modeling algorithm for performing determination of the relationship between the vehicle density and the vehicle speed.

[0029] In accordance with another aspect of the present application, a method for predicting traffic information includes a data calculation operation of deriving inter-vehicle distance, headway, and vehicle density data using a plurality of sensors installed on a vehicle, and a traffic information prediction operation of deriving driving speed data corresponding to the vehicle density and predicting traffic information. BRIEF DESCRIPTION OF DRAWINGS

[0030] The above and other objects, features and advantages of the present application will be more apparent from the following detailed description taken in conjunction with the accompanying drawings, in which:

[0031] Figure 1 is a control block diagram of a traffic information prediction apparatus according to an embodiment of the present application;

[0032] Figure 2 is a schematic diagram for illustrating inter-vehicle distance according to an embodiment of the present application;

[0033] Figure 3A and Figure 3B is a schematic diagram for illustrating headway according to an embodiment of the present application;

[0034] Figure 4 is a graph illustrating a t-distribution based statistic according to an example;

[0035] Figure 5 is a schematic diagram for illustrating a speed-density relationship according to an embodiment of the present application;

[0036] Figure 6A , Figure 6B , Figure 6C and Figure 6D is a graph illustrating a method of predicting speed based on density according to an embodiment of the present application; and

[0037] Figure 7 is a control flow diagram illustrating a traffic information prediction method according to an embodiment of the present application. DETAILED DESCRIPTION

[0038] Some embodiments of the present application will be described below in detail with reference to the accompanying drawings. In adding reference numerals to components in each drawing, it should be noted that the same or equivalent components are designated by the same reference numerals even though they are shown on different drawings. Further, in describing some embodiments of the present application, detailed description of related known configurations or functions will be omitted when it is determined that the same would cause obscuring of the understanding of embodiments of the present application.

[0039] In describing some embodiments of the present application, terms such as first, second, A, B, (a), (b), etc. can be used. These terms are used only to distinguish one component from another component, and the terms do not limit the nature, order or sequence of the components. Unless otherwise indicated in the specification, all terms used herein, including technical and scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

[0040] Hereinafter, a traffic information prediction device according to some embodiments of the present application will be described with reference to the accompanying drawings. Figures 1 to 7 Embodiments of the present application will be described in detail.

[0041] Figure 1 is a control block diagram of a traffic information prediction device according to some embodiments of the present application.

[0042] As described above, in the case of predicting a speed using a past pattern speed, a traffic environment can change according to the characteristics of each period, and the assumption that a similar speed will be maintained in the same time interval becomes a factor that increases the probability of an error occurring. Therefore, in traffic engineering, it is necessary to use density, which is a new variable that most objectively determines a traffic situation.

[0043] The traffic information prediction device according to the present application can estimate a vehicle density based on a vehicle sensing technique and image data, and improve the prediction accuracy of traffic information based on the estimated vehicle density. As illustrated, the traffic information prediction device according to some embodiments of the present application can include a data calculation device 100 and a prediction device 200 that predicts traffic information using information calculated by the data calculation device 100. Such a traffic information prediction device can communicate with sensors of a vehicle, a camera attached to the inside or outside of the vehicle, or the like, or can communicate with a control module that controls these sensors in a wired or wireless manner. The traffic information prediction device can be implemented as an external server that can communicate with a vehicle, or can be implemented as a terminal on which an application program capable of implementing the prediction method described below is executed.

[0044] The data calculation device 100 can derive a vehicle spacing, a headway, and a vehicle density using a plurality of sensors mounted on a vehicle, and for this purpose, the data calculation device 100 can include a vehicle spacing acquisition device 110, a headway calculation device 120, and a density calculation device 130.

[0045] The prediction device 200 can derive driving speed data corresponding to the vehicle density and predict traffic information, such as a speed. For this purpose, the prediction device 200 can include a speed derivation device 210 and a speed prediction device 220.

[0046] For convenience of description, components are distinguished from each other in function. One or more components can be implemented as one module or chip, and can transmit / receive data through electronic communication between physical components that implement functions.

[0047] Hereinafter, a speed prediction method based on density will be described with reference to Figures 2 to 6D

[0048] Figure 2 is a schematic diagram for illustrating inter-vehicle distance of some embodiments of the present application.

[0049] The inter-vehicle distance acquisition device 110 can acquire respective inter-vehicle distances to the front vehicle and the rear vehicle from the front and rear sensors of the vehicle. The inter-vehicle distance acquisition device 110 can acquire information on the inter-vehicle distance from the vehicle sensors, and can be implemented as the vehicle sensors themselves or a component including the vehicle sensors.

[0050] As shown in Figure 2 , in the case of the vehicle 1, the inter-vehicle distance to the front vehicle is 60 m, and the inter-vehicle distance to the rear vehicle is 50 m. In the case of the vehicle 2, the inter-vehicle distance to the front vehicle is 40 m, and the inter-vehicle distance to the rear vehicle is 30 m.

[0051] In some embodiments of the present application, the distance information from one vehicle (i.e., one probe) to the front vehicle and the rear vehicle can be acquired. That is, the state of three vehicles can be identified by one inter-vehicle distance information. Such an inter-vehicle distance can be used to derive a headway. The vehicle located in the middle of the three vehicles (i.e., the vehicle that provides information on the inter-vehicle distance) can be referred to as a reference vehicle.

[0052] According to another example, information on the inter-vehicle distance can be acquired from the vehicle sensors of one of the two vehicles, or information on the inter-vehicle distance can be acquired from a single vehicle. When the inter-vehicle distance information derived from multiple sensors overlaps, an average value can be derived as the inter-vehicle distance.

[0053] Figure 3A and Figure 3B is a schematic diagram for illustrating headway of some embodiments of the present application.

[0054] The headway calculation device 120 in the data calculation device 100 can calculate the headway based on the length of the front vehicle or the rear vehicle (not shown in the figure) acquired by the camera of the vehicle and the inter-vehicle distance.

[0055] As shown in Figure 3A , the camera of the vehicle can photograph the front vehicle, and the camera or the headway calculation device 120 can derive the length of the front vehicle or the rear vehicle from the photographed image. ​

[0056] When the vehicle length of the preceding vehicle is derived as described above, the headway calculation device 120 can derive the headway from the reference vehicle to the preceding vehicle and the headway from the reference vehicle to the following vehicle using the vehicle distance between the reference vehicle and the preceding vehicle and the vehicle distance between the reference vehicle and the following vehicle.

[0057] As Figure 3B illustrated, the headway between the reference vehicle and the preceding vehicle can be derived as the sum of the length of the preceding vehicle or the following vehicle and the vehicle distance from the reference vehicle to the preceding vehicle, and the headway between the reference vehicle and the following vehicle can be derived as the sum of the length of the reference vehicle and the vehicle distance from the reference vehicle to the following vehicle. The length of the reference vehicle can be default information of the vehicle itself that does not need to be acquired by the sensor. Such a headway can be used as direct data for deriving the vehicle density within a certain road section.

[0058] Also, in the case of the headway, a plurality of information associated with the preceding vehicle and the following vehicle can be acquired using information derived from one reference vehicle. That is, information corresponding to three probes is acquired from one probe, so that N(n x 2) headway information of n x 3 vehicles in the overall vehicle can be acquired using n probes.

[0059] When the headway is calculated as described above, the density calculation device 130 derives the vehicle density based on the headway. The density calculation device 130 estimates the overall average headway using N headway samples, and estimates the vehicle density. The vehicle density can be expressed as a Service level or LOS.

[0060] When the number of samples of the headway is N, the average of N headway samples is E(x), the standard deviation of N headway samples is s, and the average headway μ of the overall vehicle can be expressed as a statistic T that follows a t-distribution. The statistic T can be expressed in a mathematical formula as follows.

[0061] [mathematical formula 1]

[0062]

[0063] Mathematical formula 1 can be expressed as a graph describing a statistic that follows a t-distribution, as Figure 4 illustrated.

[0064] When the degree of freedom is N-1 and the limit value of the 95% confidence level is α, T can be expressed as mathematical formula 2 with respect to α. In this regard, the average headway μ of the overall vehicle can be expressed as mathematical formula 3, which ranges from a to b.

[0065] [mathematical formula 2]

[0066] -α ≤ T ≤ α

[0067] [Math. 3]

[0068] a ≤ μ ≤ b

[0069] When the average headway μ of the population is derived as shown in Math. 3, the density calculation device 130 can derive the density K of a specific road section having a specific length.

[0070] [Math. 4]

[0071]

[0072] L denotes the length of the specific road section, a denotes the minimum value of the average headway μ of the population, and b denotes the maximum value of the average headway μ of the population.

[0073] When the vehicle density of the specific road section is derived as shown in Math. 4, the vehicle density of a plurality of road sections can be estimated as shown in the following table. Table 1 shows the service level corresponding to the vehicle density.

[0074] [Table 1]

[0075]

[0076] A different service level can be derived corresponding to the vehicle density, and as can be seen from Table 1, the traffic volume decreases as the density decreases.

[0077] Although the density index is the most objective measure for identifying the traffic condition, there is a limitation that cannot be directly collected at the road site. However, in some embodiments of the present application, the vehicle density can be easily measured in real time based on the headway.

[0078] When the vehicle density is thus estimated, the prediction device 200 predicts the traffic information, i.e., the speed, using the estimated vehicle density.

[0079] To this end, the speed derivation device 210 adds the density data on the vehicle density to the speed data corresponding to a specific time zone, constructs density-speed pattern data representing at least one speed data corresponding to the vehicle density, and derives a representative speed representing the at least one speed data for each vehicle density.

[0080] As shown in Table 2, the speed derivation device 210 can generate the speed and density data for the same day and time zone in the past one week for a specific road section.

[0081] [Table 2]

[0082]

[0083]

[0084] In addition, the speed derivation device 210 can construct density- speed pattern data, in addition to the existing time-speed pattern as shown in Table 3, and calculate a representative speed for each service level. The representative speed can be calculated as an average of speeds belonging to a specific service level.

[0085] [Table 3]

[0086]

[0087] As shown in Table 3, based on the data constructed by the speed derivation device 210, a speed corresponding to density can be calculated instead of the existing speed for each time interval, so that a speed based on vehicle density can be predicted.

[0088] The density-speed pattern data of Table 3 can be graphically represented.

[0089] Figure 5 is a graph showing a speed-density relationship for some embodiments of the present application.

[0090] As shown in the graph, when the density is service level D, the distance to the preceding vehicle can be 60 m, and when the density is service level B, the distance to the preceding vehicle can be 100 m. In the case of different densities as described above, when the vehicle speed is 60 Km / h, the traffic situation can be different even though the speed is the same. Since the speed deviation is large and the speed difference is generated based on the driver's tendency, it is not possible to determine the traffic situation only with speed information. However, even if the driver travels at the same speed, the traffic situation can be identified based on the vehicle density, and a representative speed corresponding to the density can be derived for a specific road section.

[0091] Figure 6A 、 Figure 6B 、 Figure 6C and Figure 6D is a graph showing a method of predicting a speed based on density for some embodiments of the present application.

[0092] The speed prediction device 220 can more accurately predict the speed in a specific time interval by correcting the existing speed using the representative speed calculated by the speed derivation device 210. The speed prediction device 220 can predict the vehicle speed at the current time point based on the pre-constructed time-density pattern, the pre-constructed time-speed pattern, and the density-speed pattern data.

[0093] Figure 6A shows a time-speed pattern graph for the same time as the existing one, which uses the pattern speed for the same time interval, so that an error as large as the difference between the real-time speed and the past pattern speed can occur, as shown by the diagonal portion in Figure 6A .

[0094] To prevent this, the speed prediction device 220 can derive a real-time time point T1 at which the density is different from the density of the past pattern and a density K1 corresponding to the real-time time point T1 from the pre-constructed time-density pattern in Figure 6B As illustrated, it can be seen that the density based on the past pattern is 19 at the real-time time point T1, but the density is actually 25.

[0095] The speed prediction device 220 can derive a speed V1 corresponding to the density 25 using a density-speed pattern graph as illustrated in Figure 6C As illustrated in the density-speed graph in Figure 6C As illustrated in the density-speed graph in

[0096] When the speed corresponding to the real-time density is derived, the speed prediction device 220 can derive a modified time point corresponding to the speed derived from the time-speed pattern graph illustrated in Figure 6D by shifting the time-speed pattern graph by the difference ΔT between the current time point and the modified time point. Since the time-speed pattern graph is shifted by ΔT, the existing error can be minimized. As can be seen, the speed at the time point at which the density is different from the existing density is corrected from 56.7 to 43.8.

[0097] In summary, when the real-time density measured in real time does not conform to the time-density pattern, the speed prediction device 220 can derive a first vehicle speed V1 corresponding to a first real-time density K1 of a first time point T1 that does not conform to the time-density pattern based on the density-speed pattern data, derive a second time point T2 corresponding to the first vehicle speed V1 from the time-speed pattern, and shift the time-speed pattern by the difference between the second time point T2 and the first time point T1. Since the graph is moved to a time corresponding to a speed reflecting the current vehicle density, more accurate speed prediction is possible.

[0098] According to an example, such a speed prediction device 220 can include an algorithm for performing deep learning modeling that identifies a relationship between vehicle density and vehicle speed. That is, the speed prediction device 220 can be implemented with a deep learning algorithm that can correct the time-speed pattern by identifying a relationship between density and speed.

[0099] Figure 7 is a control flow diagram illustrating a traffic information prediction method of some embodiments of the present application. In conjunction with Figure 7The traffic information prediction method of some embodiments of the present application is summarized as follows.

[0100] First, the data computing device 100 can derive the vehicle distance, the headway, and the vehicle density using a plurality of sensors installed on the vehicle (S710).

[0101] The vehicle distance acquisition device 110 can acquire the respective vehicle distances to the front and rear vehicles from the front and rear sensors of the vehicle, and the headway computing device 120 can calculate the headways to the front and rear vehicles based on the lengths of the front or rear vehicles and the vehicle distances acquired by the camera of the vehicle.

[0102] The headway computing device 120 can calculate the headway between the reference vehicle and the front vehicle by adding the vehicle distance to the front vehicle to the vehicle length of the front vehicle, and can calculate the headway between the reference vehicle and the rear vehicle by adding the vehicle distance to the rear vehicle to the vehicle length of the reference vehicle.

[0103] In some embodiments of the present application, two vehicle distance data and two headway data of three probes are calculated from one probe.

[0104] Thereafter, the density computing device 130 can calculate the vehicle density based on the headways.

[0105] Specifically, the density computing device 130 estimates the average headway of the population based on N headway samples, and calculates the vehicle density of a specific road section.

[0106] When the vehicle density is derived, the speed deriving device 210 can add the density data on the vehicle density to the speed data corresponding to a specific time interval, construct the density-speed pattern data representing at least one speed data corresponding to the vehicle density, and derive a representative speed representing at least one speed data per vehicle density (S720).

[0107] The speed predicting device 220 can predict the vehicle speed at the current time point based on the pre-constructed time-density pattern, the pre-constructed time-speed pattern, and the density-speed pattern data (S730).

[0108] When the real-time density measured in real time does not conform to the time-density pattern, the speed predicting device 220 can derive a first vehicle speed corresponding to a first real-time density of a first time point not conforming to the time-density pattern based on the density-speed pattern data. Thereafter, the speed predicting device 220 can derive a second time point corresponding to the first vehicle speed from the time-speed pattern, and shift the time-speed pattern by the difference between the second time point and the first time point, thereby predicting the speed reflecting the density.

[0109] As described above, the present application summarizes the possibility of generating density data capable of predicting traffic conditions through vehicle sensing technology, and proposes a method of overcoming the limitations of current traffic information prediction based on the generated density. To this end, density data is constructed based on vehicle detector data of a vehicle front / rear sensor, a pattern speed with the same density is searched for, and pattern correction logic is developed and applied.

[0110] The above description is merely an illustration of the technical idea of the present application, and those skilled in the art can make various modifications and changes without departing from the essential characteristics of the present application.

[0111] Therefore, some embodiments of the present application are not intended to limit the technical idea of the present application but are for the purpose of illustrating the present application, and the scope of the technical idea of the present application is not limited by the embodiments of the present application. The scope of the present application should be understood to be included in the scope of the appended claims, and all technical ideas falling within the scope of the claims should be understood to be included in the scope of the present application.

[0112] In some embodiments of the present application, an apparatus and method for predicting traffic information are provided, which are capable of objectively determining traffic conditions through inter-vehicle distance data between vehicles and generating an additional variable (density) that can be used for speed prediction.

[0113] Further, in some embodiments of the present application, an apparatus and method for predicting traffic information are provided, which are capable of making reliable density estimation based on statistical techniques.

[0114] Further, in some embodiments of the present application, an apparatus and method for predicting traffic information are provided, which are capable of deriving driving speed under the same traffic conditions by providing pattern speed based on density.

[0115] Further, in some embodiments of the present application, an apparatus and method for predicting traffic information are provided, which are capable of determining continuity of traffic conditions based on current density states and developing a prediction model for future traffic information.

[0116] Therefore, a deep learning model can be utilized to recognize the relationship between density and speed, rather than through a simple calculation of driving speed at the same density, and a technique of predicting driving speed based on real-time density data can be improved with the objectivity of density for each traffic condition.

[0117] Further, various effects identified directly or indirectly through the present document can be provided.

[0118] In the foregoing, although the present application has been described by way of some embodiments and drawings of the present application, the present application is not limited thereto, but can be variously modified and changed by those skilled in the art to which the present application pertains without departing from the spirit and scope of the present application claimed by the appended claims.

Claims

1. A device for predicting traffic information, comprising: a data computing device configured to derive vehicle headway, vehicle headway, and vehicle density using a plurality of sensors installed on a vehicle; and a prediction device configured to: derive travel speed data corresponding to vehicle density; predict traffic information, wherein the prediction device further comprises: a speed deriving device configured to: add density data on vehicle density to speed data corresponding to a specific time interval; construct density-speed pattern data representing at least one speed data corresponding to vehicle density; and derive a representative speed representing at least one speed data for each vehicle density, and a speed prediction device configured to predict vehicle speed at a current time point based on density-speed pattern data, a pre-constructed time-density pattern, and a pre-constructed time-speed pattern, the speed prediction device is further configured to: when a real-time density measured in real time does not conform to the time-density pattern, derive a first vehicle speed corresponding to a first real-time density at a first time point that does not conform to the time-density pattern based on the density-speed pattern data; derive a second time point corresponding to the first vehicle speed from the time-speed pattern; and translate the time-speed pattern by calculating a difference between the second time point and the first time point.

2. The apparatus for predicting traffic information according to claim 1, wherein, the data computing device further comprises: a vehicle headway obtaining device configured to obtain respective vehicle headways to a preceding vehicle and a following vehicle from a front sensor of the vehicle and a rear sensor of the vehicle; a vehicle headway calculating device configured to calculate vehicle headway based on vehicle headway and a length of the preceding vehicle or the following vehicle obtained by a camera of the vehicle; a density calculating device configured to calculate vehicle density based on vehicle headway.

3. The apparatus for predicting traffic information according to claim 2, wherein, the vehicle headway calculating device is configured to calculate two vehicle headway data of three probes from one probe.

4. The apparatus for predicting traffic information according to claim 2, wherein, the vehicle headway calculating device is configured to: calculate vehicle headway between the vehicle and the preceding vehicle by adding vehicle headway of the vehicle to the preceding vehicle to a vehicle length of the preceding vehicle; and calculate vehicle headway between the vehicle and the following vehicle by adding vehicle headway of the vehicle to the following vehicle to a vehicle length of the vehicle.

5. The apparatus for predicting traffic information according to claim 2, wherein, the density calculating device is configured to: estimate an average vehicle headway of a population based on N vehicle headway samples; and calculate vehicle density of a specific road section.

6. The apparatus for predicting traffic information according to claim 5, wherein, the density calculating device is configured to: calculate vehicle density of a specific road section according to the following formula: wherein K is vehicle density of a specific road section, L is a length of a specific road section, a is a minimum value of vehicle headway μ of a population, and b is a maximum value of vehicle headway μ of a population.

7. The apparatus for predicting traffic information according to claim 1, wherein, the speed prediction device comprises: a deep learning model configured to determine a relationship between vehicle density and vehicle speed. 8.A method for predicting traffic information, comprising: deriving vehicle headway, vehicle headway, and vehicle density using a plurality of sensors installed on a vehicle; deriving travel speed data corresponding to vehicle density and predicting traffic information, wherein predicting traffic information comprises: adding density data on vehicle density to speed data corresponding to a specific time interval; constructing density-velocity pattern data representing at least one speed data corresponding to a vehicle density; deriving a representative speed representing the at least one speed data for each vehicle density; predicting a vehicle speed at a current time point based on the density-velocity pattern data, a pre-constructed time-density pattern, and a pre-constructed time-velocity pattern, the predicting traffic information further comprises: deriving a first vehicle speed corresponding to a first real-time density at a first time point not conforming to the time-density pattern based on the density-velocity pattern data when the real-time density measured in real-time does not conform to the time-density pattern; deriving a second time point corresponding to the first vehicle speed from the time-velocity pattern; performing a translation on the time-velocity pattern by calculating a difference between the second time point and the first time point.

9. The method of claim 8, wherein, deriving the vehicle spacing, the headway, and the vehicle density comprises: obtaining respective vehicle spacings to a front vehicle and a rear vehicle from front and rear sensors of the vehicle; calculating the headway based on the vehicle spacings and a length of the front or rear vehicle obtained by a camera of the vehicle; calculating the vehicle density based on the headway.

10. The method of claim 9, wherein, calculating the headway comprises: calculating two vehicle spacing data of three detectors from one detector.

11. The method of claim 9, wherein, calculating the headway comprises: calculating a headway between the vehicle and the front vehicle by adding the vehicle spacing of the vehicle to the front vehicle and a vehicle length of the front vehicle; calculating a headway between the vehicle and the rear vehicle by adding the vehicle spacing of the vehicle to the rear vehicle and a vehicle length of the vehicle.

12. The method of claim 9, wherein, calculating the vehicle density comprises: estimating an average headway of the population based on N headway samples; calculating the vehicle density for a specific road segment.

13. The method of claim 12, wherein, calculating the vehicle density for a specific road segment comprises: calculating the vehicle density according to the following formula: where K is the vehicle density for the specific road segment, L is a length of the specific road segment, a is a minimum value of the headway μ of the population, and b is a maximum value of the headway μ of the population.

14. The method of claim 8, wherein, the predicting traffic information comprises: performing deep learning modeling configured to determine a relationship between the vehicle density and the vehicle speed.

Images