Traffic speed prediction device and traffic speed prediction method

Abstract

The invention discloses a traffic speed prediction device and a traffic speed prediction method. The traffic speed prediction device may include: a communication device configured to receive probe data from a probe vehicle traveling on a road; a storage device configured to store data and an algorithm for predicting a traffic speed; and at least one processor electrically connected to the communication device and the storage device. The processor may be configured to predict a traffic speed on the road using traffic speed data, traffic volume data, and congestion data obtained based on the probe data, and obtain congestion data by creating a target road network including a target collection section collecting the probe data and determining a congestion matrix of the target road network.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to and the benefit of Korean Patent Application No. 10-2023-0107061 filed in the Korean Intellectual Property Office on August 16, 2023, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present disclosure relates to a traffic speed prediction device and method thereof, and more particularly, to a technology for predicting future traffic speed by determining the shock wave velocity within a congested road network urban road. Background Art

[0004] Traffic speed data at a specific time in the past and / or traffic volume data at a specific time in the past can be used to predict traffic speed at a specific time in the future (e.g., through deep learning). However, some future prediction techniques (e.g., using traditional deep learning) have the following problem: for a pair of inputs and outputs that have a clear causal relationship with past vehicle speeds, the accuracy of the prediction ability is high, but for a pair of inputs and outputs with unclear causal relationships, the accuracy of the prediction ability cannot be guaranteed.

[0005] Additionally or alternatively, if the relationships between road segments are not reflected in a road network with many intersections, the accuracy of predicted traffic speeds may be reduced. Summary of the invention

[0006] The following summary presents a brief summary of certain features. This summary is not an extensive overview and is not intended to identify key or important elements.

[0007] Systems, devices, and methods for traffic speed prediction are described. A traffic speed prediction device may include: a communication device configured to receive probe data from a probe vehicle traveling on a road; a storage device configured to store an algorithm for predicting traffic speed and the probe data; and at least one processor coupled to the communication device and the storage device. The at least one processor may be configured to: predict traffic speed on a road based on traffic speed data, traffic volume data, and congestion data, wherein the traffic speed data, traffic volume data, and congestion data are based on the probe data; and obtain the congestion data by: generating a target road network including target collection road segments corresponding to locations where the probe data are collected; and determining a congestion matrix for the target road network.

[0008] In addition or alternatively, a traffic speed prediction method may include: a processor receiving detection data from a detection vehicle traveling on a road; the processor acquiring traffic speed data and traffic volume data based on the detection data; the processor generating a target road network, the target road network including target collection road sections corresponding to locations where the detection vehicle collects detection data; the processor determining congestion data of a congestion matrix including the target road network; and the processor predicting the traffic speed on the road using the traffic speed data, traffic volume data and congestion data.

[0009] These and other features and advantages are described in more detail below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A block diagram representing an example traffic speed prediction apparatus is shown.

[0011] Figure 2 An example diagram depicting remaining traffic and departing traffic for predicting traffic speed is shown.

[0012] Figure 3 An example of a traffic speed prediction model is shown.

[0013] Figure 4 An example of a congestion matrix model for predicting traffic speeds is shown.

[0014] Figure 5A Showing road segments and traffic flows for an example urban road network.

[0015] Figure 5B An example view showing structuring road segments and traffic flows of a city road network into a graph.

[0016] Figure 6 A flow chart representing an example traffic speed prediction method is shown.

[0017] Figure 7 A flow chart representing an example method of determining a congestion matrix for traffic speed predictions is shown.

[0018] Figure 8 An example computing system is shown. DETAILED DESCRIPTION

[0019] Hereinafter, some examples of the present disclosure will be described in detail with reference to the exemplary drawings. It should be noted that when adding reference numerals to the constituent elements of each drawing, the same constituent elements have the same reference numerals as much as possible even if they are shown on different drawings. When describing an example, when it is judged that a detailed description of a known configuration or function related to the example may make the subject matter of the present disclosure unclear, the detailed description will be omitted.

[0020] When describing the constituent elements according to the examples, terms such as "first", "second", "A", "B", "(a)", "(b)" may be used. These terms are only used to distinguish the constituent elements from other constituent elements, and these terms do not limit the properties, order or sequence of the constituent elements. In addition, unless defined differently, all terms used in this document, including technical terms or scientific terms, have the same meaning as those generally understood by technicians in the field to which the present disclosure belongs (those skilled in the art). Terms defined in general dictionaries should be interpreted as having meanings that match the contextual meanings in the relevant technical field, and should not be interpreted as having ideal or overly formalized meanings unless clearly defined in this specification.

[0021] In the following, reference will be made to Figures 1 to 8 Various examples of the present disclosure are described in detail.

[0022] Figure 1 A block diagram representing an example traffic speed prediction apparatus is shown.

[0023] The traffic speed prediction device 100 according to the present disclosure can be implemented inside and / or outside the vehicle. The traffic speed prediction device 100 can be formed integrally with the internal control unit of the vehicle, and / or can be implemented as a configuration and / or device separate from the vehicle. The traffic speed prediction device 100 can be installed in the vehicle and / or attached to the vehicle, and / or a part of the traffic speed prediction device 100 can be formed integrally with the vehicle, and another part can be implemented and / or installed and / or attached to the vehicle as a configuration and / or device separate from the vehicle. The traffic speed prediction device 100 can be implemented as an integrated and / or separate hardware device, connected to the control unit of the vehicle by a connection device (e.g., wired and / or wireless communication).

[0024] The traffic speed prediction device 100 can be configured to predict the traffic speed on the road using traffic speed data, traffic volume data and congestion data obtained based on detection data, and can obtain congestion data by creating a target road network including target collection sections for collecting detection data and determining a congestion matrix of the target road network.

[0025] In this case, the detection data may include GPS data, coordinate data and / or time data.

[0026] Additionally or alternatively, the traffic speed data may include the traffic speed data of the target collection section and the previous collection section in the past predetermined time period. The traffic volume data may include the traffic volume data of the target collection section and the previous collection section in the past predetermined time period, and may include the leaving traffic volume and the remaining traffic volume, etc. The congestion data may include congestion information based on the shock wave speed propagating between sections.

[0027] Reference Figure 1 The traffic speed prediction device 100 may include a communication device 110 , a storage device 120 , an interface device 130 , and a processor 140 .

[0028] The traffic speed prediction apparatus 100 may be implemented as a single unit, for example, by coupling components to each other, and / or some components may be omitted.

[0029] The communication device 110 may be a device implemented in hardware and / or software, and implemented using various electronic circuits that send and receive signals through wireless and / or wired connections. The communication device 110 may send information to and / or receive information from devices inside the traffic speed prediction device 100 (e.g., the storage device 120, the interface device 130, and / or the processor 140) and / or devices outside the traffic speed prediction device 100 (e.g., devices remote / independent of the traffic speed prediction device 100) based on network communication technology. For example, the internal network communication technology may include controller area network (CAN) communication, local interconnect network (LIN) communication, flex-ray communication, etc.

[0030] Additionally or alternatively, the communication device 110 may communicate with the external probe vehicles 200 , 300 , and 400 , a server (not shown), and an infrastructure (not shown) by utilizing at least one of a mobile communication module, a wireless Internet module, short-range communication, or any combination thereof.

[0031] The mobile communication technology can communicate with the detection vehicles 200, 300 and 400 through a mobile communication network established according to mobile communication technology standards and / or communication methods (e.g., Global System for Mobile Communications (GSM), Code Division Multiple Access (CDMA), Code Division Multiple Access 2000 (CDMA 2000), Enhanced Voice Data Optimized or Enhanced Voice Data Only (EV-DO), Wideband CDMA (WCDMA), High Speed ​​Downlink Packet Access (HSDPA), High Speed ​​Uplink Packet Access (HSUPA), Long Term Evolution (LTE), Long Term Evolution-Advanced (LTE-A), Fourth Generation Mobile Communications (4G), Fifth Generation Mobile Communications (5G)).

[0032] The wireless Internet module can be a module for wireless Internet access, and can communicate with the detection vehicles 200, 300 and 400 through wireless local area network (WLAN), wireless fidelity (Wi-Fi), Wi-Fi direct connection, Digital Living Network Alliance (DLNA), wireless broadband (WiBro), world interoperability for microwave access (WiMAX), high-speed downlink packet access (HSDPA), high-speed uplink packet access (HSUPA), long-term evolution (LTE), long-term evolution-enhanced (LTE-A), etc.

[0033] The short-range communication module can be connected by using Bluetooth TM ), radio frequency identification (RFID), infrared data communication (IrDA), ultra-wideband (UWB), ZigBee, near field communication (NFC), wireless universal serial bus (USB) technology or at least one of any combination thereof to support short-range communication.

[0034] For example, the communication device 110 may receive the probe data from the probe vehicles 200, 300, and 400. In this case, the probe data may include GPS data, coordinate data, and / or time data.

[0035] The storage device 120 may store data and / or algorithms required for the processor 140 to operate according to the present specification. For example, the storage device 120 may store a traffic speed prediction model and a congestion matrix model for predicting traffic speed. In addition or alternatively, the storage device 120 may store input data used by the traffic speed prediction device 100 to determine the prediction result and / or result data output by the traffic speed prediction device 100. In addition or alternatively, the storage device 120 may store various logics, algorithms and / or programs for processing input data and output data to predict traffic speed.

[0036] Additionally or alternatively, the storage device 120 may store data and algorithms for generating a congestion matrix.

[0037] The storage device 120 may include a storage medium of at least one memory type, such as a flash memory, a hard disk, a micro, a card type (e.g., a secure digital (SD) card or an extreme digital (XD) card), a random access memory (RAM), a static RAM (SRAM), a read-only memory (ROM), a programmable ROM (PROM), an electrically erasable PROM (EEPROM), a magnetic memory (MRAM), a magnetic disk, and an optical disk. The storage device 120 may include a non-transitory computer-readable storage medium storing instructions that, when executed (e.g., by the processor 140), may configure the processor to perform one or more operations disclosed herein.

[0038] The interface device 130 may include an input device for receiving control commands from a user and an output device for outputting the operating state of the device 100 and its results. Here, the input device may include a key, and may include a mouse, a joystick, a jog shuttle, a stylus, etc. In addition, the input device may include a soft key implemented on a display. For example, the interface device 130 may display the predicted traffic speed for each road segment (e.g., based on an input requesting a predicted traffic time and / or requesting an output of a direction to a location of interest).

[0039] The output device may include a display, and may also include a voice output device such as a speaker. In this case, in response to the display being provided with a touch sensor formed of a touch film, a touch sheet, or a touch pad, the display may be operated as a touch screen, and may be implemented in a form in which the input device and the output device are integrated.

[0040] In this case, the display may include at least one of a liquid crystal display (LCD), a thin film transistor liquid crystal display (TFTLCD), an organic light emitting diode display (OLED display), a flexible display, a field emission display (FED), a 3D display, or any combination thereof.

[0041] The processor 140 may be communicatively (e.g., electrically) connected to the communication device 110, the storage device 120, and / or the interface device 130. The processor 140 may electrically control each component through the communication (e.g., electrical) connection. The processor 140 may include circuits configured to execute software commands and / or instructions (e.g., software commands and / or instructions stored in the storage device 120) to perform various data processing and calculations described herein.

[0042] The processor 140 may process signals (e.g., signals transmitted between components of the traffic speed prediction device 100) to perform overall control so that each component performs its function as disclosed herein. The processor 140 may be implemented in the form of hardware, software, or a combination of hardware and software. For example, the processor 140 may be implemented as a microprocessor, but the present disclosure is not limited thereto.

[0043] The processor 140 may predict the traffic speed on the road using the traffic speed data, the traffic volume data, and the congestion data obtained based on the probe data, and may create (e.g., generate) a target road network including target collection sections (e.g., sections for collecting probe data) (see Figure 5B ) and determine the congestion matrix of the target road network to obtain congestion data.

[0044] The processor 140 may determine the congestion matrix based on at least one of traffic speed data within the target road network, exiting traffic volume (e.g., the number of vehicles proceeding from the target collection segment to the preceding segment), the length of each segment within the target road network, or any combination thereof.

[0045] The processor 140 may visualize the traffic flows of adjacent road segments into a directional graph structure based on the target collection road segments to generate a target road network (eg, a virtual target road network, image data representing the target road network, etc.).

[0046] The processor 140 may determine an initial shock wave velocity matrix by determining the shock wave velocities of the road segments within the target road network.

[0047] The processor 140 can determine the shock wave velocity from the first segment to the second segment (e.g., through the first segment to the second segment) based on (e.g., by utilizing) at least one of the traffic volume entering the first segment 210 among the segments, the traffic volume entering the second segment 220 among the segments (e.g., the traffic volume entering the target collection segment), the traffic flow density on the first segment, the traffic flow density on the second segment, or any combination thereof.

[0048] In response to (for example, based on) the target road network including n road segments (for example, for road segments ①, ②, ③, ④, ⑤, ⑥, ⑦, ⑧ in FIG. 5, n=8 in FIG. 5), the processor 140 can store the speed of the shock wave propagating from the first road segment ① to the second road segment ②, the speed of the shock wave propagating from the first road segment ① to the third road segment ③, and the speed of the shock wave propagating from the first road segment ① to the nth road segment. The velocity of the shock wave is used as the value of the first row of the initial shock wave velocity matrix, storing the velocity of the shock wave propagating from the second segment ② to the first segment ①, the velocity of the shock wave propagating from the second segment ② to the third segment ③, and the velocity of the shock wave propagating from the second segment ② to the nth segment The shock wave velocity is taken as the value of the second row of the initial shock wave velocity matrix, and the shock wave velocity from the nth section is stored. The speed of the shock wave propagating to the first segment ①, the speed of the shock wave propagating from the nth segment The velocity of the shock wave propagating to the second segment ② and the velocity of the shock wave propagating from the nth segment Propagate to the nth segment The initial shock wave velocity matrix (eg, Equation 7) is configured using the shock wave velocity of as the value of the nth row of the initial shock wave matrix.

[0049] The processor 140 may determine an adjacent matrix (eg, Equation 5) by reflecting the traffic flows of road segments adjacent to the target collection road segment within the target road network.

[0050] In response to (e.g., in, based on) the case where the target road network includes n road segments, the processor 140 may configure an adjacent matrix A in the form of an n×n square matrix. In this case, based on the first road segment ① in the first row of the adjacent matrix A, information related to the traffic flow of the n adjacent road segments may be stored as 0 or 1, based on the second road segment ② in the second row of the adjacent matrix, information related to the traffic volume of the n adjacent road segments may be stored as 0 or 1, and based on the nth road segment in the nth row of the adjacent matrix A Information related to the traffic volumes of n adjacent road sections is stored as "0" or "1" (eg, Equation 5).

[0051] Processor 140 may store information related to the traffic flow of the corresponding road segment as "1" in response to (for example, in, based on) the traffic flowing in a direction entering the first road segment ① among the n road segments, and may store information related to the traffic flow of the corresponding road segment as "0" in response to (for example, in, based on) the traffic flowing in a direction exiting the first road segment ①.

[0052] The processor 140 may determine an initial shock wave velocity matrix (eg, Equation 7) and a neighbor matrix (eg, Equation 5) to determine a shock wave velocity matrix (eg, Equation 8) representing the shock wave propagation velocity of each road segment.

[0053] The processor 140 may determine a congestion severity vector (eg, Equation 10) representing the average congestion severity per number of intersections of each road segment by using the shock wave velocity matrix (eg, Equation 7) and the neighbor matrix (eg, Equation 5).

[0054] The processor 140 may determine an n×1 congestion severity vector (eg, Equation 10) by utilizing (eg, based on) the number of drivable intersections of the n road segments in the adjacency matrix and the sum of the shock wave velocities of the n road segments in the shock wave velocity matrix.

[0055] The processor 140 may determine a congestion activation vector (eg, Equation 11) indicating congestion resolution or propagation by using a value obtained by adding the distance vector of each road segment to the congestion severity vector (eg, Equation 10).

[0056] The processor 140 may utilize (e.g., apply) a negative pass filter (e.g., for a congestion activation vector) to determine that congestion will be resolved within a predetermined unit time at the current segment in response to (e.g., in, based on) the sum of the congestion severity vector (e.g., Equation 10) and the distance vector of each segment being a positive value, and to determine that congestion will occur from the current segment to an adjacent segment in response to (e.g., in, based on) the sum of the congestion severity vector (e.g., Equation 10) and the distance vector of each segment being a negative value.

[0057] The processor 140 may determine a congestion vector (eg, Equation 12) by performing matrix multiplication of the congestion activation vector and the Nth-power neighboring matrix, the vector indicating the extent to which congestion spreads to adjacent road segments.

[0058] The processor 140 may configure a congestion matrix (eg, Equation 13) by determining an expected congestion vector from a current point to N unit times later.

[0059] The processor 140 can predict the traffic speed of the target road network in the next N unit time by using the traffic speed data and traffic volume data of the target collected road section in the past M hours and the congestion data of the target road network in the next N unit time.

[0060] Therefore, according to the present disclosure, compared with the prior art, by predicting future traffic speeds using a congestion matrix that effectively reflects the relationship between road sections in a road network with multiple intersections, the accuracy of traffic speed prediction results can be improved.

[0061] According to the present disclosure, by determining the shock wave velocity of each road segment, the length of congestion propagation per unit time (e.g., the distance that the starting point of congestion becomes longer) can be determined. In response to (e.g., in, based on) the situation where congestion does not exceed the road segment length (e.g., based on determining that congestion does not propagate through the road segment length per unit time), the congestion activation vector can be deactivated and may not be calculated in the congestion matrix, thereby preventing accumulation and divergence.

[0062] Additionally or alternatively, according to the present disclosure, the final congestion matrix may be obtained by matrix multiplying the congestion activation vector with the adjacent matrix of the road network, and thus may reflect the interdependence between the road segments in the target road network.

[0063] Additionally or alternatively, the congestion matrix may have the same shape as the shape of the traffic speed to be predicted N unit times later, which may give a direct relationship to each element of the traffic speed to be predicted, and thus, by predicting future traffic speeds using a highly causal congestion matrix, the accuracy quality of future traffic speed predictions may be improved.

[0064] Figure 2 An example diagram depicting remaining traffic and departing traffic for predicting traffic speed is shown.

[0065] Reference Figure 2 , the road section is divided into a previously collected road section 110 and a target collected road section 220. In this case, the target collected road section 220 refers to a road section where detection data is collected, and the previously collected road section 110 refers to a road section of a previous target collected road section.

[0066] In addition or alternatively, the target collection section 220 can include a section connected with the section ahead of the collection detection data. Hereinafter, in this article, the target collection section 220 can be referred to as the second section, target road, target section, target road section, target road, target section, target collection section, etc. In addition or alternatively, the section ahead of the target collection section can also be referred to as the first section, previously collected section, etc.

[0067] The exit traffic volume may be defined as the number of all probe vehicles that have exited the target collection road section 220. Since the number of probe vehicles that have exited the target collection road section 220 within a specified time is 1, an example in which the exit traffic volume is 1 is disclosed.

[0068] The remaining traffic volume may be defined by the number of probe vehicles obtained by subtracting the number of probe vehicles passing through the second point 125 as the last point of the target collection section 220 from the number of probe vehicles passing through the first point 115 as the last point of the previous collection section 110 of the target collection section 220. That is, the remaining traffic volume may be defined as the number of probe vehicles traveling in the target collection section 220.

[0069] exist Figure 2 , the number of probe vehicles traveling in the target collection road section 220 within a specified time may be 3, and the remaining traffic volume may be defined as 3.

[0070] Figure 3 An example of a traffic speed prediction model is shown.

[0071] Reference Figure 3 The traffic speed prediction device 100 may control at least one artificial neural network using a processor (not shown), and may include a traffic speed prediction model 301 .

[0072] For example, the traffic speed prediction device 100 may input data into an artificial neural network, and may provide a traffic speed prediction function based on vehicle driving conditions (eg, traffic speed and / or traffic volume) via data output through various layers included in the artificial neural network.

[0073] The traffic speed prediction model 301 may be a component implemented as an artificial neural network structure including at least one layer (eg, an input layer, an output layer, and a plurality of hidden layers disposed between the input layer and the output layer).

[0074] The traffic speed prediction device 100 can input the traffic speed data S of the past M time, the traffic volume data F of the past M time, and the congestion matrix J as the congestion data of the entire road network in the future N unit time into the input layer of the traffic speed prediction model 301. N , and can obtain the speed data S of the entire road network in the future N unit time output to the output layer through multiple layers N .

[0075] Congestion Matrix J NCongestion data of the entire road network in the next N unit time may be included, which may include a congestion matrix that determines congestion in the next up to N time based on the current shock wave speed. In addition or alternatively, all data input to the traffic speed prediction model 301 may be combined and / or input through a chain operation.

[0076] Figure 4 An example of a congestion matrix model for predicting traffic speeds is shown.

[0077] Reference Figure 4 The congestion matrix J may be output in response to (eg, based on) the traffic speed S, the leaving traffic volume f, and the length l of each road segment in the current road network being input to the congestion matrix model 401. N .

[0078] In order to obtain the congestion matrix J N The traffic speed prediction device 100 can estimate the propagation time within the urban roads of the congested road network based on the shock wave theory.

[0079] First, to describe the shock wave theory, a shock wave refers to a wave (transition phenomenon) generated due to a speed change between different traffic flows.

[0080] In response to (eg, based on, in response to) a situation where smooth traffic flow and congested traffic flow on a road change, a wave (shock wave) may be generated. The shock wave may precisely refer to the propagation speed of the congestion start point.

[0081] (Equation 1)

[0082]

[0083] Among them, q a represents the traffic volume leaving point a, q b represents the traffic volume leaving point b, k a represents the density of traffic flow at point a, k b Represents the density of traffic flow at point b.

[0084] (Equation 2)

[0085]

[0086] Among them, k n represents the road density of the nth road based on the remaining traffic volume per unit time, f n represents the function of the departure traffic volume of the nth road and time (e.g., time unit, time / time unit increment, the time unit increment in the example is 5 minutes), v n represents the driving speed of the nth road, l nrepresents the driving distance of the nth road, t n Indicates the driving time of the nth road.

[0087] Substituting Equation 2, which is the road density of the n-th road based on the remaining traffic volume, into Equation 1, the shock wave velocity can be calculated as shown in the following Equation 3.

[0088] (Equation 3)

[0089]

[0090]

[0091] The congestion starting point from point a to point b in the road segment is represented by the shock wave speed W ab Movement. The amount of congestion can be determined by determining the distance the starting point of the congestion moves per unit time.

[0092] In this case, the shock wave velocity W ab It can be a value obtained by dividing the difference in traffic volume between road sections by the difference in physical capacity of the road sections, as shown in the following formula 4.

[0093] (Equation 4)

[0094]

[0095] When the traffic flows from point a to point b and a bottleneck occurs at point b, Wab becomes a negative value.

[0096] Figure 5A shows the road segments and traffic flows of an example urban road network, Figure 5B An example view showing structuring road segments and traffic flows of a city road network into a graph.

[0097] like Figure 5A As shown, the urban roads including road segments and intersections can be mapped as follows Figure 5B The directional pattern structure shown.

[0098] like Figure 5A As shown, the road network includes road segments ①, ②, ③, ④, ⑤, ⑥, ⑦, and ⑧ (for example, users / vehicles, etc. can move to destinations through these road segments), and traffic flow can be represented by reflecting edges, where the edges are the ends of each road segment. For example, in response to (for example, in, based on) the situation where the edges of node link ① are 501 and 502 respectively, node links ③, ⑤, and ⑥ represent traffic flows entering edge 501, and node links ② and ④ represent traffic flows flowing out of edge 502.

[0099] like Figure 5BAs shown, two physically connected sections can be connected by edges and diagrammatically. For sections ② and ④, based on section ①, arrows are drawn from ① to ② and from ① to ④ respectively, indicating the direction of exiting from section ①. In this case, several intersections centered on edges 501 and 502 can be known.

[0100] For sections ③, ⑤ and ⑥, the arrows point toward section ①, and for sections ⑦ and ⑧, the arrows point toward section ⑥.

[0101] An exemplary traffic flow may be formed to include a neighbor matrix as shown in Equation 5 below.

[0102] (Equation 5)

[0103]

[0104] because Figure 5B The traffic flow schematically shown in FIG. 5 has 8 road segments, so the adjacent matrix A in Equation 5 is disclosed as an example of an 8×8 matrix.

[0105] However, in response to (eg, in, based on) the situation that the road segments are not limited to 8 and there are n road segments, it can be configured as an n×n matrix.

[0106] The first row of the adjacent matrix A has the value "0 0 1 0 1 1 0 0", which reflects the number of intersections connected to the road segment ①. In other words, the value of the road segment where the traffic flows in the direction of the road segment ① should be "1". Figure 5B , the traffic flow to segment ① corresponds to segments ③, ⑤, and ⑥, so "1" is stored in the third, fifth, and sixth columns of the first row of the adjacent matrix A. Since only one segment ① has traffic flow directed to segment ②, the value of the second row of the adjacent matrix A is "1 0 0 00 0 0". Similarly, since only one segment ① has traffic flow directed to segment ④, the value of the fourth row of the adjacent matrix A is "1 0 0 0 0 0 0". The values ​​of all rows of the adjacent matrix A are stored in the same way as above.

[0107] The future velocity prediction apparatus 100 may determine a shock wave velocity matrix.

[0108] The future speed prediction apparatus 100 may determine the shock wave speed Wab between road points (eg, between point a and point b) by utilizing (eg, based on) the departing traffic volume f and density k per unit time (eg, 5 minutes), as shown in Equation 6 below.

[0109] (Equation 6)

[0110]

[0111] In this case, fa represents the departure traffic volume at point a, f b represents the departure traffic volume at point b, t a represents the passing time of point a, t b represents the passing time of point b, l a represents the length of the road section including point a, l b Indicates the length of the road segment including point b.

[0112] The future speed prediction device 100 can determine a first value, which is determined by the departure traffic volume f at point a. a and the passing time t of point a a divided by the length of the road segment including point a a It turns out that a second value can also be determined, which is determined by the leaving traffic volume f at point b. b Multiply by the transit time t of point b b Divide by the length of the road segment including point b, l b The value obtained by subtracting the second value from the first value is called the third value, which can be obtained by the amount of traffic leaving point a, f a Subtract the outgoing traffic volume f from point b b , divided by the third value, and multiplied by 1 / 0.012 to determine the shock wave velocity Wab.

[0113] By determining the shock wave velocity of all roads using Equation 6, the initial shock wave velocity matrix W can be defined as shown in Equation 7 init , the matrix is ​​a square matrix.

[0114] (Equation 7)

[0115]

[0116] W 11 represents the shock wave velocity value (0) from section ① to section ①, W 12 represents the shock wave velocity value from section ① to section ②, and so on. In this way, the shock wave velocity matrix can be obtained by determining the shock wave velocities of all sections. However, the initial shock wave velocity matrix W in Equation 7 init May not reflect the physical connections between roads.

[0117] Therefore, the future speed prediction device 100 can determine the shock wave velocity matrix W propagating between roads by reflecting the physical connection between roads. propagate , as shown in Formula 8 below.

[0118] (Equation 8)

[0119]

[0120] That is, the future velocity prediction device 100 can calculate the initial shock wave velocity matrix W within the unit time T. init By element-by-element multiplication with the adjacent matrix A taking into account the physical connection, the shock wave velocity matrix W including the shock wave propagating once between the physically connected roads is determined. propagate .

[0121] Next, the future velocity prediction device 100 can use the shock wave velocity matrix W propagate Define the congestion severity vector G T As shown in Equation 9 below, the congestion severity G can be determined by determining the adjacent matrix A and the node feature vector E.

[0122] (Equation 9)

[0123] G=(A,E)

[0124] Among them, G represents the severity of congestion, A represents the adjacent matrix, and E represents the node feature vector.

[0125] (Equation 10)

[0126]

[0127] In the future speed prediction device 100, the congestion severity vector G T The element g n It can be obtained by the sum of the shock wave velocities to be propagated (∑ j W propagate，1：j ) and the number of intersections that can be driven at point n (∑ j A 1：j ). In this case, n may include the number of road sections.

[0128] For example, the sum of the shock wave velocities (∑ j W propagate，1：j ) is the shock wave velocity matrix W in Eq. 8 propagate The sum of all values ​​from the 1st column to the jth column of the first row, that is, the value W 13 , W 15 and W 16 The sum.

[0129] The number of intersections that can be driven at point n (∑ j A 1：j ) is the sum of all columns from the 1st column to the Jth column of the first row of the adjacent matrix A. The first row of the adjacent matrix A is "0 0 1 0 1 1 0 0", so adding the values ​​of each column equals 1+1+1, which is 3. Therefore, the congestion severity G T,1 becomes (W 13 +W 15 +W 16) / 3.

[0130] Congestion severity G T,1 It becomes the average value of the shock wave velocity that will propagate once in the direction of each adjacent road.

[0131] As shown in Equation 11, the future speed prediction device 100 can use the congestion severity vector G T To define the congestion activation vector J T .

[0132] (Equation 11)

[0133]

[0134] The future speed prediction device 100 can calculate the distance vector l of each road segment by n and the congestion severity vector G T The summed values ​​are applied to a filter σ(x) that passes only negative numbers to determine the congestion activation vector J T .

[0135] That is, the future speed prediction device 100 can calculate the distance vector l of each road segment by n and the average congestion severity g T,n Add to determine the congestion activation vector J T , the congestion activation vector J T Used as an indicator of the excess traffic volume that would saturate the current road segment and cause congestion within a unit of time.

[0136] Therefore, the distance vector l of each road segment is n and the average congestion severity g T,n The positive value obtained by adding (g T,n +l n ) indicates that the current road congestion is resolved within a unit time (e.g., 5 minutes). On the other hand, the distance vector l of each road segment is n and the average congestion severity g T,n A negative value obtained by adding together indicates that congestion occurs beyond the current road to adjacent roads.

[0137] The future speed prediction device 100 can use the congestion activation vector J T Determine the congestion vector J T,N . Congestion vector J T,N is the congestion activation vector J T A vector of degrees to propagate once to neighboring roads.

[0138] The future speed prediction device 100 can determine the congestion activation vector J based on the current traffic propagation. T,0 With the Nth power adjacent matrix AN Perform matrix multiplication to determine the congestion vector J T,N In this case, N represents the unit time.

[0139] (Equation 12)

[0140] J T，1 =AJ T，0 ; J T，N =A N J T，0

[0141] (Equation 13)

[0142] J N =[J T，1 ...J T，N ]

[0143] The future speed prediction apparatus 100 may configure (eg, generate) a congestion matrix J by determining an expected congestion vector from the current time to N unit times later. N .

[0144] That is, the congestion activation vector J T,0 Refers to the degree of congestion propagation based on the current traffic volume propagation, the congestion activation vector J T,1 It refers to the degree of congestion propagation after a predetermined unit time (e.g., 5 minutes). The congestion activation vector J T,2 The congestion activation vector J refers to the degree of congestion propagation in response to (eg, after, based on) a predetermined unit time twice (eg, 5 minutes*2=10 minutes). T,N Refers to the degree of congestion propagation in response to (eg, upon, based on) a predetermined unit time N times (eg, after N*5 minutes).

[0145] Therefore, the degree of congestion propagation based on the current traffic volume, the degree of congestion propagation after 5 minutes, the degree of congestion propagation after 10 minutes, and the degree of congestion propagation after 5N minutes can be configured as a congestion matrix (J N ).

[0146] In the following, reference will be made to Figure 6 A traffic speed prediction method according to an example of the present disclosure is described. Figure 6 A flow chart representing an example traffic speed prediction method is shown.

[0147] In the following, it is assumed Figure 1 The traffic speed prediction device 100 performs Figure 6 Additionally or alternatively, Figure 6In the description of , the operations described as being performed by the device can be understood as being controlled by the processor 140 of the traffic speed prediction device 100. In the following example, the operations of steps S101 to S105 can be performed in sequence, but not necessarily in sequence. For example, the order of each operation can be changed, and at least two operations can be performed in parallel.

[0148] Reference Figure 6 , the traffic speed prediction device 100 may determine a second road section connected to the first road section and having collected detection data as a target collection road section (S101). For example, the first road section may include a road section ahead of the second road section.

[0149] The traffic speed prediction apparatus 100 may determine the target collection section based on at least one of direction information included in link information of the first section and the second section, whether probe data is detected in the first section, or a combination thereof.

[0150] The traffic speed prediction device 100 may determine whether the second road section is connected to the first road section differently according to the characteristics of the road.

[0151] In response to (e.g., in, based on) the case where the second road segment is a city road, the traffic speed prediction device 100 may determine whether the second road segment is connected to the first road segment based on whether the direction included in the link information of the first road segment (i.e., the road segment ahead of the second road segment) matches the direction included in the link information of the second road segment. In particular, the traffic speed prediction device 100 may determine that the second road segment is connected to the first road segment in response to (e.g., in, based on) the case where the direction included in the link information of the first road segment matches the direction included in the link information of the second road segment.

[0152] In response to (e.g., in, based on) the case where the second road section is an urban road, the traffic speed prediction device 100 can determine whether the second road section is connected to the first road section based on whether the probe data is detected in the first road section (i.e., the road section ahead of the second road section). In particular, in response to (e.g., in, based on) the case where the probe data is detected in the first road section and the probe data is detected in the second road section, the traffic speed prediction device 100 can determine that the second road section is connected to the first road section.

[0153] In the traffic speed prediction device 100, in response to (e.g., in, based on) the second road section being a highway, the processor may determine whether the second road section is connected to the first road section based on whether the probe data is detected in the first road section (i.e., the road section ahead of the second road section). In particular, in response to (e.g., in, based on) the detection of the probe data in the first road section and the detection of the probe data in the second road section, the processor may determine that the second road section is connected to the first road section.

[0154] In the traffic speed prediction device 100, in response to (for example, in, based on) a situation where the second road segment is a highway and the first road segment, which is a road ahead of the second road segment, is a city road, the processor can determine that the second road segment is connected to the first road segment in a state where a direction enclosed in the link information of the first road segment matches a direction included in the link information of the second road segment.

[0155] The traffic speed prediction device 100 can output first output data (S102) by using traffic speed data of a predetermined time (eg, a first time) obtained based on the probe data collected in the first section and the second section. The first output data includes traffic speed data for the past M hours.

[0156] The traffic speed prediction apparatus 100 may output second output data using the traffic volume data at the first time obtained based on the probe data collected at the first road section and the second road section ( S103 ).

[0157] In response to (e.g., in, based on) the presence of probe data in the first road segment, the traffic speed prediction device 100 can determine the remaining traffic volume corresponding to the number of probe vehicles traveling in the second road segment by utilizing the probe data in the second road segment and the probe data in the first road segment.

[0158] For example, the remaining traffic volume may be determined by subtracting the number of probe vehicles that passed the last point of the first road segment from the number of probe vehicles that passed the last point of the second road segment.

[0159] In response to (e.g., in, based on) the absence of probe data in the first road segment, the traffic speed prediction device 100 may determine the leaving traffic volume corresponding to the number of probe vehicles passing the last point of the second road segment. In this case, the second output data may not include the remaining traffic volume, but may include the leaving traffic volume.

[0160] The traffic speed prediction apparatus 100 may output third data using the congestion data at the first time obtained based on the probe data collected at the first road section and the second road section ( S104 ).

[0161] The traffic speed prediction apparatus 100 may predict the traffic speed of the road including the second road section by applying the first output data, the second output data, and the third output data to the traffic speed prediction model ( S105 ).

[0162] By inputting the first output data, the second output data, and the third output data into the traffic speed prediction model, the traffic speed prediction device 100 can predict the traffic speed at a specified time in the future (eg, N unit time in the future).

[0163] Therefore, the traffic speed prediction device can utilize the remaining traffic volume, traffic speed and congestion data in response to (eg, upon, based on) a situation where vehicles enter from multiple roads such as a city road intersection to facilitate prediction of future traffic speed.

[0164] In the following, reference will be made to Figure 7 A traffic speed prediction method according to an example of the present disclosure is described. Figure 7 A flow chart representing an example method of determining a traffic speed prediction congestion matrix is ​​shown.

[0165] In the following, it is assumed Figure 1 The traffic speed prediction device 100 performs Figure 7 Additionally or alternatively, Figure 7 In the description of , the operation described as being performed by the device can be understood as being controlled by the processor 140 of the traffic speed prediction device 100. In the following example, the operations of steps S201 to S207 can be performed in sequence, but not necessarily in sequence. For example, the order of each operation can be changed, and at least two operations can be performed in parallel.

[0166] Reference Figure 7 , the traffic speed prediction device 100 selects a target collection section (S201). In this case, the traffic speed prediction device 100 may select a section (or a target collection section) where traffic volume data can be collected from all sections of the road.

[0167] The traffic speed prediction device 100 may determine that the outgoing traffic volume and the remaining traffic volume of the target collection section are collected (S202). In response to (e.g., in, based on) the situation where the outgoing traffic volume and the remaining traffic volume of the target collection section are not collected, the traffic speed prediction device 100 may select another section as the target collection section (S203).

[0168] In response to (eg, based on) collecting the leaving traffic volume and the remaining traffic volume of the target collection section, the traffic speed prediction device 100 may determine whether the following steps can be completed: Figure 5B The graph expressing the connectivity of the target collection segment is structured ( S204 ).

[0169] In response to (eg, in, based on) a situation where graph structuring is impossible, the traffic speed prediction apparatus 100 selects a new target collection section as another section ( S203 ).

[0170] In response to (eg, in, based on) the situation where graph structuring is possible, the traffic speed prediction device 100 configures the target road network (S205). In this case, the configuration of the target road network may represent performing the following steps on the target road network: Figure 5B The graph shown is structured.

[0171] The traffic speed prediction device 100 defines an adjacent matrix of the target road network and defines a shock wave speed matrix to determine a congestion vector of the target road network up to N unit times in the future ( S206 ).

[0172] Next, the traffic speed prediction device 100 predicts the traffic speed of the target road network in the next N unit time by using the traffic speed data of the target collection section in the past M hours, the traffic volume data in the past M hours and the congestion vector of the target road network in the next N unit time (S207).

[0173] Therefore, according to the present disclosure, by using not only traffic speed data and traffic volume data but also a congestion matrix to predict the future traffic speed of the target road network, the prediction accuracy of the future traffic speed can be improved.

[0174] In addition or optionally, according to the present disclosure, a congestion matrix can be determined by performing matrix multiplication of a congestion activation vector and an adjacent matrix of a road network, and a congestion matrix that takes causal relationships into account can be used by effectively reflecting the relationship between road segments in a road network with many intersections, thereby improving the prediction accuracy of future traffic speeds.

[0175] In addition or alternatively, the present disclosure can determine the shock wave speed of each road segment, and can determine the length of congestion propagation per predetermined unit time, and disable it when the congestion propagation length does not exceed the road segment length, and is therefore uncertain in the congestion matrix, thereby preventing accumulation and divergence phenomena.

[0176] Figure 8 An example computing system is shown.

[0177] Reference Figure 8 The computing system 1000 includes at least one processor 1100 , a memory 1300 , a user interface input device 1400 , a user interface output device 1500 , a storage device 1600 , and a network interface 1700 connected via a bus 1200 .

[0178] The processor 1100 may be a central processing unit (CPU) or a semiconductor device that processes commands stored in the memory 1300 and / or the storage device 1600. The memory 1300 and the storage device 1600 may include various types of volatile or non-volatile storage media. For example, the memory 1300 may include a read-only memory (ROM) 1310 and a random access memory (RAM) 1320.

[0179] Therefore, the steps of the method or algorithm described in conjunction with the examples disclosed herein may be directly implemented as a hardware or software module or a combination of both executed by the processor 1100. The software module may reside in a storage medium (i.e., memory 1300 and / or storage device 1600) such as RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, a hard disk, a removable disk, and a CD-ROM.

[0180] An exemplary storage medium is coupled to the processor 1100, and the processor 1100 can read information from the storage medium and can write information to the storage medium. Optionally, the storage medium can be integrated with the processor 1100. The processor and the storage medium can reside in an application specific integrated circuit (ASIC). The ASIC can reside in a user terminal. Optionally, the processor and the storage medium can reside in a user terminal as separate components.

[0181] The present disclosure provides a traffic speed prediction device and method thereof, which can accurately predict future traffic speeds by using past traffic speeds of road sections, past traffic volumes of road sections, and congestion data of road networks.

[0182] The present disclosure provides a traffic speed prediction device and method thereof, which can generate a congestion matrix by estimating the propagation time of a congested road network based on shock wave theory, and use the congestion matrix to predict future traffic conditions.

[0183] The technical objectives of the present disclosure are not limited to the objectives mentioned herein, and those skilled in the art can clearly understand other technical objectives not mentioned through the description of the claims.

[0184] The present disclosure provides a traffic speed prediction device, which includes: a communication device configured to receive detection data from a detection vehicle traveling on a road; a storage device configured to store data and algorithms for predicting traffic speed; and at least one processor electrically connected to the communication device and the storage device, wherein the processor predicts the traffic speed on the road using traffic speed data, traffic volume data, and congestion data obtained based on the detection data, and obtains congestion data by creating a target road network including target collection sections from which the detection data is collected and determining a congestion matrix of the target road network.

[0185] The processor may be configured to determine a congestion matrix by utilizing at least one of traffic speed data within the target road network, outgoing traffic volume as the number of vehicles proceeding from the target collection road segment to the front road segment, the length of each road segment within the target road network, or any combination thereof.

[0186] The processor may be configured to generate a target road network by visualizing traffic flows of adjacent road segments as a directional graph structure based on the target collection road segments.

[0187] The processor may be configured to determine an initial shock wave velocity matrix by determining shock wave velocities of road segments within the target road network.

[0188] The processor can be configured to determine the shock wave speed from the first segment to the second segment by utilizing at least one of the traffic volume entering the first segment among the segments, the traffic volume entering the second segment among the segments, the density of traffic flow on the first segment, the density of traffic flow on the second segment, or any combination thereof.

[0189] The processor can be configured to, in response to a situation in which the target road network includes n road sections, for example, store the shock wave velocity propagating from the first section to the second section, the shock wave velocity propagating from the first section to the third section, and the shock wave velocity propagating from the first section to the n-th section as values ​​of the first row of the shock wave velocity matrix; store the shock wave velocity propagating from the second section to the first section, the shock wave velocity propagating from the second section to the third section, and the shock wave velocity propagating from the second section to the n-th section as values ​​of the second row of the shock wave velocity matrix; and store the shock wave velocity propagating from the n-th section to the 1st section, the shock wave velocity propagating from the n-th section to the 3rd section, and the shock wave velocity propagating from the n-th section to the n-th section as values ​​of the nth row of the shock wave velocity matrix to configure an initial shock wave velocity matrix.

[0190] The processor may be configured to determine the adjacent matrix by reflecting the traffic flow of road segments adjacent to the target collection road segment within the target road network.

[0191] The processor can be configured to, in response to the situation that the target road network includes n road sections, configure the adjacent matrix into an n×n square matrix form, and store information related to the traffic flow of the n adjacent road sections as 0 or 1 based on the first road section in the first row of the adjacent matrix, store information related to the traffic flow of the n adjacent road sections as 0 or 1 based on the second road section in the second row of the adjacent matrix, and store information related to the traffic flow of the n adjacent road sections as 0 or 1 based on the nth road section in the nth row of the adjacent matrix.

[0192] The processor can be configured to, in response to the traffic flow flowing in a direction entering the first road section among the n road sections, store the information related to the traffic flow of the corresponding road section as "1"; in response to the traffic flow flowing in a direction exiting the first road section among the n road sections, store the information related to the traffic flow of the corresponding road section as "0".

[0193] The processor may be configured to determine an initial shock wave velocity matrix and a neighboring matrix to determine a shock wave velocity matrix representing the velocity at which the shock wave propagates for each road segment.

[0194] The processor may be configured to determine a congestion severity vector representing an average congestion severity per number of intersections for each road segment by utilizing the shock wave velocity matrix and the adjacency matrix.

[0195] The processor may be configured to determine an n×1 congestion severity vector by utilizing the number of drivable intersections of the n road segments in the adjacency matrix and the sum of the shock wave speeds of the n road segments in the shock wave speed matrix.

[0196] The processor may be configured to determine a congestion activation vector indicating resolution or propagation of congestion by using a value obtained by adding the distance vector of each road segment to the congestion severity vector.

[0197] The processor can be configured to utilize a negative transfer filter to determine, in response to a situation where the sum of the congestion severity vector and the distance vector of each road section is a positive value, that the congestion in the current road section will be resolved within a predetermined unit time; and to determine, in response to a situation where the sum of the congestion severity vector and the distance vector of each road section is a negative value, that congestion will occur from the current road section to an adjacent road section.

[0198] The processor may be configured to determine a congestion vector by performing matrix multiplication of the congestion activation vector and an N-th power neighboring matrix, the congestion vector indicating the extent to which congestion spreads to neighboring road segments.

[0199] The processor may be configured to form a congestion matrix by determining an expected congestion vector from a current point to N units of time later.

[0200] The processor can be configured to predict the traffic speed of the target road network in the next N unit time by utilizing the traffic speed data and traffic volume data of the target collected road section in the past M hours and the congestion data of the target road network in the next N unit time.

[0201] The present disclosure provides a traffic speed prediction method, which includes: a processor receiving detection data from a detection vehicle traveling on a road; the processor obtaining traffic speed data and traffic volume data based on the detection data; the processor generating a target road network, the target road network including a target collection section for collecting the detection data; the processor obtaining congestion data by determining a congestion matrix of the target road network; and the processor predicting the traffic speed on the road using the traffic speed data, the traffic volume data and the congestion data.

[0202] Obtaining congestion data by determining a congestion matrix for a target road network may include determining the congestion matrix by utilizing at least one of traffic speed data within the target road network, outgoing traffic volume as the number of vehicles proceeding from a target collection segment to a preceding segment, the length of each segment within the target road network, or any combination thereof.

[0203] Generating a target road network including target collection sections for collecting probe data may include generating the target road network by visualizing traffic flows of adjacent sections as a directional graph structure based on the target collection sections.

[0204] According to the present technology, the past traffic speed of a road section, the past traffic volume of a road section and the congestion data of the road network can be used to accurately predict the future traffic speed, thereby improving the future traffic speed prediction performance.

[0205] According to the present technology, it is also possible to generate a congestion matrix by estimating the propagation time of a congested road network based on shock wave theory, and to predict future traffic conditions by using the congestion matrix, thereby improving the performance of predicting future traffic speeds.

[0206] Furthermore, various effects that can be determined directly or indirectly through this document can be provided.

[0207] The above description merely illustrates the technical idea of ​​the present disclosure, and those skilled in the art to which the present disclosure belongs may make various modifications and changes without departing from the basic features of the present disclosure.

[0208] Therefore, the examples disclosed in the present disclosure are not intended to limit the technical ideas of the present disclosure, but are used to explain the technical ideas of the present disclosure, and the scope of the technical ideas of the present disclosure is not limited by these examples. The protection scope of the present disclosure should be interpreted by the attached claims, and all technical ideas within the equivalent scope should be interpreted as included in the scope of the present disclosure.

Claims

1. A traffic speed prediction device, comprising: a communication device configured to receive probe data from a probe vehicle traveling on a road; a storage device configured to store an algorithm for predicting traffic speed and the detection data; as well as at least one processor coupled to the communication device and the storage device, Wherein, the at least one processor is configured to: predicting a traffic speed on the road based on traffic speed data, traffic volume data, and congestion data, wherein the traffic speed data, the traffic volume data, and the congestion data are based on the probe data; and The congestion data is obtained by: generating a target road network including target collection road segments corresponding to locations from which the probe data were collected; and A congestion matrix of the target road network is determined.

2. The traffic speed prediction device according to claim 1, wherein: The processor determines the congestion matrix based on at least one of: Traffic speed data within the target road network, The outgoing traffic volume representing the number of vehicles that proceed from the target collection segment to the preceding segment, or The length of each road segment in the target road network.

3. The traffic speed prediction device according to claim 1, wherein: The processor is configured to generate the target road network by visualizing the traffic flows of the adjacent road segments of the target collection road segment as a directional graph structure.

4. The traffic speed prediction device according to claim 1, wherein: The processor is configured to generate an initial shockwave velocity matrix based on shockwave velocities of road segments within the target road network.

5. The traffic speed prediction device according to claim 4, wherein: The processor is configured to determine a shock wave velocity from a first road segment to a second road segment based on at least one of: a volume of traffic entering the first road segment, a volume of traffic entering the second road segment, a density of traffic flow on the first road segment, or a density of traffic flow on the second road segment.

6. The traffic speed prediction device according to claim 4, wherein: The processor is configured to configure the initial shock wave velocity matrix of n road segments in the target road network by: In the first row of the shock wave velocity matrix is ​​stored: The velocity of the shock wave propagating from the first segment to the second segment, The velocity of the shock wave propagating from the first segment to the third segment, and The velocity of the shock wave propagating from the first segment to the nth segment; In the second row of the shock wave velocity matrix is ​​stored: The velocity of the shock wave propagating from the second segment to the first segment, the velocity of the shock wave propagating from the second segment to the third segment, and The velocity of the shock wave propagating from the second section to the nth section; And store in the nth row of the shock wave velocity matrix: The velocity of the shock wave propagating from the nth segment to the first segment, The velocity of the shock wave propagating from the nth segment to the third segment, and The velocity of the shock wave propagating from the nth segment to the nth segment.

7. The traffic speed prediction device according to claim 1, wherein: The processor is configured to generate an adjacent matrix by reflecting traffic flows of road segments adjacent to the target collection road segment in the target road network.

8. The traffic speed prediction device according to claim 7, wherein: The processor is configured to configure the adjacency matrix into an n×n square matrix form based on n road segments in the target road network, wherein each entry of the adjacency matrix includes information indicating whether traffic flows from the road segment corresponding to the column of the entry to the road segment corresponding to the row of the entry.

9. The traffic speed prediction device according to claim 8, wherein: The information includes a value of 1 indicating that traffic flows from the road segment corresponding to the column of the item into the road segment corresponding to the row of the item, and the information includes a value of 0 indicating that traffic does not flow from the road segment corresponding to the column of the item into the road segment corresponding to the row of the item.

10. The traffic speed prediction device according to claim 7, wherein: The processor is configured to: Based on the initial shock wave velocity matrix and the neighbor matrix, a shock wave velocity matrix representing the velocity of shock wave propagation for each road segment is determined.

11. The traffic speed prediction device according to claim 10, wherein: The processor is configured to determine a congestion severity vector based on the shock wave velocity matrix and the neighbor matrix, the congestion severity vector representing an average congestion severity per number of intersections for each road segment.

12. The traffic speed prediction device according to claim 10, wherein: The processor is configured to determine an n×1 congestion severity vector generated based on a sum of shock wave velocities of the n road segments in the shock wave velocity matrix and a number of drivable intersections of the n road segments in the neighbor matrix.

13. The traffic speed prediction device according to claim 11, wherein: The processor is configured to determine a congestion activation vector indicating resolution or propagation of congestion based on a sum of a congestion severity vector and a distance vector of each road segment.

14. The traffic speed prediction device according to claim 13, wherein: The processor is configured to apply a negative transfer filter to the congestion activation vector to: Based on the fact that the sum of the congestion severity vector and the distance vector of each road section is a positive value, it is determined that the congestion in the current road section will be resolved within a predetermined unit time, or Based on the fact that the sum of the congestion severity vector and the distance vector of each road section is a negative value, it is determined that congestion will occur from the current road section to the adjacent road section.

15. The traffic speed prediction device according to claim 13, wherein: The processor is configured to determine a congestion vector by performing matrix multiplication of the congestion activation vector and an N-th power adjacent matrix, wherein the congestion vector represents the extent to which congestion spreads to adjacent road segments.

16. The traffic speed prediction device according to claim 15, wherein: The processor is configured to form a congestion matrix by determining an expected congestion vector from a current point to N time units later.

17. The traffic speed prediction device according to claim 1, wherein: The processor is configured to predict the traffic speed of the target road network in the next N unit time based on the traffic speed data and traffic volume data of the target collection section in the past M unit time and the congestion data of the target road network in the next N unit time.

18. A traffic speed prediction method, comprising: receiving, by a processor, detection data from a detection vehicle traveling on a road; obtaining, by the processor, traffic speed data and traffic volume data based on the detection data; generating, by the processor, a target road network including target collection road sections corresponding to locations where the probe vehicle collects the probe data; Determining, by the processor, congestion data including a congestion matrix of the target road network; as well as The processor predicts a traffic speed on the road using the traffic speed data, the traffic volume data, and the congestion data.

19. The traffic speed prediction method according to claim 18, wherein: Determining congestion data including a congestion matrix of the target road network includes determining, by the processor, the congestion matrix based on at least one of the following: Traffic speed data within the target road network, The outgoing traffic volume based on the number of vehicles advancing from the target collection section to the preceding section, or The length of each road segment in the target road network.

20. The traffic speed prediction method according to claim 18, wherein: The generating the target road network including the target collection road segment includes visualizing the traffic flow into a directional graph structure based on the traffic flow of adjacent road segments related to the target collection road segment.

Images