Vehicle path determination method, device and medium

Abstract

The invention relates to the technical field of intelligent traffic, and discloses a vehicle path determination method and device, and a medium, and the method employs the real flow to reflect the real condition of a target road section, and supplements the traffic demand of vehicle missing detection or false detection. And constructing a traffic residual network according to the real physical connection between the road sections, the ID of the corresponding road section and the residual error between the real traffic and the current traffic, and calculating a gain according to the constructed network, so that when the gain is in a first state, a new path is generated by traversing the road sections to form an updated path set. And updating the traffic residual network by using the updated path set, circularly calculating gains until the gains are all in a second state, namely, when no new path is generated, determining the vehicle path information of the target road section. According to the scheme, the real situation of the target road section is reflected through the real flow, and the missed or false detected path is complemented by using the flow residual network, so that the traffic travel demand of the target road section is accurately reflected, and the accuracy of vehicle path determination is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a method, device and medium for determining a vehicle path. Background Art

[0002] The vehicle path can restore the travel demand characteristics of motor vehicles with high confidence, and can specifically obtain the traffic flow and travel paths within the region. The acquisition of traffic flow and travel paths within the region can usually be achieved by deploying intelligent network equipment, using the intelligent network environment to provide rich vehicle trajectories, and theoretically achieving complete traffic demand modeling.

[0003] In the related art, the vehicle path is usually determined by identifying the license plate or vehicle ID information through image recognition technology or other methods. However, in this method, the corresponding vehicle path is often lost due to the influence of changes in algorithms or external environmental factors, such as missed detection and false detection caused by the vehicle or license plate being blocked or the signal being lost. As a result, due to the low accuracy of the vehicle path, the traffic demand modeled using the path is relatively small and cannot reflect the actual traffic demand. Therefore, there is a corresponding demand for determining the completed vehicle path in the related art. Summary of the invention

[0004] In view of this, the present invention provides a method, device, and medium for determining a vehicle path to solve the technical problems of missed detection and false detection of vehicle paths in related technologies.

[0005] In a first aspect, the present invention provides a method for determining a vehicle path, the method comprising: determining the initial segment sequence information and the current flow of the current path set based on the initial path set of the target segment; constructing a flow residual network based on the initial path set, the real physical connection between the segments, the current flow, and the turning real flow of the target segment within a preset time interval; based on the flow residual network, determining the gain between each segment using the turning real flow, the current flow, and the initial segment sequence information; and determining the vehicle path information of the completed target segment based on the gain.

[0006] In combination with the first aspect, in a possible implementation of the first aspect, based on the gain, determining the vehicle path information of the completed target road segment, including: in response to the gain being in a first state, forming an updated path set according to the gain, updating the traffic residual network based on the updated path set and the real physical connection between the road segments, and returning the steps of determining the gain between each road segment based on it; in response to the gain being in a second state, determining the vehicle path information of the target road segment based on the latest updated path set.

[0007] In combination with the first aspect, in a possible implementation of the first aspect, a flow residual network is constructed based on the real physical connection between the sections of the initial path set, the current flow, and the real flow of turning to the target section within a preset time interval, including: determining multiple sections corresponding to the target section based on the initial path set; determining the flow residual based on the real flow of turning and the current flow; determining the physical connection relationship between the sections, the sampling probability corresponding to each section, the ID of each section, and the flow residual corresponding to the physical connection relationship based on the multiple sections and the flow residual; constructing the flow residual network based on the physical connection relationship between the sections, the sampling probability, the ID of each section, and the flow residual corresponding to the physical connection relationship.

[0008] In combination with the first aspect, in a possible implementation method of the first aspect, based on the traffic residual network, the gain between each road section is determined by utilizing the actual traffic, current traffic, and initial road section sequence information, including: based on the traffic residual, utilizing the physical connection relationship between the road sections corresponding to the initial road section sequence information and the sampling probability corresponding to each road section, to determine the gain between each road section.

[0009] In combination with the first aspect, in a possible implementation of the first aspect, in response to the gain being in a first state, an updated path set is formed according to the gain, based on the updated path set and the real physical connection between the road segments, the flow residual network is updated, and the step of determining the gain between each road segment based on its return includes: in response to the existence of a gain exceeding a preset gain threshold, selecting a road segment sequence exceeding the preset gain threshold, and determining the sum of the gains of the corresponding path sequence; based on the sum of the gains, determining a gain target; based on the gain target, using the gains, traversing adjacent road segments in the road segment sequence information to form a newly added path; adding the newly added path to the path set to form an updated path set; based on the updated path set and the real physical connection between the road segments, updating the flow residual network, and based on its return, determining the step of determining the gain between each road segment.

[0010] In combination with the first aspect, in a possible implementation manner of the first aspect, the method further includes: parsing vehicle trajectory data based on vehicle identification information to determine a matching trajectory; and forming an initial path set based on trajectory points in the matching trajectory.

[0011] In combination with the first aspect, in a possible implementation of the first aspect, based on the vehicle identification information, the vehicle trajectory data is parsed to determine the matching trajectory, including: extracting a trajectory sequence based on the vehicle identification information; mapping the trajectory sequence to a road section, and filtering the trajectory sequence based on the road section to determine the matching trajectory.

[0012] In combination with the first aspect, in a possible implementation of the first aspect, an initial path set is formed based on trajectory points in the matching trajectory, including: based on the trajectory points, calculating vehicle delays between adjacent trajectory points in time sequence; based on each vehicle delay, determining multiple corresponding paths; and forming an initial path set based on the multiple paths.

[0013] In a second aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the vehicle path determination method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0014] In a third aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the vehicle path determination method of the first aspect or any corresponding embodiment thereof.

[0015] The technical solution of the present invention has the following advantages: The present invention provides a method, device, and medium for determining a vehicle path. The method determines the initial segment sequence information and the current flow through the initial path set of the target segment, and uses the initial path set and the real physical connection between the segments to construct a flow residual network, thereby determining the gain between each segment through the flow residual network by turning to the real flow, the current flow, and the initial segment sequence information, and when the gain is in a first state, an updated path set is formed through the gain, and the flow residual network is constructed again based on the updated path set until all gains are in a second state, thereby determining the vehicle path information of the target segment. In this process, the real flow is used to reflect the real situation of the target segment, and the flow residual network is constructed with the real physical connection between the segments, the corresponding segment ID, and the residual of the real flow and the current flow, and the gain is calculated by the constructed network to complete the path, and the vehicle path information of the completed target segment is determined. Therefore, based on the actual situation of the target section, the missed or misdetected paths are supplemented through the traffic residual network to determine the vehicle path information of the target section, so that the supplemented vehicle path information of the target section can supplement the missed or misdetected vehicle paths, thereby accurately reflecting the traffic travel needs of the target section and improving the accuracy of vehicle path determination. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 is a flow chart of a method for determining a vehicle path according to an embodiment of the present invention; Figure 2 The determination of the vehicle path is a schematic diagram of the hardware structure of the computer device of the embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0019] According to an embodiment of the present invention, an embodiment of a method for determining a vehicle path is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0020] This embodiment provides a method for determining a vehicle path. Figure 1 As shown, the method comprises the following steps: S101. Based on an initial path set of a target road segment, determine initial road segment sequence information and current traffic of a current path set.

[0021] Specifically, the initial path set of the target road section refers to a preset initial path set; or, the initial path set is formed by parsing the vehicle trajectory data through the vehicle identification information and using the trajectory points in the vehicle trajectory data. The path set contains multiple path sequences of the target road section, the frequency of the corresponding path sequence, and the distribution of vehicle models of the corresponding path sequence. Each road section sequence is recorded as R , R ={ R 1, R 2,…, R p}, the frequency of the corresponding path sequence is recorded as C , C ={ C 1, C 2,…, C p},in, R p Indicates p A sequence of paths, Cp Indicates p The frequency of the path sequence, the distribution of the corresponding path sequence is recorded as T , T ={ T 1, T 2,…, T p}, T p Indicates p The distribution of vehicles in the path sequence, for example {car:0.7,bus:0.3}. It should be understood that R There are no duplicate items in . If there are identical road segment sequences, the corresponding frequency will increase, that is, the frequency is used to represent the number of times the corresponding road segment sequence appears.

[0022] Specifically, based on the initial path set, determining the initial road segment sequence information refers to determining the path sequence contained in the initial path set, the frequency of the corresponding path sequence, and the vehicle type distribution of the corresponding path sequence.

[0023] Specifically, based on the initial path set of the target road segment, determining the current flow of the current path set means determining the driving trajectory of each road segment in the road segment sequence passed by the vehicle in the target road segment and the traffic flow of each road segment through the initial path set, thereby forming the current flow. The current flow is recorded as f now For example, the sequence of road sections that vehicles pass through on the target road section within the preset time interval is { l 2, l 3, l 5}、{ l 1, l 2, l 3}, where l 1. l 2. l 3. l 5 represents section 1, section 2, section 3, and section 5 respectively, so the current flow is l 1, l 2:1, l 2, l 3:2, l 3, l 5:1, which means that there is one vehicle traveling from section 1 to section 2, two vehicles traveling from section 2 to section 3, and one vehicle traveling from section 3 to section 5.

[0024] It should be understood that the current path set includes an initial path set and an updated path set. The difference between the initial path set and the updated path set is that the updated path set is an updated path set formed by adding the updated path to the initial path set after the updated path. For example, the initial path set is q, and in response to the current path set being q, the calculated gain is in the first state, and the updated path determined according to the gain is q1, then the updated path set at this time is: q+q1, and the current path set at this time is q+q1. If q+q1 is used as the updated path set, the calculated gain is still in the first state, and the updated path determined according to the gain is q2, then the updated path set at this time is: q+q1+ q2, and so on, and no further description is given.

[0025] S102: construct a flow residual network based on the initial path set, the real physical connection between the road sections, the current flow, and the real flow of the target road section within a preset time interval.

[0026] Specifically, the real physical connection between road sections refers to the connected areas between road sections connected in the target road section. Taking the road section sequence in the above embodiment as an example, the real physical connection between road sections exists between road section one and road section two, road section two and road section three, and road section three and road section five.

[0027] Specifically, based on the initial path set, the real physical connection between road sections, the current traffic, and the real traffic turned to the target section within a preset time interval, constructing a traffic residual network means determining multiple sections corresponding to the target section through the initial path set, and using multiple sections to determine the physical connection relationship between the sections, the sampling probability and ID corresponding to each section, so as to construct a traffic residual network with the physical connection relationship between the sections as the edge and the ID corresponding to each section, wherein each edge has three corresponding characteristics, namely the current traffic, the real traffic turned to and the traffic residual, and the traffic residual is the difference between the real traffic turned to and the current traffic.

[0028] S103, based on the flow residual network, using the actual flow of the turn, the current flow, and the initial section sequence information, determine the gain between each section.

[0029] Specifically, based on the flow residual network, the gain between each road section is determined by using the real flow, current flow, and initial road section sequence information. This means determining the flow residual by using the real flow and current flow, and determining the gain between each road section through the flow residual and the physical connection relationship corresponding to the initial road section sequence information and each sampling probability.

[0030] Specifically, the determination of the actual turning flow rate of the target road section within the preset time interval is based on the intelligent network connection device, which obtains the actual turning flow rate of the target road section within the preset time interval. The preset time interval refers to a unit time interval, which can be set according to the actual working conditions, usually 5 to 15 minutes. Intelligent network connection equipment refers to equipment such as cameras, lidars, signal collectors, etc. installed on the side of the road. Through the corresponding equipment, information including vehicle ID, vehicle model, departure location, departure time, arrival location, vehicle coordinates, road section sequence, etc. can be obtained. The actual turning flow rate is recorded as f real .

[0031] Specifically, based on the intelligent network connection device, obtaining the real turning flow of the target road section within the preset time interval means using the departure position, arrival position, vehicle coordinates, and road section sequence obtained by the roadside acquisition device to determine the driving trajectory of each road section in the road section sequence passed by the vehicle in the target road section and the traffic flow of each road section, thereby forming the real turning flow. For example, the road section sequence passed by the vehicle in the target road section within the preset time interval is respectively { l 1, l 2, l 3}、{ l 1, l 2, l 4}, where l 1. l 2. l 3. l 4 represent sections 1 to 4 respectively, so the actual turning flow is l 1, l 2:2, l 2, l 3:1, l 3, l 4:1, which means that there are two vehicles traveling from section 1 to section 2, one vehicle traveling from section 2 to section 3, and one vehicle traveling from section 3 to section 4.

[0032] S104: Determine vehicle path information of the target road section based on the gain.

[0033] Specifically, based on the gain, determining the vehicle path information of the target road section means completing the vehicle path according to the state of the gain until no new path is generated, and taking the full path information at the current moment as the vehicle path information of the target road section. According to the state of the gain, the vehicle path corresponding to the gain being in the first state is completed. When the gain is in the first state, a new path is generated by traversing the road section to form an updated path set. The flow residual network is updated with the updated path set, and the gain is calculated cyclically until the gain is in the second state, that is, when no new path is generated, taking the full path information at the current moment as the vehicle path information of the target road section.

[0034] The present invention provides a method, device, and medium for determining a vehicle path. The method determines the initial segment sequence information and the current flow through the initial path set of the target segment, and uses the initial path set and the real physical connection between the segments to construct a flow residual network, thereby determining the gain between each segment through the flow residual network by turning to the real flow, the current flow, and the initial segment sequence information, and when the gain is in a first state, an updated path set is formed through the gain, and the flow residual network is constructed again based on the updated path set until all gains are in a second state, thereby determining the vehicle path information of the target segment. In this process, the real flow is used to reflect the real situation of the target segment, and the flow residual network is constructed with the real physical connection between the segments, the corresponding segment ID, and the residual of the real flow and the current flow, and the gain is calculated by the constructed network to complete the path, and the vehicle path information of the completed target segment is determined. Therefore, based on the actual situation of the target section, the missed or misdetected paths are supplemented through the traffic residual network to determine the vehicle path information of the target section, so that the supplemented vehicle path information of the target section can supplement the missed or misdetected vehicle paths, thereby accurately reflecting the traffic travel needs of the target section and improving the accuracy of vehicle path determination.

[0035] In an optional implementation, determining the vehicle path information of the completed target road section based on the gain includes: In response to the gain being in a first state, an updated path set is formed according to the gain, a flow residual network is updated based on the updated path set and the real physical connection between the road segments, and the gain between each road segment is determined based on the return thereof; in response to the gain being in a second state, vehicle path information of the target road segment is determined based on the latest updated path set.

[0036] Specifically, the gain being in the first state means that the gain exceeds a preset gain threshold.

[0037] Specifically, forming an updated path set based on gain means selecting a road segment sequence that exceeds a preset gain threshold, determining the sum of gains of the corresponding path sequence, determining a gain target based on the sum of gains, and constructing a new path through the gain target and using the gain, thereby forming an updated path set after adding the new path to the original path set.

[0038] Specifically, the step of updating the flow residual network based on the updated path set and the real physical connection between the road segments, and returning and determining the gain between each road segment based on it refers to bringing the updated path set into the flow residual network again, that is, updating the flow residual network with the updated path set, and determining the gain between each road segment, thereby repeating this cycle until the gains are all in the second state.

[0039] Specifically, the gain being in the second state means that the gains are all below a preset gain threshold. Determining the vehicle path information of the target road section based on the latest updated path set means taking the path in the updated path set when the gain is in the second state as the vehicle path information of the target road section, and outputting it in the form of <vehicle id, vehicle model, departure position, departure time, arrival position, arrival time, road section sequence>.

[0040] In an optional implementation, a flow residual network is constructed based on an initial path set, real physical connections between road sections, current flow, and real flow of a target road section within a preset time interval, including: Based on the initial path set, multiple sections corresponding to the target section are determined; based on the actual steering flow and the current flow, the flow residual is determined; based on the multiple sections and the flow residual, the physical connection relationship between the sections, the sampling probability corresponding to each section, the ID of each section and the flow residual corresponding to the physical connection relationship are determined; based on the physical connection relationship between the sections, the sampling probability, the ID of each section, and the flow residual corresponding to the physical connection relationship, a flow residual network is constructed.

[0041] Specifically, determining multiple road segments corresponding to the target road segment based on the initial path set means using each road segment sequence in the initial path set to match the target road segment to determine multiple road segments corresponding to the target road segment.

[0042] Specifically, based on the actual flow and the current flow, determining the flow residual means using the difference between the actual flow and the current flow to determine the flow residual, recorded as f real - f now .

[0043] Specifically, based on multiple road segments and flow residuals, determining the physical connection relationship between road segments, the sampling probability corresponding to each road segment, the ID of each road segment, and the flow residual corresponding to the physical connection relationship means taking the connected areas between road segments connected in the target road segment as edges to perform the first-level traversal, and for each edge, further traverse all road segment sequences passing through the corresponding edge in multiple road segments to form a road segment sequence set, and calculate the sampling probability of each road segment sequence in the road segment sequence set and the ID of each road segment. Among them, each edge is recorded as E , E ∈( l m , l n ), m , n are positive integers, l m , l nRespectively represent m Section 1 n The road segment sequence set is denoted as R `, R `={ R `1, R `2… R ` s}, R ` s Indicates the first s The path sequence is sampled with probability R prob . And because each edge in the flow residual network has three corresponding characteristics, namely the current flow, the actual flow turned to, and the flow residual, therefore, after determining the physical connection relationship between the road sections, the flow residual corresponding to the physical connection relationship can be determined. It should be understood that the multiple road sections determined by the initial path set here are based on the initial road section sequence information, and the multiple road sections corresponding to the target road sections are determined, thereby performing a two-layer traversal based on the initial road section sequence information, so that the traversal process is based on the initial road section sequence information, reducing the complexity of data traversal, thereby improving the computing speed. And in determining the corresponding E and R prob Afterwards, it is necessary to count the distribution of vehicle models again to provide a data basis for the subsequent construction of new routes.

[0044] Specifically, the sampling probability is calculated by the proportion of the frequency of the corresponding road segment in the corresponding road segment sequence, for example, { l 1, l 2, l 3} has a frequency of 100 vehicles, { l 2, l 3, l 4} has a frequency of 200 vehicles, l 3, l 4} has a frequency of 250 vehicles. If the first layer traverses to section 2, then all the section sequences that pass through section 2 are determined. l 1, l 2, l 3}、{ l 2, l 3, l 4}. And then { l 1, l 2, l 3} and { l 2, l 3, l The sampling probabilities of 4} are calculated as: 100 / (100+200) and 200 / (100+200).

[0045] Specifically, based on the physical connection relationship between road segments, sampling probability, and the ID flow residual of each road segment, the flow residual network is constructed by taking the connected areas between the road segments connected in the target road segment as edges and using the road segment ID corresponding to each road segment to construct the flow residual network. The constructed flow residual network is recorded as G res ( L , E ), where L represents the road segment ID corresponding to each road segment, and each edge E has a corresponding flow residual.

[0046] In an optional implementation, based on the flow residual network, using the real flow, current flow, and initial segment sequence information, the gain between each segment is determined, including: Based on the flow residual, the gain between each road section is determined by utilizing the physical connection relationship between the road sections corresponding to the initial road section sequence information and the sampling probability corresponding to each road section.

[0047] Specifically, the gain between each road segment is calculated by formula (1): (1) in, represents the sum of the gains of all road segment sequences in path S, Rs represents the road segment sequence, which is used to describe all road segments passed by a vehicle path. Represents the current flow residual, represents the original flow residual, Indicates s The probability of a path sequence being sampled.

[0048] Specifically, the original flow residual For Update R ` s The previous flow residual and the current flow residual For Update R ` s The residual flow after R ` s Update refers to the process of generating new paths based on the gain to form an updated path set. Therefore, the original flow residual Indicates the flow residual before updating the flow residual network with the updated path set in this cycle, the current flow residual Represents the flow residual after updating the flow residual network with the updated path set in this cycle.

[0049] Specifically, in response to the gain being below the preset gain threshold, determining the vehicle path information of the target road section means that in the flow residual network constructed by updating the path set, when the gain between each road section is below the preset gain threshold, each path is saved in the form of <vehicle id, vehicle type, departure position, departure time, arrival position, arrival time, road section sequence> as the vehicle path information of the target road section. The preset gain threshold can be set according to the actual working conditions, and this embodiment does not specifically limit this, and zero is usually selected as the preset gain threshold.

[0050] In an optional embodiment, in response to the gain being in the first state, forming an updated path set according to the gain, updating the flow residual network based on the updated path set and the real physical connection between the road segments, and returning the step of determining the gain between each road segment based on the updated path set, comprises: In response to the existence of a gain exceeding a preset gain threshold, a segment sequence exceeding the preset gain threshold is selected to determine the sum of gains of the corresponding path sequence; based on the sum of gains, a gain target is determined; based on the gain target, adjacent segments in the segment sequence information are traversed using the gains to form a new path; the new path is added to the path set to form an updated path set; based on the updated path set and the real physical connection between the segments, a flow residual network is updated, and based on the return, a step of determining the gain between each segment is performed.

[0051] Specifically, in response to the existence of a gain exceeding a preset gain threshold, a sequence of road sections exceeding the preset gain threshold is selected, and the sum of gains of the corresponding path sequence is determined. Determining the gain target means that in the traffic residual network updated by updating the path set, when there is a gain between sections exceeding the preset gain threshold, a sequence of road sections exceeding the preset gain threshold is selected, the sum of gains of the corresponding sections is calculated, and the gain target is determined.

[0052] Specifically, selecting a road segment sequence exceeding a preset gain threshold includes selecting a road segment sequence with the largest corresponding gain or selecting a road segment sequence in a weighted sampling manner according to a value corresponding to the gain.

[0053] Specifically, based on the sum of gains, determining the gain target means taking the gain corresponding to the selected section sequence as the gain target. It should be understood that in this embodiment, since the number of selected section sequences exceeding the preset gain threshold is one section sequence, the gain corresponding to the selected section sequence is numerically the same as the sum of gains in this embodiment. The number of selected section sequences exceeding the preset gain threshold can be selected according to actual working conditions, and this embodiment will not be described in detail.

[0054] Specifically, based on the gain target, using the gain, traversing adjacent road sections in the road section sequence information, forming a new path means generating a random number in the range of [0, gain target), traversing the gain between each road section according to the weight ratio, accumulating the weight until the accumulated weight exceeds or equals the generated random number, selecting the road section corresponding to the currently traversed gain as the new road section sequence, combining the vehicle type distribution counted again in the above embodiment, and randomly generating a vehicle ID to form a new path.

[0055] Specifically, adding a new path to the path set to form an updated path set means that if the original path set is q, in response to the current path set being q, the calculated gain is in the first state, and the updated path determined according to the gain is q1, then the updated path set at this time is: q+q1, and the current path set at this time is q+q1. If q+q1 is used as the updated path set, the calculated gain is still in the first state, and the updated path determined according to the gain is q2, then the updated path set at this time is: q+q1+ q2, and so on, which will not be repeated.

[0056] In an optional embodiment, the method further comprises: Based on the vehicle identification information, the vehicle trajectory data is parsed to determine the matching trajectory; based on the trajectory points in the matching trajectory, an initial path set is formed.

[0057] Specifically, parsing the vehicle trajectory data based on the vehicle identification information and determining the matching trajectory refers to extracting a trajectory sequence based on the vehicle identification information, screening the trajectory sequence, and determining the matching trajectory.

[0058] Specifically, forming an initial path set based on trajectory points in the matching trajectory means calculating vehicle delays between adjacent trajectory points in timing through trajectory points, dividing the matching trajectory into trip chains by vehicle delays, and determining multiple paths, thereby forming an initial path set.

[0059] In an optional implementation, based on the vehicle identification information, parsing the vehicle trajectory data and determining the matching trajectory includes: Based on the vehicle identification information, a trajectory sequence is extracted; the trajectory sequence is mapped to a road segment, and the trajectory sequence is filtered based on the road segment to determine a matching trajectory.

[0060] Specifically, based on vehicle identification information, extracting trajectory sequences means using the vehicle license plate as identification information to extract trajectory sequences corresponding to each vehicle. Each trajectory sequence corresponding to a vehicle contains multiple trajectory points, and the set of trajectory points is denoted as S , S ={ s 1, s 2, s 3,s 4,…, s},in, s Indicates n Track points.

[0061] Specifically, mapping the trajectory sequence to the road segment and filtering the trajectory sequence based on the road segment to determine the matching trajectory means mapping the trajectory points to each road segment of the target road segment, removing duplicates by road segment, and only retaining the first trajectory point of the trajectory on the road segment to obtain a new matching trajectory. Snet , Snet The expression can be: s 1- edge 1, s 2- edge 1, s 3- edge 2, s 4- edge 2,…, s - edgen}, the matching trajectory is recorded as S`net , Snet The expression can be: s 1- edge 1, s 3- edge 2,… s - edgem}.

[0062] In an optional implementation, forming an initial path set based on the track points in the matching track includes: Based on the trajectory points, the vehicle delays between adjacent trajectory points in time sequence are calculated; based on each vehicle delay, multiple corresponding paths are determined; based on the multiple paths, an initial path set is formed.

[0063] Specifically, the vehicle delay between adjacent trajectory points in time series is calculated using formula (2): (2) Among them, timeloss represents the vehicle delay between adjacent trajectory points in time sequence. l s1,s2 represents the shortest driving distance between adjacent trajectory points, v represents the average velocity between adjacent trajectory points, t 1. t 2 respectively represent the time corresponding to the two trajectory points.

[0064] Specifically, based on each vehicle delay, determining multiple corresponding paths means that if the vehicle delay between adjacent trajectory points in time sequence exceeds the vehicle delay threshold, the corresponding matching trajectory is divided into a trip chain to form multiple corresponding paths.

[0065] Specifically, as shown in Table 1, Table 1 exemplarily shows a plurality of corresponding paths formed after the corresponding matches are divided into itinerary chains.

[0066] Table 1

[0067] Among them, Edge1, Edge2, and Edge5 respectively represent the sections in the target section.

[0068] Specifically, forming an initial path set based on multiple paths means saving each path according to <vehicle id, vehicle type, departure location, departure time, arrival location, arrival time, road section sequence> to form an initial path set.

[0069] The embodiment of the present invention also provides a computer device, see Figure 2 , Figure 2 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 2 As shown, the computer device includes: one or more processors 201, memory 202, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Each component utilizes different buses to communicate with each other, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in the memory or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 2 A processor 201 is taken as an example.

[0070] The processor 201 may be a central processing unit, a network processor or a combination thereof. The processor 201 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0071] The memory 202 stores instructions executable by at least one processor 201, so that the at least one processor 201 executes the method shown in the above embodiment.

[0072] The memory 202 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 202 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 202 may optionally include a memory remotely arranged relative to the processor 201, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0073] The memory 202 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 202 may also include a combination of the above types of memory. The computer device also includes a communication interface 203, which is used for the computer device to communicate with other devices or communication networks.

[0074] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0075] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for determining a vehicle path, characterized in that: The method comprises: Based on the initial path set of the target road segment, determine the initial road segment sequence information and the current flow of the current path set; Constructing a flow residual network based on the initial path set, the real physical connection between the road sections, the current flow, and the real flow of the target road section within a preset time interval; Based on the flow residual network, using the steering real flow, the current flow, and the initial section sequence information, determining the gain between each of the sections; Based on the gain, the vehicle path information of the completed target road section is determined.

2. The method according to claim 1, characterized in that The determining, based on the gain, vehicle path information of the completed target road section includes: In response to the gain being in a first state, forming an updated path set according to the gain, updating a flow residual network based on the updated path set and the real physical connections between the road segments, and returning to the step of determining the gain between each of the road segments based on the updated path set; In response to the gain being in the second state, vehicle path information of the completed target road segment is determined based on the latest updated path set.

3. The method according to claim 1, characterized in that: The constructing of a flow residual network based on the initial path set, the real physical connection between the road sections, the current flow, and the real flow of the target road section within a preset time interval includes: Based on the initial path set, determining a plurality of road sections corresponding to the target road section; Determining a flow residual based on the steering actual flow and the current flow; Based on the plurality of road sections and the flow residuals, determining a physical connection relationship between the road sections, a sampling probability corresponding to each of the road sections, an ID of each of the road sections, and a flow residual corresponding to the physical connection relationship; A flow residual network is constructed based on the physical connection relationship between the road sections, the sampling probability, the ID of each road section, and the flow residual corresponding to the physical connection relationship.

4. The method according to claim 3, characterized in that The determining the gain between each of the road sections based on the flow residual network and using the steering real flow, the current flow, and the initial road section sequence information includes: Based on the flow residual, the gain between each of the road sections is determined by utilizing the physical connection relationship between the road sections corresponding to the initial road section sequence information and the sampling probability corresponding to each of the road sections.

5. The method according to claim 4, characterized in that In response to the gain being in the first state, forming an updated path set according to the gain, updating the flow residual network based on the updated path set and the real physical connection between the road segments, and returning the step of determining the gain between each of the road segments based on the updated path set, comprises: In response to the existence of the gain exceeding the preset gain threshold, selecting a road segment sequence exceeding the preset gain threshold, and determining a sum of gains of a corresponding path sequence; Determining a gain target based on the gain sum; Based on the gain target, using the gain, traversing adjacent road segments in the road segment sequence information to form a newly added path; Adding the newly added path to the path set to form an updated path set; Based on the updated path set and the real physical connections between the road segments, the flow residual network is updated, and based on it, the step of determining the gain between each of the road segments is returned.

6. The method according to claim 1, characterized in that The method further comprises: Based on the vehicle identification information, the vehicle trajectory data is parsed to determine the matching trajectory; Based on the trajectory points in the matching trajectory, an initial path set is formed.

7. The method according to claim 6, characterized in that The method of parsing the vehicle trajectory data based on the vehicle identification information and determining the matching trajectory includes: Extract trajectory sequence based on vehicle identification information; The trajectory sequence is mapped to a road segment, and the trajectory sequence is filtered based on the road segment to determine a matching trajectory.

8. The method according to claim 6, characterized in that The forming an initial path set based on the trajectory points in the matching trajectory includes: Based on the trajectory points, the vehicle delay between adjacent trajectory points in time sequence is calculated; Based on each of the vehicle delays, determining a plurality of corresponding routes; Based on the plurality of said paths, an initial path set is formed.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method for determining a vehicle path according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method for determining a vehicle path according to any one of claims 1 to 7.

Images