Methods, systems, and electronic devices for identifying critical paths of vehicles

Abstract

This application relates to a method, system, and electronic device for identifying vehicle critical paths. It identifies vehicle types within a preset area using a total vehicle database, obtains a set of spatiotemporal trajectories for each vehicle type based on their spatiotemporal trajectories, and generates a similarity matrix for each type of vehicle based on their spatiotemporal trajectory sets. Using a clustering algorithm, based on the similarity matrix of each vehicle type's spatiotemporal trajectory, it divides each type of vehicle into different clusters, thereby obtaining the critical path for each type of vehicle. This solves the problem of inaccurate identification results and low efficiency caused by manual identification of vehicle critical paths, and improves the efficiency of vehicle critical path identification.

Description

Technical Field

[0001] This application relates to the field of urban road traffic technology, and in particular to methods, systems and electronic devices for identifying critical paths of vehicles. Background Technology

[0002] Alleviating traffic congestion has always been one of the most important challenges facing large cities, and refined, personalized congestion mitigation measures are increasingly valued. However, the prerequisite for developing such refined measures is the analysis of vehicle trajectories. For example, the increasingly popular customized bus services require understanding travelers' travel routes to better plan bus schedules and routes. However, current technologies typically rely on manual identification of vehicle critical paths, leading to inaccurate results and low efficiency.

[0003] Currently, no effective solution has been proposed to address the problem of inaccurate identification results and low efficiency caused by manually identifying the critical path of vehicles in related technologies. Summary of the Invention

[0004] This application provides a method, system, and electronic device for identifying the critical path of a vehicle, in order to at least solve the problem in the related art that the identification of the critical path of a vehicle by manual identification leads to inaccurate identification results and low efficiency.

[0005] In a first aspect, embodiments of this application provide a method for identifying a vehicle's critical path, the method comprising:

[0006] Identify the types of vehicles within a preset area using the total vehicle database;

[0007] Based on the spatiotemporal trajectory of the vehicle, obtain the spatiotemporal trajectory set of each type of vehicle, and based on the spatiotemporal trajectory set of each type of vehicle, obtain the spatiotemporal trajectory similarity matrix of each type of vehicle.

[0008] Using a clustering algorithm, based on the spatiotemporal trajectory similarity matrix of each type of vehicle, each type of vehicle is divided into different clusters to obtain the critical path of each type of vehicle.

[0009] In some embodiments, before identifying the type of vehicles within a preset area through a total vehicle database, the method further includes:

[0010] The total vehicle database will be divided into a taxi database, a ride-hailing database, a temporary service vehicle database, a frequently crossing vehicle database, and a remaining vehicle database using the following methods;

[0011] Based on license plate rules and travel frequency, taxis are identified, and the taxi database is constructed.

[0012] The scores for all vehicles are obtained based on the number of daily inspections, total travel time, total travel distance, and total nighttime travel time.

[0013] After filtering out taxis and ride-hailing vehicles based on the minimum and maximum scores of taxis, taxis are removed, and the ride-hailing vehicle database is constructed.

[0014] Temporary service vehicles are identified based on the number of travel days, and a database of temporary service vehicles is constructed. Frequently transiting vehicles are identified based on the average daily detection frequency, and a database of frequently transiting vehicles is constructed.

[0015] Excluding taxis, ride-hailing vehicles, temporary vehicles, and vehicles that frequently cross the border, the remaining vehicles are obtained. The remaining vehicle categories include commuter vehicles, family vehicles that are not frequently used, and company business vehicles.

[0016] A decision tree model is trained based on the feature vectors and category labels of the remaining vehicles to obtain a pre-trained decision tree model. Recognition rules for the categories of the remaining vehicles are obtained based on the pre-trained decision tree model. The database of the remaining vehicles is then obtained based on the recognition rules for the categories of the remaining vehicles.

[0017] In some embodiments, obtaining the scores for all vehicles based on the daily number of vehicle inspections, total travel time, total travel distance, and total nighttime travel time includes:

[0018] Based on the number of daily vehicle inspections, total travel time, total travel distance, and total nighttime travel time, a standardized positive eigenvalue matrix is ​​constructed.

[0019] Based on the standardized positive eigenvalue matrix, the distance between each vehicle and the optimal solution and the distance between each vehicle and the worst solution are obtained.

[0020] The score for each car is obtained based on the distance between each car and the optimal solution and the distance between each car and the worst solution.

[0021] In some embodiments, prior to training the decision tree model based on the feature vectors and category labels of the remaining vehicles, the method includes:

[0022] The number of clusters is determined based on the number of remaining vehicle categories. The remaining vehicles are then subjected to cluster analysis using an iterative self-organizing data analysis algorithm to obtain clustering results and clustering features. The clustering results are used as category labels, and the clustering features are used as feature vectors for the remaining vehicles.

[0023] The clustering features are the vehicle behavior features of the remaining vehicles, which include: number of travel days, average daily detection frequency, number of travel days during morning and evening peak hours, average daily detection frequency during morning and evening peak hours, number of travel days outside of morning and evening peak hours, and average daily detection frequency outside of morning and evening peak hours.

[0024] In some embodiments, obtaining the morning and evening peak time periods within the preset area includes the following steps:

[0025] Divide a day into a preset number of time intervals;

[0026] The proportion of vehicles first detected in each time interval is obtained out of the total number of vehicles. The time intervals in which the proportion is higher than a set threshold are the morning peak.

[0027] The system obtains the proportion of the last detected vehicles to the total number of vehicles in each time interval. Time intervals with a proportion higher than a set threshold are considered evening peak hours.

[0028] In some embodiments, obtaining the spatiotemporal trajectory set for each type of vehicle based on its spatiotemporal trajectory includes:

[0029] The spatiotemporal trajectory of the vehicle includes a spatiotemporal sequence consisting of the time the vehicle is detected during the day and its latitude and longitude position at the time of detection. The spatiotemporal trajectories of all vehicles of each type constitute the spatiotemporal trajectory set of each type of vehicle.

[0030] In some embodiments, obtaining the spatiotemporal trajectory similarity matrix for each type of vehicle based on the spatiotemporal trajectory set for each type of vehicle includes:

[0031] The spatiotemporal trajectory set includes a time trajectory set and a spatial trajectory set;

[0032] The similarity between any two time trajectories in the time trajectory set is obtained by using the dynamic time warping algorithm, and the time trajectory similarity matrix is ​​obtained based on the similarity between any two time trajectories.

[0033] The similarity between any two spatial trajectories in the spatial trajectory set is obtained by using the dynamic time warping algorithm, and the spatial trajectory similarity matrix is ​​obtained based on the similarity between any two spatial trajectories.

[0034] Based on the temporal trajectory similarity matrix and the spatial trajectory similarity matrix of each type of vehicle, the spatiotemporal trajectory similarity matrix of each type of vehicle is obtained.

[0035] In some embodiments, obtaining the spatiotemporal trajectory similarity matrix based on the temporal trajectory similarity matrix and the spatial trajectory similarity matrix includes:

[0036] The sum of the temporal trajectory similarity matrix and the spatial trajectory similarity matrix, which are multiples of the size scaling factor, constitutes the spatiotemporal trajectory similarity matrix.

[0037] Secondly, embodiments of this application provide a vehicle critical path identification system, the system comprising an identification module, an acquisition module, and a segmentation module.

[0038] The identification module is used to identify the type of vehicle within a preset area through the total vehicle database;

[0039] The acquisition module is used to obtain a spatiotemporal trajectory set for each type of vehicle based on the spatiotemporal trajectory of the vehicle, and to obtain a spatiotemporal trajectory similarity matrix for each type of vehicle based on the spatiotemporal trajectory set for each type of vehicle.

[0040] The partitioning module is used to divide each type of vehicle into different clusters based on the spatiotemporal trajectory similarity matrix of each type of vehicle using a clustering algorithm, thereby obtaining the critical path of each type of vehicle.

[0041] Thirdly, embodiments of this application provide an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the vehicle critical path identification method as described in the first aspect above.

[0042] Compared to related technologies, the vehicle critical path identification method provided in this application identifies the types of vehicles within a preset area through a total vehicle database. Based on the spatiotemporal trajectories of the vehicles, a spatiotemporal trajectory set for each type of vehicle is obtained. Based on the spatiotemporal trajectory set for each type of vehicle, a spatiotemporal trajectory similarity matrix for each type of vehicle is obtained. Using a clustering algorithm, based on the spatiotemporal trajectory similarity matrix for each type of vehicle, each type of vehicle is divided into different clusters to obtain the critical path for each type of vehicle. By analyzing the critical path for each type of vehicle, the method solves the problem of inaccurate identification results and low efficiency caused by manual identification of vehicle critical paths, and improves the identification efficiency of vehicle critical paths. Attached Figure Description

[0043] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0044] Figure 1 This is a flowchart of a method for identifying the critical path of a vehicle according to an embodiment of this application;

[0045] Figure 2 This is a structural block diagram of a vehicle critical path identification system according to an embodiment of this application. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this application clearer, the application is described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the application. All other embodiments obtained by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application. Furthermore, it is understood that although the efforts made in such a development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, modifications to design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as insufficient disclosure of the content of this application.

[0047] In this application, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that is mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0048] Unless otherwise defined, the technical or scientific terms used in this application shall have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms “a,” “an,” “an,” “the,” and similar words used in this application do not indicate quantity limitation and may indicate singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units not listed, or may include other steps or units inherent to these processes, methods, products, or devices. The terms “connected,” “linked,” “coupled,” and similar words used in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. “Multiple” used in this application means two or more. “And / or” describes the relationship between related objects, indicating that three relationships may exist; for example, “A and / or B” can represent: A alone, A and B simultaneously, and B alone. The terms “first,” “second,” “third,” etc., used in this application are merely to distinguish similar objects and do not represent a specific ordering of the objects.

[0049] This embodiment provides a method for identifying the critical path of a vehicle. Figure 1This is a flowchart of a vehicle critical path identification method according to an embodiment of this application, such as... Figure 1 As shown, the method includes the following steps:

[0050] Step S101: Identify the type of vehicle within the preset area through the total vehicle database. In this embodiment, the total vehicle database includes a database for each type of vehicle. After the database for each type of vehicle is pre-built, the vehicle type corresponding to the license plate number can be determined based on the total vehicle database when the license plate number is identified.

[0051] Step S102: Based on the spatiotemporal trajectory of the vehicle, obtain the spatiotemporal trajectory set of each type of vehicle, and based on the spatiotemporal trajectory set of each type of vehicle, obtain the spatiotemporal trajectory similarity matrix of each type of vehicle; specifically, when a vehicle is detected, record the time and spatial location of the vehicle detection, and the time and spatial location corresponding to the vehicle being detected within a day constitute the spatiotemporal trajectory of the vehicle.

[0052] Step S103: Using a clustering algorithm, based on the spatiotemporal trajectory similarity matrix of each type of vehicle, each type of vehicle is divided into different clusters to obtain the critical paths for each type of vehicle. In this embodiment, the clustering algorithm can be the DBSCAN algorithm, which is a density-based noise spatial clustering algorithm. According to this algorithm, vehicles with similar spatiotemporal trajectories among each type of vehicle are divided into a cluster, thus obtaining different clusters. Each cluster represents the critical path of each type of vehicle. By dividing the vehicle types, focusing on a certain travel mode, it avoids interference from other travel modes and reduces the heterogeneity of the sample dataset.

[0053] By analyzing the critical paths of each type of vehicle, refined and personalized congestion mitigation measures can be developed. For example, critical path analysis based on a commuter mode vehicle database can support the planning and optimization of customized bus routes. In addition, it can provide a theoretical basis for setting up green wave routes. Critical path analysis based on a taxi database can be used for taxi dispatching to prevent taxis from being overly concentrated on similar routes.

[0054] Through steps S101 to S103, compared with the problem in the related art that the key path of the vehicle is identified manually, resulting in inaccurate identification results and low efficiency, in this embodiment, the types of vehicles in the preset area are identified through the total vehicle database. According to the spatio-temporal trajectories of the vehicles, the spatio-temporal trajectory sets of each type of vehicle are obtained. According to the spatio-temporal trajectory sets of each type of vehicle, the spatio-temporal trajectory similarity matrix of each type of vehicle is obtained. Using the clustering algorithm, based on the spatio-temporal trajectory similarity matrix of each type of vehicle, each type of vehicle is divided into different clusters, and the key path of each type of vehicle is obtained. Based on the analysis of the key paths of each type of vehicle, refined and personalized traffic congestion alleviation measures can be formulated, solving the problem in the related art that the key path of the vehicle is identified manually, resulting in inaccurate identification results and low efficiency, improving the identification efficiency of the vehicle key path. Moreover, the vehicle key path identification method in this embodiment can be easily migrated to any spatio-temporal travel data, such as GPS trajectory data, LBS data, etc.

[0055] In some of these embodiments, before identifying the types of vehicles in the preset area through the total vehicle database, it is necessary to first construct the database of each type of vehicle to obtain the total vehicle database, which includes a taxi database, a network car-hailing database, a temporary business vehicle database, a frequently passing-through vehicle database, and a remaining vehicle database. The process of constructing the database of each type of vehicle is as follows:

[0056] First, identify taxis and network car-hailings on the road network. Taxis often have special license plate signs. For example, taxis in Hangzhou must be "浙AT****", that is, the license plate must contain the letter T, and taxis travel more frequently. Therefore, first identify taxis on the road network based on the license plate rule and travel frequency. If vehicle i meets the rules described in formulas 1 and 2, it is marked as a taxi, and a taxi database is constructed for all identified taxis.

[0057]

[0058] where, carnum i is the license plate number of the i-th vehicle, F ti is the average daily detection frequency of vehicle i, and θ is a manually set threshold, and the recommended value is the 95% quantile of the average daily detection frequency of all vehicles.

[0059] Extract four characteristic values of all vehicles for the identification of network car-hailings, including the number of daily detections, the total travel duration, the total travel distance, and the total travel duration at night. The unit of the number of daily detections is "times", the unit of the total travel duration is "seconds", the unit of the total travel distance is "kilometers", and the unit of the travel duration at night is "seconds".

[0060] Calculate and standardize the eigenvalue matrix for all vehicles, and then use the Top-Preference-by-Similarity to an Ideal Solution (TOPSIS) method to calculate the scores for all vehicles. The steps of TOPSIS are as follows:

[0061] 1. Construct the forward eigenvalue matrix. Assuming there are n cars and m forward-directed eigenvalues, the resulting forward eigenvalue matrix X is shown in Formula 3:

[0062]

[0063] 2. Standardize the normalized eigenvalue matrix X to eliminate the influence of dimensions. The standardized normalized eigenvalue matrix is ​​Z, where the element in the i-th row and j-th column of Z is Zi. ij The calculation method is shown in Formula 4:

[0064]

[0065] 3. Calculate the distances of each vehicle to the optimal and worst solutions. Since the optimal solution is the maximum value and the worst solution is the minimum value after forwarding, define the maximum value Z. + :

[0066]

[0067] Define the minimum value Z — :

[0068]

[0069] The distance between the i-th (i=1,2,...,n) vehicle and the maximum value is shown in Formula 5:

[0070]

[0071] The distance between the i-th (i=1,2,...,n) vehicle and the minimum value is shown in Formula 6:

[0072]

[0073] Where, ω j The index weight is the weight of the j-th feature. The index weight can be determined using methods such as entropy weighting or analytic hierarchy process, or it can be manually specified. The default value is 1.

[0074] 4. Calculate the score for each car based on its distance from both the optimal and worst solutions, denoted as G. i Then sort them as shown in Formula 7:

[0075]

[0076] For the taxi samples identified above, the score for each taxi sample was calculated and statistically analyzed to obtain the minimum score G for each taxi. min The maximum score is G max Since ride-hailing services and taxis exhibit similar travel characteristics, their scores are similar. After calculating the scores for all vehicle samples, the scores fall within the range of [G]. min G max The samples within are taxi samples and ride-hailing samples. Taxi samples are removed. If ride-hailing vehicles must be local vehicles, they are further filtered by the vehicle's license plate location to remove vehicles from other areas. The remaining vehicle samples are the ride-hailing samples, and a ride-hailing database is built for all identified ride-hailing vehicles.

[0077] To build the remaining vehicle database, the morning and evening peak hours for the preset area must first be identified. The first and last detection times for each vehicle's journey on each workday must be extracted. Then, the day is divided into 48 time intervals, each 30 minutes long, indexed by numbers 1-48. Here, 1 represents 00:00-00:30, 2 represents 00:30-01:00, ..., 48 represents 23:30-24:00. These integer values ​​are defined as the detection time indices. The proportions of vehicles first detected at each time index and the proportion of vehicles last detected at each time index are calculated. A threshold α (recommended value 3.5%) is set. Time periods where the proportion of first detected vehicles is higher than α are defined as the morning peak, and time periods where the proportion of last detected vehicles is higher than α are defined as the evening peak.

[0078] Based on license plate numbers, vehicle behavior characteristics were analyzed for all vehicles except taxis and ride-hailing vehicles, resulting in the extraction of six features, including: number of travel days N. t Daily average detection frequency F t Number of days for travel during morning and evening rush hours (N) tp Average daily testing frequency during morning and evening rush hours (F) tp Number of days for travel outside of morning and evening peak hours (N) to Average daily detection frequency F during off-peak hours to .

[0079] Set a threshold β (recommended value is 2), and set N t Vehicles with a value less than β are identified as temporary service vehicles, and a database of temporary service vehicles is constructed.

[0080] Set a threshold γ (recommended value is 2) to F tVehicles with a value ≤γ are identified as frequently transiting vehicles, and a database of frequently transiting vehicles is constructed.

[0081] After excluding taxis, ride-hailing vehicles, temporary vehicles, and vehicles frequently crossing the border, the remaining vehicles are randomly sampled, with N samples randomly selected. v (Recommended value is 10000) vehicles. Use the Iterative Self-Organizing Data Analysis Algorithm (ISODATA) to perform cluster analysis, obtain clustering results and clustering features. If the remaining vehicle categories include commuter vehicles, family vehicles that are not frequently used, and company business vehicles, then the recommended number of clusters k for the ISODATA clustering algorithm is set to 3.

[0082] Clustering features are the vehicle behavior features of the remaining vehicles. The clustering results are used as category labels, and the six clustering features are used as feature vectors. A decision tree model is trained to obtain a pre-trained decision tree model. The identification rules for the remaining vehicle categories are extracted through the pre-trained decision tree model. Based on the identification rules, the remaining vehicles are profiled to obtain a database of remaining vehicles. The database of remaining vehicles includes a database of commuter vehicles, a database of family vehicles that are not frequently used, and a database of company business vehicles.

[0083] In some embodiments, obtaining the spatiotemporal trajectory set for each type of vehicle based on its spatiotemporal trajectory includes:

[0084] For vehicles Its spatiotemporal trajectory S during the day i ={[l i1 ,t i1 ],...,[l ij ,t ij ],...,[l in ,t in ]}, where l ij Let time t j The vehicle's latitude and longitude location detected by detectors on the road network was detected a total of [number] times that day. Next, spacetime trajectory It can also be divided into time trajectory ST i and spatial trajectory SS i ST i ={[t i1 ,...,t iu ,...,t in ST j ={[t j1 ,...,t jw ,...,t jm Each traveling vehicle can have its spatiotemporal trajectory S discovered. iTherefore, the spatiotemporal trajectory set of a certain type of vehicle is D={S1,...,S...} i ,...,S N}, of which there are a total The spatiotemporal trajectory of a vehicle can be divided into a time trajectory set DT and a spatial trajectory set DS, where DT = {ST1,...,ST...} i ,...,ST N}, DS={SS1,...,SS i ,...,SS N}

[0085] In some embodiments, obtaining the spatiotemporal trajectory similarity matrix for each type of vehicle based on the spatiotemporal trajectory set for each type of vehicle includes:

[0086] Spatiotemporal trajectory similarity matrix M ST Including the time trajectory similarity matrix M T Similarity matrix M of spatial trajectories S Construct a time trajectory similarity matrix M T The process is as follows:

[0087] Assume the time trajectories of the two vehicles are ST and ST respectively. i and ST j ST i ={[t i1 ,...,t iu ,...,t in ST j ={[t j1 ,...,t jw ,...,t jm Then, the time trajectory elements are calculated using Formula 8. and The distance between them.

[0088]

[0089] Where sec(.) converts time to seconds, d uw For t iu and t jw The distance between them.

[0090] To obtain the similarity between two time trajectories using the Dynamic Time Warping (DTW) algorithm, an n*m distance matrix M is constructed based on the distance between the time trajectory elements of the two vehicles. n,m As shown in Formula 9:

[0091]

[0092] Obtain the distance matrix M n,m Then, the time trajectory ST is defined by a continuous sequence of elements W. i and ST j A mapping between, where the elements of sequence W are distance matrices M. n,m The elements in the sequence W are shown in Formula 10:

[0093]

[0094] Sequence W needs to satisfy the following three conditions:

[0095] 1. Boundary conditions: w1=(1,1), w K =(n,m), that is, W is from the distance matrix M n,m d 11 Starting with element d nm End of element.

[0096] 2. Continuity: If w k-1 If the path is (a, b), then for the next point w on the path... k =(a,b), where aa≤1 and bb≤1. This constraint requires that adjacent elements in W can only be distance matrices M. n,m Adjacent elements or diagonal elements in the array.

[0097] 3. Monotonicity: For w in condition 2 k and w k-1 The following constraints must be satisfied: 0 ≤ (aa) and 0 ≤ (bb), which make the elements in W monotonic in time.

[0098] The similarity between two time trajectories calculated based on the DTW algorithm is shown in Equation 11:

[0099]

[0100] ω k This represents the weight of the k-th element. Generally, ω=2 is set from (a-1,b-1) to (a,b), while ω=1 is set from (a-1,b) or (a,b-1) to (a,b). This setting is used to compensate for the different lengths of different element sequences.

[0101] The time trajectory similarity matrix M can be constructed using Formula 11. T As shown in Formula 12:

[0102]

[0103] Where, d tij For the similarity between the i-th time trajectory and the j-th time trajectory, there are a total of NA number of vehicles were used for travel.

[0104] Constructing the spatial trajectory similarity matrix M S The process is as follows:

[0105] Assume the spatial trajectories of the two traveling vehicles are SS i and SS j SS i ={[l i1 ,...,l iu ,...,l in ]}, SS j ={[l j1 ,...,l jw ,...,l jm ]}, then the spatial trajectory element l iu and l jw The distance between them is calculated using Formula 13:

[0106]

[0107] Among them, l iu For SS i The u-th element in l jw For SS j The w-th element in the expression, dist(.), represents the straight-line distance between two latitude and longitude points.

[0108] Based on the distance calculation formula for two spatial trajectory elements, the DTW algorithm can be applied to calculate the similarity between two spatial trajectories. The calculation steps are consistent with those for calculating the similarity of time trajectories, and a spatial trajectory similarity matrix M is constructed. S As shown in Formula 14:

[0109]

[0110] Where, d sij For the similarity between the i-th spatial trajectory and the j-th spatial trajectory, there are a total of A number of vehicles were used for travel.

[0111] Based on the temporal trajectory similarity matrix and the spatial trajectory similarity matrix of each type of vehicle, the spatiotemporal trajectory similarity matrix of each type of vehicle is obtained.

[0112] Optionally, the spatiotemporal trajectory similarity matrix M can be obtained based on the temporal trajectory similarity matrix and the spatial trajectory similarity matrix. ST include:

[0113] Size scaling factor multiple time trajectory similarity matrix M T Similarity matrix M with spatial trajectory SThe sum of these forms the time trajectory similarity matrix, as shown in Formula 15:

[0114]

[0115] in, This is the size scaling factor, which can be set to different values ​​according to the similarity of time trajectories and spatial trajectories in the actual analysis, so as to realize the different weight allocation of the time dimension and spatial dimension of the spatiotemporal trajectory similarity matrix.

[0116] It should be noted that the steps shown in the above process or in the flowchart of the accompanying figures 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 may be executed in a different order than that shown here.

[0117] This embodiment also provides a vehicle critical path identification system, which is used to implement the above embodiments and preferred embodiments, and will not be repeated as described previously. As used below, the terms "module," "unit," "subunit," etc., can refer to a combination of software and / or hardware that performs a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0118] Figure 2 This is a structural block diagram of a vehicle critical path identification system according to an embodiment of this application, such as... Figure 2 As shown, the system includes an identification module 21, an acquisition module 22, and a segmentation module 23. The identification module 21 is used to identify the types of vehicles within a preset area through a total vehicle database. The acquisition module 22 is used to obtain the spatiotemporal trajectory set of each type of vehicle based on the spatiotemporal trajectory of the vehicles, and to obtain the spatiotemporal trajectory similarity matrix of each type of vehicle based on the spatiotemporal trajectory set of each type of vehicle. The segmentation module 23 is used to use a clustering algorithm to divide each type of vehicle into different clusters based on the spatiotemporal trajectory similarity matrix of each type of vehicle, and to obtain the critical path of each type of vehicle. Based on the analysis of the critical path of each type of vehicle, refined and personalized congestion mitigation measures can be formulated, which solves the problem of inaccurate identification results and low efficiency caused by manually identifying the critical path of vehicles in related technologies.

[0119] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0120] This embodiment also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0121] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0122] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated here.

[0123] Furthermore, in conjunction with the vehicle critical path identification method in the above embodiments, this application embodiment can provide a storage medium for implementation. This storage medium stores a computer program; when executed by a processor, the computer program implements any of the vehicle critical path identification methods in the above embodiments.

[0124] In one embodiment, a computer device is provided, which may be a terminal. The computer device includes a processor, memory, a network interface, a display screen, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, it implements a method for identifying critical paths in a vehicle. The display screen may be a liquid crystal display (LCD) or an e-ink display. The input device may be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0125] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0126] Those skilled in the art should understand that the technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments have been described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0127] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for identifying the critical path of a vehicle, characterized in that, The method includes: The types of vehicles within a preset area are identified by the total vehicle database. The total vehicle database will be divided into a taxi database, a ride-hailing database, a temporary service vehicle database, a frequently crossing vehicle database, and a remaining vehicle database using the following methods. Based on license plate rules and travel frequency, taxis are identified, and the taxi database is constructed. Based on the number of daily inspections, total travel time, total travel distance, and total nighttime travel time of vehicles, the score of all vehicles is obtained using the superior-inferior solution distance method. After filtering out taxis and ride-hailing vehicles based on the minimum and maximum scores of taxis, taxis are removed, and the ride-hailing vehicle database is constructed. Temporary service vehicles are identified based on the number of travel days, and a database of temporary service vehicles is constructed. Frequently transiting vehicles are identified based on the average daily detection frequency, and a database of frequently transiting vehicles is constructed. Excluding taxis, ride-hailing vehicles, temporary vehicles, and vehicles that frequently cross the border, the remaining vehicles are obtained. The remaining vehicle categories include commuter vehicles, family vehicles that are not frequently used, and company business vehicles. A decision tree model is trained based on the feature vectors and category labels of the remaining vehicles to obtain a pre-trained decision tree model. The identification rules for the remaining vehicle categories are obtained based on the pre-trained decision tree model. The remaining vehicle database is obtained based on the identification rules for the remaining vehicle categories. Based on the spatiotemporal trajectory of the vehicle, obtain the spatiotemporal trajectory set for each type of vehicle, and based on the spatiotemporal trajectory set for each type of vehicle, obtain the spatiotemporal trajectory similarity matrix for each type of vehicle. Using a clustering algorithm, based on the spatiotemporal trajectory similarity matrix of each type of vehicle, each type of vehicle is divided into different clusters to obtain the critical path of each type of vehicle.

2. The method according to claim 1, characterized in that, The scores for all vehicles, calculated based on the number of daily inspections, total travel time, total travel distance, and total nighttime travel time, include: Based on the number of daily vehicle inspections, total travel time, total travel distance, and total nighttime travel time, a standardized positive eigenvalue matrix is ​​constructed. Based on the standardized positive eigenvalue matrix, the distance between each vehicle and the optimal solution and the distance between each vehicle and the worst solution are obtained. The score for each vehicle is obtained based on the distance between each vehicle and the optimal solution and the distance between each vehicle and the worst solution.

3. The method according to claim 1, characterized in that, Before training the decision tree model based on the feature vectors and category labels of the remaining vehicles, the method includes: The number of clusters is determined based on the number of remaining vehicle categories. The remaining vehicles are then subjected to cluster analysis using an iterative self-organizing data analysis algorithm to obtain clustering results and clustering features. The clustering results are used as category labels, and the clustering features are used as feature vectors for the remaining vehicles. The clustering features are the vehicle behavior features of the remaining vehicles, which include: number of travel days, average daily detection frequency, number of travel days during morning and evening peak hours, average daily detection frequency during morning and evening peak hours, number of travel days outside of morning and evening peak hours, and average daily detection frequency outside of morning and evening peak hours.

4. The method according to claim 3, characterized in that, Obtaining the morning and evening peak hours within the preset area includes the following steps: Divide a day into a preset number of time intervals; The proportion of vehicles first detected in each time interval is obtained out of the total number of vehicles. The time intervals in which the proportion is higher than a set threshold are the morning peak. The system obtains the proportion of the last detected vehicles to the total number of vehicles in each time interval. Time intervals with a proportion higher than a set threshold are considered evening peak hours.

5. The method according to claim 1, characterized in that, The process of obtaining the spatiotemporal trajectory set for each type of vehicle based on its spatiotemporal trajectory includes: The spatiotemporal trajectory of the vehicle includes a spatiotemporal sequence consisting of the time the vehicle is detected during the day and its latitude and longitude position at the time of detection. The spatiotemporal trajectories of all vehicles of each type constitute the spatiotemporal trajectory set of each type of vehicle.

6. The method according to claim 1, characterized in that, The step of obtaining the spatiotemporal trajectory similarity matrix for each type of vehicle based on the spatiotemporal trajectory set for each type of vehicle includes: The spatiotemporal trajectory set includes a time trajectory set and a spatial trajectory set; The similarity between any two time trajectories in the time trajectory set is obtained by using the dynamic time warping algorithm, and the time trajectory similarity matrix is ​​obtained based on the similarity between any two time trajectories. The similarity between any two spatial trajectories in the spatial trajectory set is obtained by using the dynamic time warping algorithm, and the spatial trajectory similarity matrix is ​​obtained based on the similarity between any two spatial trajectories. Based on the temporal trajectory similarity matrix and the spatial trajectory similarity matrix of each type of vehicle, the spatiotemporal trajectory similarity matrix of each type of vehicle is obtained.

7. The method according to claim 6, characterized in that, Based on the temporal trajectory similarity matrix and the spatial trajectory similarity matrix, the spatiotemporal trajectory similarity matrix is ​​obtained as follows: The sum of the temporal trajectory similarity matrix and the spatial trajectory similarity matrix, which are multiples of the size scaling factor, constitutes the spatiotemporal trajectory similarity matrix.

8. A vehicle critical path identification system, characterized in that, The system includes an identification module, an acquisition module, and a segmentation module. The identification module is used to identify the type of vehicles in a preset area through the total vehicle database. The total vehicle database will be divided into a taxi database, a ride-hailing database, a temporary service vehicle database, a frequently crossing vehicle database, and a remaining vehicle database using the following method. Based on license plate rules and travel frequency, taxis are identified, and the taxi database is constructed. Based on the number of daily inspections, total travel time, total travel distance, and total nighttime travel time of vehicles, the score of all vehicles is obtained using the superior-inferior solution distance method. After filtering out taxis and ride-hailing vehicles based on the minimum and maximum scores of taxis, taxis are removed, and the ride-hailing vehicle database is constructed. Temporary service vehicles are identified based on the number of travel days, and a database of temporary service vehicles is constructed. Frequently transiting vehicles are identified based on the average daily detection frequency, and a database of frequently transiting vehicles is constructed. Excluding taxis, ride-hailing vehicles, temporary vehicles, and vehicles that frequently cross the border, the remaining vehicles are obtained. The remaining vehicle categories include commuter vehicles, family vehicles that are not frequently used, and company business vehicles. A decision tree model is trained based on the feature vectors and category labels of the remaining vehicles to obtain a pre-trained decision tree model. The identification rules for the remaining vehicle categories are obtained based on the pre-trained decision tree model. The remaining vehicle database is obtained based on the identification rules for the remaining vehicle categories. The acquisition module is used to obtain a spatiotemporal trajectory set for each type of vehicle based on the spatiotemporal trajectory of the vehicle, and to obtain a spatiotemporal trajectory similarity matrix for each type of vehicle based on the spatiotemporal trajectory set for each type of vehicle. The partitioning module is used to divide each type of vehicle into different clusters based on the spatiotemporal trajectory similarity matrix of each type of vehicle using a clustering algorithm, thereby obtaining the critical path of each type of vehicle.

9. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to perform the vehicle critical path identification method according to any one of claims 1 to 7.

Images