A method for intersection scene recognition and a computer-readable storage medium

Abstract

This application provides an intersection scene recognition method and a computer-readable storage medium, relating to the field of traffic signal control technology. The method includes: acquiring target vehicle passage data for each target lane in each direction of a target intersection during a target green light period within a control time period; obtaining a first feature to be identified, a second feature to be identified, and a third feature to be identified based on the target vehicle passage data; determining the traffic flow discrete state of the target lane based on the first feature to be identified, the traffic demand state of the target lane based on the second feature to be identified, and the traffic operation state of the target lane based on the third feature to be identified; and determining the scene type of the target intersection within the control time period based on the traffic flow discrete state, traffic demand state, and traffic operation state of the target lanes in each direction. Thus, scene recognition of the intersection can be performed based on the intersection's vehicle passage data.

Description

Technical Field

[0001] This application relates to the field of traffic signal control technology, and in particular to a method for intersection scene recognition and a computer-readable storage medium. Background Technology

[0002] In the field of traffic signal control technology, due to the influence of human life patterns, the traffic conditions at the same intersection may vary at different times of the day. For example, in urban roads, the number of vehicles traveling at intersections during the morning rush hour (such as 7:00-9:00) is usually greater than the number of vehicles traveling at intersections during other times (such as 9:00-11:00).

[0003] However, if the same traffic signal is used for intersections with different traffic conditions, there may be a mismatch between the traffic signal and the traffic conditions, resulting in traffic jams due to excessively short green light times or empty green light times.

[0004] Therefore, there is an urgent need for an intersection scene recognition method that can identify the scene of the intersection based on the vehicle passage data. Subsequently, appropriate traffic signal control strategies can be adopted at the intersection according to the scene. Summary of the Invention

[0005] The purpose of this application is to provide a method for intersection scene recognition and a computer-readable storage medium, so as to perform scene recognition of the intersection based on the vehicle passage data of the intersection. The specific technical solution is as follows:

[0006] A first aspect of this application provides a method for recognizing intersection scenes, the method comprising:

[0007] For each target lane in each direction of the target intersection, obtain the target vehicle passage data of that target lane during the target green light period within the time period to be controlled;

[0008] Based on the target vehicle passage data, a first feature to be identified is obtained, representing the probability distribution of the headway distance between vehicles in the target lane during the target green light period; a second feature to be identified is obtained, representing the change in the number of vehicles allowed to pass in the target lane during each statistical time period of the target green light period; and a third feature to be identified is obtained, representing the probability distribution of the vehicle speed in the target lane during the target green light period.

[0009] The traffic flow discrete state of the target lane is determined based on the first feature to be identified, the traffic demand state of the target lane is determined based on the second feature to be identified, and the traffic operation state of the target lane is determined based on the third feature to be identified.

[0010] Based on the discrete state of traffic flow, traffic demand, and traffic operation status of the target lanes in each direction, the scenario type of the target intersection during the time period to be controlled is determined.

[0011] In some embodiments, during the control period, the target lane has the highest number of vehicles allowed to pass in each lane of any direction of the road segment at the target intersection.

[0012] In some embodiments, before determining the traffic flow discrete state of the target lane based on the first feature to be identified, the method further includes: for each preset traffic flow discrete state, obtaining first sample vehicle passage data of the first sample lane during a first sample green light when the first sample lane is in the preset traffic flow discrete state; and obtaining a first sample feature representing the probability distribution of the headway in the first sample lane during the first sample green light based on the first sample vehicle passage data.

[0013] The step of determining the discrete state of traffic flow of the target lane based on the first feature to be identified includes: for each preset discrete state of traffic flow, calculating the similarity between the first sample feature of the preset discrete state of traffic flow and the first feature to be identified using a first similarity algorithm; wherein the first similarity algorithm includes at least one of the following: bulldozer distance algorithm and Mahalanobis distance algorithm; and determining the preset discrete state of traffic flow with the highest similarity as the discrete state of traffic flow of the target lane.

[0014] In some embodiments, obtaining a first sample feature representing the probability distribution of headway distances in the first sample lane during the first sample green light period based on the first sample vehicle passage data includes: determining, based on the first sample vehicle passage data, each headway distance at a specified position in the first sample lane during the first sample green light period as a sample headway distance; determining a preset headway distance interval to which each sample headway distance belongs; wherein each preset headway distance interval is adjacent and has the same length; and for each preset headway distance interval, calculating the ratio of sample headway distances belonging to that preset headway distance interval to all sample headway distances to obtain the first sample feature.

[0015] The step of obtaining a first feature to be identified based on the target vehicle passage data, representing the probability distribution of headway distances in the target lane during the target green light period, includes: determining, based on the target vehicle passage data, the headway distances at specified positions in the target lane during the target green light period as headway distances to be processed; determining a preset headway distance interval to which each headway distance to be processed belongs; and for each preset headway distance interval, calculating the ratio of headway distances to be processed belonging to that preset headway distance interval to all headway distances to be processed, thereby obtaining the first feature to be identified.

[0016] In some embodiments, before determining the traffic demand state of the target lane based on the second feature to be identified, the method further includes: for each preset traffic demand state, obtaining second sample vehicle passage data of the second sample lane during a second sample green light when the second sample lane is in the preset traffic demand state; and obtaining a second sample feature based on the second sample vehicle passage data, representing the change in the number of vehicles allowed to pass through the second sample lane in each statistical time period during the second sample green light.

[0017] The step of determining the traffic demand state of the target lane based on the second feature to be identified includes: for each preset traffic demand state, calculating the similarity between the second sample feature of the preset traffic demand state and the second feature to be identified; and determining the preset traffic demand state with the highest similarity as the traffic demand state of the target lane.

[0018] In some embodiments, the second sample green light period includes multiple sample green light cycles; the target green light period includes multiple target green light cycles;

[0019] The step of obtaining a second sample feature based on the second sample vehicle passage data, representing the change in the number of vehicles released in the second sample lane during each statistical time period of the second sample green light period, includes: for each sample green light cycle, determining the number of vehicles released in the second sample lane during each statistical time period of the sample green light cycle based on the second sample vehicle passage data of that sample green light cycle, as the number of vehicles released during each statistical time period of the sample green light cycle; for any statistical time period, calculating the average number of vehicles released during that statistical time period of each sample green light cycle, to obtain the second sample feature representing the change in the number of vehicles released in the second sample lane during each statistical time period of the second sample green light period;

[0020] The step of obtaining a second identifying feature based on the target vehicle passage data, representing the change in the number of vehicles released in the target lane within each statistical time period of the target green light period, includes: for each target green light cycle, determining the number of vehicles released in the target lane within each statistical time period of the target green light cycle based on the target vehicle passage data of that target green light cycle, as the number of vehicles released within each statistical time period of the target green light cycle; for any statistical time period, calculating the average number of vehicles released within that statistical time period of each target green light cycle, to obtain the second identifying feature representing the change in the number of vehicles released in the target lane within each statistical time period of the target green light period.

[0021] In some embodiments, before determining the traffic operation state of the target lane based on the third feature to be identified, the method further includes: for each preset traffic operation state, acquiring third sample vehicle passage data of the third sample lane during the third sample green light period when the third sample lane is in the preset traffic operation state; and obtaining a third sample feature representing the probability distribution of vehicle speeds in the third sample lane during the third sample green light period based on the third sample vehicle passage data.

[0022] The step of determining the traffic operation status of the target lane based on the third feature to be identified includes: for each preset traffic operation status, calculating the similarity between the third sample feature of the preset traffic operation status and the third feature to be identified using a second similarity algorithm; wherein the second similarity algorithm includes at least one of the following: bulldozing distance algorithm and Mahalanobis distance algorithm; and determining the preset traffic operation status with the highest similarity as the traffic operation status of the target lane.

[0023] In some embodiments, obtaining a third sample feature representing the probability distribution of vehicle speeds in the third sample lane during the third sample green light period based on the third sample vehicle passage data includes: determining the speed of each vehicle in the third sample lane during the third sample green light period based on the third sample vehicle passage data, as a sample speed; determining a preset speed interval to which each sample speed belongs; wherein the preset speed intervals are adjacent and have the same length; and for each preset speed interval, calculating the ratio of the sample speeds belonging to that preset speed interval to all sample speeds to obtain the third sample feature.

[0024] The step of obtaining a third feature to be identified based on the target vehicle passage data, representing the probability distribution of vehicle speeds in the target lane during the target green light period, includes: determining the speed of each vehicle in the target lane during the target green light period based on the target vehicle passage data, as the speed to be processed; determining a preset speed range to which each speed to be processed belongs; and for each preset speed range, calculating the ratio of the speed to be processed belonging to that preset speed range among all speeds to be processed, to obtain the third feature to be identified.

[0025] In some embodiments, the traffic flow discrete state is: discrete, weak car-following, or strong car-following; the traffic demand state is: low traffic demand or high traffic demand; the traffic operation state is: slow-moving or smooth-flowing; determining the scenario type of the target intersection within the control time period based on the traffic flow discrete state, traffic demand state, and traffic operation state of the target lanes in each direction includes:

[0026] If at least one target lane in each direction of the target intersection meets the first screening condition, then the scenario type of the target intersection during the control period is determined to be queuing equilibrium type; wherein, the first screening condition indicates that the traffic flow discrete state is strong car-following, the traffic demand state is high traffic demand, and the traffic operation state is slow. If none of the target lanes in each direction of the target intersection meet the first screening condition, and at least one target lane meets the second screening condition, then the scenario type of the target intersection during the control period is determined to be capacity type; wherein, the second screening condition indicates that the traffic flow discrete state is strong car-following or weak car-following, the traffic demand state is high traffic demand, and the traffic operation state is smooth. If none of the target lanes in each direction of the target intersection meet the first screening condition, and none of the target lanes meet the second screening condition, then the scenario type of the target intersection during the control period is determined to be efficiency improvement type.

[0027] In some embodiments, the method further includes:

[0028] If the scenario type of the target intersection during the control period is queue balancing, a first type of traffic signal control strategy is adopted at the target intersection; wherein, the first type of traffic signal control strategy is used to shorten the queue length of vehicles at the target intersection. If the scenario type of the target intersection during the control period is capacity-related, a second type of traffic signal control strategy is adopted at the target intersection; wherein, the second type of traffic signal control strategy is used to improve the vehicle capacity of the target intersection. If the scenario type of the target intersection during the control period is efficiency-improvement, a third type of traffic signal control strategy is adopted at the target intersection; wherein, the third type of traffic signal control strategy is used to reduce the delay time of vehicles at the target intersection.

[0029] A second aspect of this application provides an intersection scene recognition device, the device comprising:

[0030] The target vehicle passage data acquisition module is used to acquire target vehicle passage data for each target lane in each direction of the target intersection during the target green light period within the control time period. The target feature acquisition module is used to obtain, based on the target vehicle passage data, a first target feature representing the probability distribution of headway distance in the target lane during the target green light period, a second target feature representing the change in the number of vehicles allowed to pass in the target lane within each statistical time period of the target green light period, and a third target feature representing the probability distribution of vehicle speed in the target lane during the target green light period. The state determination module is used to determine the discrete traffic flow state of the target lane based on the first target feature, the traffic demand state of the target lane based on the second target feature, and the traffic operation state of the target lane based on the third target feature. The scenario determination module is used to determine the scenario type of the target intersection within the control time period based on the discrete traffic flow state, traffic demand state, and traffic operation state of the target lanes in each direction.

[0031] In some embodiments, during the control period, the target lane has the highest number of vehicles allowed to pass in each lane of any direction of the road segment at the target intersection.

[0032] In some embodiments, the apparatus further includes: a first sample vehicle passage data acquisition module, configured to, before determining the traffic flow discrete state of the target lane based on the first feature to be identified, acquire, for each preset traffic flow discrete state, first sample vehicle passage data of the first sample lane during a first sample green light period when the first sample lane is in the preset traffic flow discrete state; a first sample feature acquisition module, configured to, based on the first sample vehicle passage data, obtain a first sample feature representing the probability distribution of the headway distance in the first sample lane during the first sample green light period; the state determination module is specifically configured to: for each preset traffic flow discrete state, calculate the similarity between the first sample feature of the preset traffic flow discrete state and the first feature to be identified using a first similarity algorithm; wherein, the first similarity algorithm includes at least one of the following: bulldozing distance algorithm and Mahalanobis distance algorithm; and determine the preset traffic flow discrete state with the highest similarity as the traffic flow discrete state of the target lane.

[0033] In some embodiments, the sample feature acquisition module is specifically used for: determining, based on the first sample vehicle passage data, the headway of each vehicle at a specified position in the first sample lane during the first sample green light period, as sample headway; determining a preset headway interval to which each sample headway belongs; wherein, each preset headway interval is adjacent and has the same length; for each preset headway interval, calculating the ratio of the sample headway belonging to that preset headway interval to all sample headway intervals, to obtain a first sample feature; the feature acquisition module to be identified is specifically used for: determining, based on the target vehicle passage data, the headway of each vehicle at a specified position in the target lane during the target green light period, as headway to be processed; determining a preset headway interval to which each headway to be processed belongs; for each preset headway interval, calculating the ratio of the headway to be processed belonging to that preset headway interval to all headway intervals, to obtain a first feature to be identified.

[0034] In some embodiments, the apparatus further includes: a second sample vehicle passage data acquisition module, configured to, before determining the traffic demand state of the target lane based on the second feature to be identified, acquire, for each preset traffic demand state, second sample vehicle passage data of the second sample lane during a second sample green light period when the second sample lane is in the preset traffic demand state; a second sample feature acquisition module, configured to, based on the second sample vehicle passage data, obtain a second sample feature representing the change in the number of vehicles allowed to pass through the second sample lane in each statistical time period during the second sample green light period; the state determination module is specifically configured to: for each preset traffic demand state, calculate the similarity between the second sample feature of the preset traffic demand state and the second feature to be identified; and determine the preset traffic demand state with the highest similarity as the traffic demand state of the target lane.

[0035] In some embodiments, the second sample green light period includes multiple sample green light cycles; the target green light period includes multiple target green light cycles; the second sample feature acquisition module is specifically used for: for each sample green light cycle, based on the second sample vehicle passage data of the sample green light cycle, determining the number of vehicles released in the second sample lane within each statistical time period of the sample green light cycle, as the number of vehicles released within each statistical time period of the sample green light cycle; for any statistical time period, calculating the average number of vehicles released within that statistical time period of each sample green light cycle, to obtain a second sample feature representing the change in the number of vehicles released in the second sample lane within each statistical time period of the second sample green light period; the feature to be identified acquisition module is specifically used for: for each target green light cycle, based on the target vehicle passage data of the target green light cycle, determining the number of vehicles released in the target lane within each statistical time period of the target green light cycle, as the number of vehicles released within each statistical time period of the target green light cycle; for any statistical time period, calculating the average number of vehicles released within that statistical time period of each target green light cycle, to obtain a second feature representing the change in the number of vehicles released in the target lane within each statistical time period of the target green light period.

[0036] In some embodiments, the apparatus further includes: a third sample vehicle passage data acquisition module, configured to, for each preset traffic operation state, acquire third sample vehicle passage data of the third sample lane during a third sample green light period when the third sample lane is in the preset traffic operation state, before determining the traffic operation state of the target lane based on the third feature to be identified; a third sample feature acquisition module, configured to, based on the third sample vehicle passage data, obtain a third sample feature representing the probability distribution of vehicle speeds in the third sample lane during the third sample green light period; the state determination module is specifically configured to: for each preset traffic operation state, calculate the similarity between the third sample feature of the preset traffic operation state and the third feature to be identified using a second similarity algorithm; wherein the second similarity algorithm includes at least one of the following: bulldozing distance algorithm and Mahalanobis distance algorithm; and determine the preset traffic operation state with the highest similarity as the traffic operation state of the target lane.

[0037] In some embodiments, the third sample feature acquisition module is specifically used for: determining the speed of each vehicle in the third sample lane during the third sample green light period based on the third sample vehicle passage data, as a sample driving speed; determining a preset driving speed interval to which each sample driving speed belongs; wherein, each preset driving speed interval is adjacent and has the same length; for each preset driving speed interval, calculating the ratio of the sample driving speed belonging to that preset driving speed interval to all sample driving speeds, to obtain the third sample feature; the feature acquisition module to be identified is specifically used for: determining the speed of each vehicle in the target lane during the target green light period based on the target vehicle passage data, as a driving speed to be processed; determining a preset driving speed interval to which each driving speed to be processed belongs; for each preset driving speed interval, calculating the ratio of the driving speed to be processed belonging to that preset driving speed interval to all driving speeds to be processed, to obtain the third sample feature.

[0038] In some embodiments, the traffic flow discrete state is: discrete, weak car-following, or strong car-following; the traffic demand state is: low traffic demand or high traffic demand; the traffic operation state is: slow or smooth; the scenario determination module is specifically used to: if at least one target lane in each direction of the target intersection meets the first screening condition, then determine the scenario type of the target intersection in the control time period as queue equilibrium type; wherein, the first screening condition indicates: the traffic flow discrete state is strong car-following, the traffic demand state is high traffic demand, and the traffic operation state is slow; if the target lane in each direction of the target intersection meets the first screening condition, then determine the scenario type of the target intersection in the control time period as queue equilibrium type. If there are no target lanes that meet the first screening condition, and at least one target lane that meets the second screening condition, then the scenario type of the target intersection during the control period is determined to be capacity-related. The second screening condition indicates that the traffic flow discrete state is strong or weak car-following, the traffic demand state is high traffic demand, and the traffic operation state is smooth. If there are no target lanes that meet the first screening condition and no target lanes that meet the second screening condition in any direction of the target intersection, then the scenario type of the target intersection during the control period is determined to be efficiency-enhancing.

[0039] In some embodiments, the apparatus further includes: a first control strategy determination module, configured to adopt a first type of traffic signal control strategy at the target intersection if the scenario type of the target intersection during the control period is queue balancing; wherein the first type of traffic signal control strategy is used to shorten the queue length of vehicles at the target intersection; a second control strategy determination module, configured to adopt a second type of traffic signal control strategy at the target intersection if the scenario type of the target intersection during the control period is capacity-related; wherein the second type of traffic signal control strategy is used to improve the vehicle capacity of the target intersection; and a third control strategy determination module, configured to adopt a third type of traffic signal control strategy at the target intersection if the scenario type of the target intersection during the control period is efficiency-improvement; wherein the third type of traffic signal control strategy is used to reduce the delay time of vehicles at the target intersection.

[0040] A third aspect of this application provides an electronic device, including:

[0041] Memory, used to store computer programs;

[0042] The processor, when executing a program stored in memory, implements any of the intersection scene recognition methods described above.

[0043] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the intersection scene recognition methods described above.

[0044] This application also provides a computer program product containing instructions that, when run on a computer, cause the computer to execute any of the intersection scene recognition methods described above.

[0045] Beneficial effects of the embodiments in this application:

[0046] This application provides an intersection scene recognition method, which includes: acquiring target vehicle passage data for each target lane in each direction of the target intersection during the target green light period within the control time period; based on the target vehicle passage data, obtaining a first feature to be identified representing the probability distribution of headway in the target lane during the target green light period, a second feature to be identified representing the change in the number of vehicles released in the target lane during each statistical time period of the target green light period, and a third feature to be identified representing the probability distribution of vehicle speed in the target lane during the target green light period; determining the traffic flow discrete state of the target lane based on the first feature to be identified, determining the traffic demand state of the target lane based on the second feature to be identified, and determining the traffic operation state of the target lane based on the third feature to be identified; and determining the scene type of the target intersection within the control time period based on the traffic flow discrete state, traffic demand state, and traffic operation state of the target lanes in each direction.

[0047] Based on the above processing, for each target lane in each direction of the target intersection, the first feature to be identified, the second feature to be identified, and the third feature to be identified can be obtained based on the target vehicle data of the target lane during the target green light period.

[0048] Since the first feature to be identified represents the probability distribution of headway distances in the target lane during the target green light period, and this probability distribution reflects the proportion of vehicles in the traffic flow that are constrained to varying degrees by the vehicle in front, the discrete state of traffic flow in the target lane can be determined based on the first feature to be identified. Since the second feature to be identified represents the change in the number of vehicles allowed to pass in the target lane during each statistical time period of the target green light period, and this change in number reflects the queue length of vehicles in the target lane, the traffic demand state of the target lane can be determined based on the second feature to be identified. Since the third feature to be identified represents the probability distribution of vehicle speeds in the target lane during the target green light period, and this probability distribution reflects the proportion of vehicles with different speeds in the traffic flow, the traffic operation state of the target lane can be determined based on the third feature to be identified.

[0049] Furthermore, for any target lane, it can be described from three aspects: traffic flow discrete state, traffic demand state, and traffic operation state. This allows us to determine the scenario type of the target intersection during the control period based on the traffic flow discrete state, traffic demand state, and traffic operation state of the target lanes in each direction. Thus, it is possible to obtain the scenario type of the target intersection during the control period based on vehicle passage data.

[0050] Of course, implementing any product or method of this application does not necessarily require achieving all of the advantages described above at the same time. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0052] Figure 1 This is a first flowchart of the intersection scene recognition method provided in the embodiments of this application;

[0053] Figure 2 A flowchart illustrating how to obtain a first feature to be identified based on target vehicle data is provided in this embodiment of the application.

[0054] Figure 3 A flowchart illustrating how to obtain a second feature to be identified based on target vehicle data is provided in this application embodiment;

[0055] Figure 4 A flowchart illustrating how to obtain a third feature to be identified based on target vehicle data is provided in this embodiment of the application.

[0056] Figure 5 A schematic diagram of a process for obtaining the first, second, and third features to be identified of a target road segment in each direction, provided in an embodiment of this application;

[0057] Figure 6 A flowchart for determining the discrete state of traffic flow of a target lane based on a first feature to be identified is provided in an embodiment of this application.

[0058] Figure 7 A flowchart illustrating how to determine the traffic demand state of a target lane based on a second feature to be identified, as provided in this application embodiment;

[0059] Figure 8 A flowchart illustrating how to determine the traffic operation status of a target lane based on a third feature to be identified, as provided in this application embodiment;

[0060] Figure 9 This is a second flowchart of the intersection scene recognition method provided in the embodiments of this application;

[0061] Figure 10(a) is a schematic diagram of lanes under different discrete traffic flow states provided in the embodiments of this application;

[0062] Figure 10(b1) is a schematic diagram showing the change in the number of vehicles released in a lane under low traffic demand according to an embodiment of this application;

[0063] Figure 10(b2) is a schematic diagram showing the change in the number of vehicles released in a lane under high traffic demand according to an embodiment of this application;

[0064] Figure 10(c) is a schematic diagram of roads under different traffic operation states provided in the embodiments of this application;

[0065] Figure 11 This is a schematic diagram of the structure of a control strategy determination device provided in an embodiment of this application;

[0066] Figure 12 A structural diagram of an intersection scene recognition device provided in an embodiment of this application;

[0067] Figure 13 This is a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0068] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of this application.

[0069] In the field of traffic signal control technology, due to the influence of human life patterns, the traffic conditions at the same intersection may vary at different times of the day. For example, in urban roads, the number of vehicles traveling at intersections during the morning rush hour (such as 7:00-9:00) is usually greater than the number of vehicles traveling at intersections during other times (such as 9:00-11:00).

[0070] However, if the same traffic signal is used for intersections with different traffic conditions, there may be a mismatch between the traffic signal and the traffic conditions, resulting in traffic jams due to excessively short green light times or empty green light times.

[0071] Therefore, there is an urgent need for an intersection scene recognition method that can identify the scene of the intersection based on the vehicle passage data. Subsequently, appropriate traffic signal control strategies can be adopted at the intersection according to the scene.

[0072] This application provides an intersection scene recognition method. This method can be applied to an electronic device, which can be a traffic signal control system, or the electronic device can communicate with the traffic signal control system and acquire vehicle passage data of the intersection controlled by the traffic signal control system.

[0073] See Figure 1 , Figure 1 This is a first flowchart of an intersection scene recognition method provided in an embodiment of this application. The method includes the following steps:

[0074] S101: For each direction of the target lane at the target intersection, obtain the target vehicle passage data during the target green light period within the control time period for that target lane.

[0075] S102: Based on the target vehicle data, obtain the first feature to be identified, representing the probability distribution of the headway distance in the target lane during the target green light period; the second feature to be identified, representing the change in the number of vehicles released in the target lane during each statistical time period of the target green light period; and the third feature to be identified, representing the probability distribution of the vehicle speed in the target lane during the target green light period.

[0076] S103: Determine the discrete state of traffic flow of the target lane based on the first feature to be identified, determine the traffic demand state of the target lane based on the second feature to be identified, and determine the traffic operation state of the target lane based on the third feature to be identified.

[0077] S104: Determine the scenario type of the target intersection during the control period based on the discrete state of traffic flow, traffic demand, and traffic operation status of the target lanes in each direction.

[0078] Based on the above processing, for each target lane in each direction of the target intersection, a first, second, and third feature to be identified can be obtained based on the target vehicle passage data during the target green light period. Since the first feature to be identified represents the probability distribution of headway distance in the target lane during the target green light period, and this probability distribution reflects the proportion of vehicles in the traffic flow constrained to varying degrees by the vehicle in front, the discrete state of the traffic flow for that target lane can be determined based on the first feature to be identified. Since the second feature to be identified represents the change in the number of vehicles allowed to pass through the target lane during each statistical time period of the target green light period, and this change in number reflects the queue length of vehicles in the target lane, the traffic demand state for that target lane can be determined based on the second feature to be identified. Since the third feature to be identified represents the probability distribution of vehicle speed in the target lane during the target green light period, and this probability distribution reflects the proportion of vehicles with different speeds in the traffic flow, the traffic operation state for that target lane can be determined based on the third feature to be identified. Therefore, for any target lane, it can be described from three aspects: discrete state of traffic flow, traffic demand state, and traffic operation state. This allows us to determine the scenario type of the target intersection during the control period based on the discrete state of traffic flow, traffic demand, and traffic operation status of the target lanes in each direction. Thus, it enables us to obtain the scenario type of the target intersection during the control period based on vehicle passage data.

[0079] For step S101, the target intersection is the intersection that requires scene type identification. For example, in urban roads, an intersection on a main urban road can be selected as the target intersection.

[0080] Any intersection contains road segments in multiple directions; that is, the road segments included in a target intersection (which can be called target road segments) can be determined according to the direction of the road segments. For example, a common cross intersection contains road segments in four directions (such as road segments in the four directions of east, west, north, and south), and a T-shaped intersection contains road segments in three directions.

[0081] Each road segment may contain at least one lane. Accordingly, for each target road segment in each direction within a target intersection, a lane can be selected from the lanes contained in that target road segment as the target lane for that direction. The specific process for determining the target lane in a target road segment will be described in subsequent embodiments.

[0082] Understandably, due to the influence of human activities (e.g., more vehicles on the road during the day and fewer vehicles at night), the scene type of the target intersection is not fixed. Therefore, the scene type of the target intersection to be identified within a certain time period (which can be called the control time period) can be selected according to actual needs.

[0083] In one implementation, the day can be divided into multiple time periods. For example, the day can be divided into multiple time periods, with each hour representing a time period. Each of these time periods can then be considered a time period to be controlled.

[0084] Alternatively, technicians can preset a fixed time period as the control period based on actual needs. For example, according to human life patterns, the morning peak time period (such as 7:00-9:00) and / or the evening peak time period (such as 17:00-20:00) can be determined each day. Then, the morning peak time period and / or the evening peak time period can be determined as the control period.

[0085] Understandably, when determining the traffic signal control strategy for a specific time period on a given day, vehicle passage data from historical periods within that time period can be obtained, and steps S101-S104 can be executed to obtain the scenario type of the target intersection during the time period. For example, for a target intersection, if the time period to be controlled is the morning rush hour on a Tuesday of a certain week, vehicle passage data from the morning rush hour on the Tuesday of the previous week can be obtained, or vehicle passage data from the morning rush hour on Monday of the current week can be obtained, and steps S101-S104 can be executed. Then, based on the obtained scenario type, the traffic signal control strategy for the time period to be controlled on Tuesday of that week can be determined.

[0086] The specific method for determining the control strategy for traffic signals will be explained in subsequent embodiments.

[0087] In some embodiments, the electronic device can acquire signal timing data of the target intersection. The signal timing data includes the correspondence between intersection identifiers, road segment identifiers, lane identifiers, green light start time, and green light end time. Specifically, there is a one-to-one correspondence between intersection identifiers, a one-to-one correspondence between road segment identifiers within any intersection, and a one-to-one correspondence between lane identifiers within any road segment.

[0088] Furthermore, the electronic device can determine, based on the signage of the target intersection, the signage of the target road segment, and the signage of each lane within the target road segment, all green light cycles (i.e., the target green light cycles in this application) included in the signal timing data for each lane within the target road segment of the target intersection. A green light cycle represents the duration of the green light within a traffic signal cycle.

[0089] For each target lane in each direction of the target intersection, after obtaining all the green light cycles included in the time period to be controlled for that target lane, the time corresponding to one or more green light cycles can be determined as the target green light period.

[0090] The electronic equipment can also acquire vehicle passage data at the target intersection during the control period. This data includes: intersection markings, road segment markings, lane markings, lane types, and the corresponding passage times. Specifically, the passage time corresponding to a lane marking indicates the time it takes for each vehicle permitted to travel in that lane to reach a designated location within that lane (such as the lane's stop line).

[0091] Furthermore, the electronic equipment can determine the vehicle passage data of each lane in the target road segment of the target intersection during the time period to be controlled, based on the signs of the target intersection, the signs of the target road segment, and the signs of each lane in the target road segment, from the vehicle passage data.

[0092] In addition, vehicle data can also include: license plate number and travel time within a road segment. Specifically, the travel time for a given license plate number indicates the time the vehicle with that license plate entered the road segment and the time it exited the road segment.

[0093] Furthermore, based on the signal timing data of the target intersection and the vehicle passage data of the target intersection during the waiting period, the vehicle passage data of the target lane during the target green light period (i.e., the target vehicle passage data) can be obtained.

[0094] In some embodiments, during the control period, the target lane has the highest number of vehicles allowed to pass in each lane of any direction of the road segment at the target intersection.

[0095] In this embodiment of the application, for each lane in the target road segment, the number of vehicle passage times in the vehicle passage data of that lane during the control period can be used as the number of vehicles allowed to pass through that lane during the control period (which can be called the release number). Furthermore, the lane corresponding to the largest release number can be determined as the target lane of the target road segment, which also enables the determination of target lanes in different directions.

[0096] Based on the above processing, when determining the target lanes for each direction, the lane with the largest number of vehicles allowed to pass during the control period can be identified as the target lane for each lane in the target road segment. Therefore, the traffic status of the target lane can accurately reflect the traffic status of the road segment to which the target lane belongs. Subsequently, when determining the traffic signal control strategy based on the scenario type of the target intersection determined by the traffic status of each target lane, it can ensure that the needs of the road segments in each direction within the target intersection are met.

[0097] In some embodiments, for each lane in the target road segment, after obtaining the number of lanes allowed to pass, the ratio of the number of lanes allowed to pass to the lane's capacity can be calculated as the lane's saturation level, and the lane corresponding to the highest saturation level is determined as the target lane. The capacity of any lane is preset by technicians based on the actual conditions of that lane.

[0098] Based on the above processing, when determining the target lanes in each direction, the lane with the highest saturation can be identified as the target lane for each lane in the target road segment. Therefore, the traffic status of the target lane can accurately reflect the traffic status of the road segment to which the target lane belongs. Subsequently, when determining the traffic signal control strategy based on the scenario type of the target intersection determined by the traffic status of each target lane, it can ensure that the needs of the road segments in each direction within the target intersection are met.

[0099] For step S102, for each target lane, the target vehicle data of the target lane can be processed to obtain a first feature to be identified, representing the probability distribution of the headway of the vehicles in the target lane during the target green light period; a second feature to be identified, representing the change in the number of vehicles released in the target lane during each statistical time period of the target green light period; and a third feature to be identified, representing the probability distribution of the vehicle speed in the target lane during the target green light period.

[0100] In some embodiments, see Figure 2 , Figure 2This application provides a flowchart for obtaining a first feature to be identified based on target vehicle passage data. The step of obtaining a first feature to be identified, representing the probability distribution of headway distances in the target lane during a target green light period, based on the target vehicle passage data, includes:

[0101] S201: Based on the target vehicle passage data, determine the headway of each vehicle at a specified position in the target lane during the target green light period, and use it as the headway to be processed.

[0102] S202: Determine the preset headway interval to which each headway to be processed belongs.

[0103] Among them, each preset headway interval is adjacent and the length of each preset headway interval is the same.

[0104] S203: For each preset headway time interval, calculate the ratio of the headway time interval to be processed belonging to that preset headway time interval to all headway time intervals to be processed, and obtain the first feature to be identified.

[0105] Headway indicates the time difference between the front ends of adjacent vehicles passing a specified position (such as the stop line) in the same lane.

[0106] Regarding step S201, since the target vehicle passage data includes: the vehicle passage time of the target lane within the time period to be controlled (which can be called the first vehicle passage time), and the vehicle passage time represents: the time it takes for each vehicle released in the target lane to travel to a specified position (such as the stop line of the lane) in the target lane. Therefore, the vehicle passage time within the target green light period in the first vehicle passage time (which can be called the second vehicle passage time) can be determined, and the difference between every two adjacent second vehicle passage times can be calculated to obtain the headway of each vehicle at the specified position in the target lane during the target green light period (i.e., the headway of the vehicles to be processed).

[0107] For example, the headway between any two vehicles can be calculated using the following formula (1):

[0108] T i，i+1 =pass_time i+1 -pass_time i (1)

[0109] Among them, T i,i+1 The pass_time represents the headway between the i-th vehicle and the (i+1)-th vehicle. i+1 The time taken for the (i+1)th vehicle to pass through the specified location in the target lane is represented by `pass_time`. i This represents the time it takes for the i-th vehicle to pass through the specified location in the target lane.

[0110] In step S202, the preset headway time interval to which each headway to be processed belongs can be determined. These preset headway time intervals are adjacent and have the same length.

[0111] For example, the length of each preset headway interval is 3 seconds, and the preset headway intervals are [0,3), [3,6), [6,9), etc. Accordingly, when the headway interval to be processed is 5 seconds, it can be determined that the headway interval to be processed belongs to the preset headway interval [3,6); when the headway interval to be processed is 8 seconds, it can be determined that the headway interval to be processed belongs to the preset headway interval [6,9].

[0112] For step S203, for each preset headway interval, calculate the ratio of the headway to be processed belonging to that preset headway interval to all headway to be processed. Accordingly, based on the ratio corresponding to each preset headway interval, determine the probability distribution of the headway to be processed (i.e., the probability distribution of headway in the target lane during the target green light period).

[0113] For example, each ratio can be calculated according to the following formula (2):

[0114]

[0115] Where P(r) is the ratio of the headway time to be processed belonging to the r-th preset headway time interval to all headway time intervals to be processed, n r This represents the number of unprocessed headway times belonging to the r-th preset headway time interval, and N represents the total number of unprocessed headway times.

[0116] Therefore, the probability distribution X of the time distance between the front and rear of the vehicle to be processed can be expressed as:

[0117] X={r, P(r)}, r=1, 2, 3,...

[0118] Where r represents the r-th preset headway interval, and P(r) is the ratio of the headway to be processed belonging to the r-th preset headway interval to all headway to be processed.

[0119] In some embodiments, a target green light period includes multiple target green light cycles. See also Figure 3 , Figure 3 This application provides a flowchart for obtaining a second feature to be identified based on target vehicle passage data. The step of obtaining a second feature representing the change in the number of vehicles allowed to pass through the target lane during each statistical time period of the target green light period, based on the target vehicle passage data, includes:

[0120] S301: For each target green light cycle, based on the target vehicle passage data of that target green light cycle, determine the number of vehicles allowed to pass through the target lane in each statistical time period of that target green light cycle, and use this number as the number of vehicles allowed to pass through in each statistical time period of that target green light cycle.

[0121] S302: For any statistical time period, calculate the average number of vehicles released during that statistical time period for each target green light cycle, and obtain the second feature to be identified, which represents the change in the number of vehicles released in the target lane during each statistical time period of the target green light period.

[0122] In this embodiment of the application, each target green light cycle can be divided according to a preset duration to determine the statistical time periods included in each target green light cycle.

[0123] For example, the preset duration can be 5 seconds. Accordingly, for a target green light cycle, the first statistical time period included in the target green light cycle is 0 to 5 seconds in the target green light cycle, the second statistical time period included in the target green light cycle is 5 to 10 seconds in the target green light cycle, and so on, until all statistical time periods included in the target green light cycle are determined.

[0124] Accordingly, since the target vehicle passage data includes the vehicle passage time of the target lane within the time period to be controlled (i.e., the first vehicle passage time), for each statistical time period of each target green light cycle, the number of vehicle passage times within each statistical time period of the target green light cycle can be determined, that is, the number of vehicles released in the target lane within each statistical time period of the target green light cycle can be determined as the number of vehicles released in each statistical time period of the target green light cycle.

[0125] For any given statistical time period, calculate the average number of vehicles allowed during that statistical time period within each target green light cycle.

[0126] For example, for the first statistical time period mentioned above, the average number of vehicles released during the first statistical time period within each target green light cycle can be calculated (this can be called the average value corresponding to the first statistical time period). Similarly, for the second statistical time period, the average number of vehicles released during the second statistical time period within each target green light cycle can be calculated. This process continues until the average value corresponding to each statistical time period is obtained, thus revealing the change in the number of vehicles released in the target lane within each statistical time period of the target green light cycle (this can be called the periodic traffic flow rate). Furthermore, a second feature to be identified, representing the change in the number of vehicles released in the target lane within each statistical time period of the target green light cycle, can be obtained.

[0127] For example, the change in the number of vehicles allowed to pass through the target lane during each statistical time period of the target green light period can be represented as follows:

[0128] flow=[flow1,flow2,flow3,…”

[0129] Among them, flow i Let represent the mean value corresponding to the i-th statistical time period.

[0130] In some embodiments, see Figure 4 , Figure 4 This application provides a flowchart for obtaining a third feature to be identified based on target vehicle passage data. The step of obtaining a third feature representing the probability distribution of vehicle speeds in the target lane during a target green light period, based on target vehicle passage data, includes:

[0131] S401: Based on the target vehicle passing data, determine the speed of each vehicle in the target lane during the target green light period, and use it as the speed to be processed.

[0132] S402: Determine the preset driving speed range to which each driving speed to be processed belongs.

[0133] Among them, the preset driving speed ranges are adjacent and the length of each preset driving speed range is the same.

[0134] S403: For each preset driving speed range, calculate the ratio of the driving speed to be processed belonging to that preset driving speed range to all driving speeds to be processed, and obtain the third feature to be identified.

[0135] Regarding step S401, since the target vehicle data includes license plate number and road segment travel time, for each vehicle, the travel time in the target road segment is obtained based on the time the vehicle corresponding to that license plate number entered the road segment and the time the vehicle corresponding to that license plate number exited the road segment. The ratio of the length of the target road segment to the travel time is calculated as the vehicle's speed. Furthermore, the speed of each vehicle in the target lane during the target green light period can be determined as the speed to be processed.

[0136] In step S402, the preset driving speed range to which each driving speed to be processed belongs can be determined. These preset driving speed ranges are adjacent and have the same length.

[0137] For example, the length of each preset driving speed range is 5 m / s (meters per second), and the preset driving speed ranges are [0,5), [5,10), [10,15), [15,20), etc. Accordingly, when a driving speed to be processed is 19 m / s, it can be determined that the driving speed to be processed belongs to the preset driving speed range [15,20); when a driving speed to be processed is 32 m / s, it can be determined that the driving speed to be processed belongs to the preset driving speed range [30,35].

[0138] For step S403, for each preset driving speed range, the ratio of the driving speed to be processed belonging to that preset driving speed range to all driving speeds to be processed is calculated. Accordingly, the probability distribution of the driving speed to be processed (i.e., the probability distribution of vehicle speeds in the target lane during the target green light period) can be determined based on the ratio corresponding to each preset driving speed range.

[0139] For example, each ratio can be calculated according to the following formula (3):

[0140]

[0141] Where P(j) is the ratio of the driving speed to be processed belonging to the j-th preset driving speed range to all driving speeds to be processed, and n j This represents the number of unprocessed driving speeds belonging to the j-th preset driving speed range, and M represents the total number of unprocessed driving speeds.

[0142] Therefore, the probability distribution Y of the driving speed to be processed can be expressed as:

[0143] Y={j,P(j)},j=1,2,3,...

[0144] Where j represents the j-th preset driving speed range, and P(j) is the ratio corresponding to the j-th preset driving speed range.

[0145] Regarding step S103, the traffic status of the target lane can be reflected by the discrete state of traffic flow, the state of traffic demand, and the state of traffic operation of the target lane.

[0146] The discrete state of traffic flow is used to represent whether the latter vehicle is constrained by the former vehicle among two adjacent vehicles in the same lane. It can be understood that the larger the headway between two adjacent vehicles, the greater the distance between them, and the less constrained the latter vehicle is by the former. Conversely, the smaller the headway between two adjacent vehicles, the smaller the distance between them, and the greater the constrained the latter vehicle is by the former.

[0147] Traffic demand status indicates whether a lane can clear all queuing vehicles within a single green light period. Understandably, if the number of vehicles requiring passage is small within a green light period, the lanes can clear all vehicles within the first few statistical time intervals; therefore, there are instances of empty lanes within the green light cycle. Correspondingly, this is manifested as the number of vehicles cleared initially increasing, then stabilizing, and finally decreasing within each statistical time interval, following the order of the statistical time intervals. Conversely, if the number of vehicles requiring passage is large, the lanes cannot clear all vehicles within a single green light cycle; therefore, there are no instances of empty lanes within the green light cycle. Correspondingly, this is manifested as the number of vehicles cleared initially increasing, then stabilizing, until the end of the green light cycle, following the order of the statistical time intervals.

[0148] Traffic flow status is used to indicate the speed of vehicles traveling in a lane. It can be understood that the less external interference a vehicle experiences, the faster its speed. Conversely, the more external interference a vehicle experiences, the slower its speed. For example, vehicles on highways generally do not experience interference from pedestrians and other vehicles, and therefore travel at higher speeds; vehicles on city roads need to yield to pedestrians and oncoming traffic, and therefore travel at lower speeds.

[0149] The specific process of determining the discrete state of traffic flow of the target lane based on the first feature to be identified, determining the traffic demand state of the target lane based on the second feature to be identified, and determining the traffic operation state of the target lane based on the third feature to be identified will be described in subsequent embodiments.

[0150] Regarding step S104, since the target intersection contains target lanes in multiple directions, and the traffic conditions of each target lane may be different, after determining the discrete traffic flow state, traffic demand state, and traffic operation state of the target lanes in each direction of the target intersection according to steps S101-S103 above, the scenario type of the target intersection during the control period can be determined by combining the discrete traffic flow state, traffic demand state, and traffic operation state of the target lanes in each direction. The specific process of determining the scenario type of the target intersection during the control period will be described in subsequent embodiments.

[0151] See Figure 5 , Figure 5 This is a schematic diagram of a process for obtaining the first, second, and third features to be identified of a target road segment in each direction, as provided in an embodiment of this application.

[0152] S501: Start.

[0153] S502: Obtain vehicle passage data and signal timing data at the intersection.

[0154] That is, to obtain the signal timing data of the target intersection, as well as the vehicle passage data of the target intersection during the time period to be controlled.

[0155] S503: Traverse the critical lanes in all directions.

[0156] That is, for any target road segment in any direction at the target intersection, determine the target lane in that target road segment.

[0157] S504: Obtain vehicle passage data during each green light period.

[0158] That is, acquiring the target vehicle passage data for the target lane during the target green light period within the control time period. The target green light period includes at least multiple target green light cycles.

[0159] S505: Extract the first feature to be identified, the second feature to be identified, and the third feature to be identified.

[0160] That is, the target vehicle data of the target lane is processed to obtain a first feature to be identified, representing the probability distribution of the headway of vehicles in the target lane during the target green light period; a second feature to be identified, representing the change in the number of vehicles released in the target lane during each statistical time period of the target green light period; and a third feature to be identified, representing the probability distribution of the vehicle speed in the target lane during the target green light period.

[0161] S506: Determine whether all directions have been traversed. If yes, proceed to step S507; otherwise, return to step S503.

[0162] S507: End.

[0163] In some embodiments, see Figure 6 , Figure 6 This is a flowchart illustrating how to determine the discrete state of traffic flow for a target lane based on a first feature to be identified, as provided in an embodiment of this application. Figure 6 middle:

[0164] S601: For each preset traffic flow discrete state, obtain the first sample vehicle passage data of the first sample lane during the first sample green light period when the first sample lane is in the preset traffic flow discrete state.

[0165] S602: Based on the first sample vehicle passage data, obtain the first sample feature representing the probability distribution of the headway distance in the first sample lane during the first sample green light period.

[0166] S603: For each preset traffic flow discrete state, the similarity between the first sample feature and the first feature to be identified in the preset traffic flow discrete state is calculated using the first similarity algorithm.

[0167] The first similarity algorithm includes at least one of the following: bulldozing distance algorithm and Mahalanobis distance algorithm.

[0168] S604: The preset traffic flow discrete state with the highest similarity is determined as the traffic flow discrete state of the target lane.

[0169] In this embodiment, a technician can determine, based on practical experience, that the first sample lane is in a certain preset traffic flow discrete state within the first sample time period. Then, vehicle passage data (i.e., first sample vehicle passage data) of the first sample lane during the first sample green light period can be obtained as the first sample feature of the preset traffic flow discrete state. The process of determining the first sample green light period within the first sample time period can refer to the process of determining the target green light period within the controllable time period described above, and will not be repeated here. The length of the first sample time period is the same as the length of the controllable time period.

[0170] Furthermore, based on the first sample vehicle passage data, a first sample feature representing the probability distribution of the headway distance between vehicles in the first sample lane during the first sample green light period can be obtained. The specific process of obtaining the first sample feature will be described in subsequent embodiments.

[0171] For each preset discrete traffic flow state, a first similarity algorithm can be used to obtain the similarity between a first sample feature and a first feature to be identified for that preset discrete traffic flow state. The first similarity algorithm includes at least one of the following: bulldozer distance algorithm, Mahalanobis distance algorithm, and other algorithms used to calculate the similarity between two probability distributions.

[0172] When the first similarity algorithm includes only one of the algorithms mentioned above for calculating the similarity between two probability distributions, the similarity between the first sample feature and the first feature to be identified in the preset traffic flow discrete state can be calculated using the first similarity algorithm.

[0173] When the first similarity algorithm includes multiple algorithms for calculating the similarity between two probability distributions, for each first similarity algorithm, a first similarity between the first sample feature of the preset traffic flow discrete state and the first feature to be identified can be calculated using that first similarity algorithm. Furthermore, the similarity between the first sample feature of the preset traffic flow discrete state and the first feature to be identified can be obtained by combining the calculated multiple first similarities. For example, the weighted sum of the calculated multiple first similarities can be used to obtain the similarity between the first sample feature of the preset traffic flow discrete state and the first feature to be identified.

[0174] In one implementation, if the preset traffic flow discrete state has multiple first sample features, the similarity between each first sample feature of the preset traffic flow discrete state and the first feature to be identified can be calculated. Based on the average of these multiple similarities, the similarity between the first feature to be identified and the first sample features of the preset traffic flow discrete state can be obtained. This further improves the accuracy of determining the traffic flow discrete state of the target lane.

[0175] The greater the similarity between the first feature to be identified and the first sample feature of the preset traffic flow discrete state, the closer the traffic flow discrete state of the target lane is to the preset traffic flow discrete state. Therefore, the preset traffic flow discrete state with the highest similarity can be determined as the traffic flow discrete state of the target lane.

[0176] Based on the above processing, the traffic flow discrete state of the target lane can be determined by calculating the similarity between the first feature to be identified and the first sample feature of each preset traffic flow discrete state, thus ensuring the accuracy of the determined traffic flow discrete state of the target lane.

[0177] In some embodiments, the first feature to be identified can be input into a pre-trained neural network model to obtain the discrete traffic flow state of the target lane. This neural network model is trained based on the first sample feature and its corresponding sample label, where the sample label represents a preset discrete traffic flow state corresponding to the first sample feature. Thus, processing the first feature to be identified using a neural network model to determine the discrete traffic flow state of the target lane can improve the efficiency of intersection scene recognition.

[0178] In addition, in related technologies, when determining the discrete state of traffic flow in a lane, the discrete state of traffic flow in the lane can be determined by judging whether the headway of vehicles in the lane is greater than a certain preset threshold. However, since the actual conditions of different roads are different, the preset threshold may also be different, making it difficult to set such a threshold.

[0179] In some embodiments, a traffic flow discrete state sample library can be pre-established to store the first sample features of each of the aforementioned preset traffic flow discrete states. That is, steps S601-S604 can be executed based on the first sample features of each of the preset traffic flow discrete states recorded in the traffic flow discrete state sample library.

[0180] Accordingly, after determining the discrete state of traffic flow in the target lane, the first feature to be identified for that target lane can be recorded in the discrete state traffic flow sample library, and its corresponding discrete state of traffic flow can be marked for use in the subsequent scene type identification process of other intersections. In this way, the discrete state traffic flow sample library can be continuously updated, improving the accuracy of determining the discrete state of traffic flow in lanes.

[0181] It is understandable that, compared with the above-mentioned method of determining the discrete state of traffic flow of target road segments based on neural network models, the method of determining the discrete state of traffic flow of lanes provided in this application embodiment does not require pre-labeling a large amount of sample data, which can reduce the cost of manual labeling.

[0182] In addition, compared with the method of determining the discrete state of traffic flow of lanes by threshold, the method of determining the discrete state of traffic flow of lanes provided in this application embodiment can combine macro data and micro data to jointly determine the discrete state of traffic flow of lanes. In this way, the accuracy of the determined discrete state of traffic flow of lanes can be improved, and thus the accuracy of the scene type of the target intersection can be improved.

[0183] In some embodiments, step S602 includes:

[0184] Step 1: Based on the first sample vehicle passage data, determine the headway of each vehicle at a specified position in the first sample lane during the first sample green light period, and use it as the sample headway.

[0185] Step 2: Determine the preset headway interval to which each sample headway belongs.

[0186] Among them, each preset headway interval is adjacent and the length of each preset headway interval is the same.

[0187] Step 3: For each preset headway interval, calculate the ratio of the sample headway of the vehicle belonging to that preset headway interval to the total sample headway, and obtain the first sample feature.

[0188] In this embodiment of the application, the process of obtaining the first sample feature based on the first sample vehicle data is similar to the process of obtaining the first feature to be identified based on the target vehicle data in steps S201-S203 above, and will not be described in detail here.

[0189] In this way, the method of determining the first sample feature of the preset traffic flow discrete state is consistent with the method of determining the first feature to be identified. When calculating the similarity between the first sample feature of the preset traffic flow discrete state and the first feature to be identified in the subsequent calculation, the similarity between the target road segment and the first sample road segment in each preset traffic flow discrete state can be accurately reflected, thus ensuring the accuracy of identification.

[0190] In some embodiments, see Figure 7 , Figure 7 This is a flowchart illustrating how to determine the traffic demand status of a target lane based on a second feature to be identified, as provided in an embodiment of this application. Figure 7 middle:

[0191] S701: For each preset traffic demand state, obtain the second sample vehicle passage data of the second sample lane during the second sample green light period when the second sample lane is in the preset traffic demand state.

[0192] S702: Based on the second sample vehicle passage data, obtain the second sample feature representing the change in the number of vehicles allowed to pass through the second sample lane during each statistical time period of the second sample green light period.

[0193] The steps for determining the traffic demand state of the target lane based on the second feature to be identified include:

[0194] S703: For each preset traffic demand state, calculate the similarity between the second sample feature of the preset traffic demand state and the second feature to be identified.

[0195] S704: The preset traffic demand state with the highest similarity is determined as the traffic demand state of the target lane.

[0196] In this embodiment, a technician can determine, based on practical experience, that the second sample lane is in a certain preset traffic demand state within the second sample time period. Then, vehicle passage data (i.e., second sample vehicle passage data) of the second sample lane during the second sample green light period can be obtained as the second sample feature of the preset traffic demand state. The process of determining the second sample green light period within the second sample time period can refer to the process of determining the target green light period within the control period described above, and will not be repeated here. The length of the second sample time period is the same as the length of the control period.

[0197] Furthermore, based on the second sample vehicle passage data, a second sample feature can be obtained, representing the change in the number of vehicles allowed to pass through the second sample lanes during each statistical time period of the second sample green light period. The specific process of obtaining the second sample feature will be described in subsequent embodiments.

[0198] For each preset traffic demand state, the similarity between the second sample feature and the second feature to be identified can be calculated based on a preset algorithm used to calculate the similarity between two sequences. The preset algorithm can be a Dynamic Time Warping (DTW) algorithm.

[0199] In one implementation, if the preset traffic demand state has multiple second sample features, the similarity between each second sample feature of the preset traffic demand state and the second feature to be identified can be calculated. Then, the mean of these multiple similarities can be calculated, and the traffic demand state of the target lane can be determined based on this mean. This further improves the accuracy of the determined discrete traffic flow state of the target lane.

[0200] The greater the similarity between the second feature to be identified and the second sample feature of the preset traffic demand state, the closer the traffic demand state of the target lane is to the preset traffic demand state. Therefore, the preset traffic demand state with the highest similarity can be determined as the traffic demand state of the target lane.

[0201] Based on the above processing, the traffic demand status of the target lane can be determined by calculating the similarity between the second feature to be identified and the second sample features of each preset traffic demand status, thus ensuring the accuracy of the determined traffic demand status of the target lane.

[0202] In some embodiments, the second feature to be identified can be input into a pre-trained neural network model to obtain the traffic demand state of the target lane. This neural network model is trained based on the second sample feature and its corresponding sample label, where the sample label represents a preset traffic demand state corresponding to the second sample feature. Thus, processing the second feature to be identified using a neural network model to determine the traffic demand state of the target lane can improve the efficiency of intersection scene recognition.

[0203] In addition, in related technologies, when determining the traffic demand status of a lane, the traffic demand status of the lane can be determined by judging whether the queue length of vehicles in the lane is greater than a certain preset threshold. However, since the actual conditions of different roads are different, the preset threshold may also be different, making it difficult to set such a threshold.

[0204] In some embodiments, a traffic demand state sample library can be pre-established to store the second sample features of each of the aforementioned preset traffic demand states. That is, steps S701-S704 can be performed based on the second sample features of each of the preset traffic demand states recorded in the traffic demand state sample library.

[0205] Accordingly, after determining the traffic demand state of the target lane, the second feature to be identified for that target lane can be recorded in the traffic demand state sample database, and its corresponding traffic demand state can be marked for use in the subsequent scene type identification process of other intersections. In this way, the traffic demand state sample database can be continuously updated, improving the accuracy of determining the traffic demand state of lanes.

[0206] It is understandable that, compared to the above-mentioned method of determining the traffic demand status of the target lane based on a neural network model, the method of determining the traffic demand status of the lane provided in this application embodiment does not require pre-labeling a large amount of sample data, thus reducing the cost of manual labeling.

[0207] In addition, compared with the method of determining the traffic demand status of a lane by using a threshold, the method of determining the traffic demand status of a lane provided in this application embodiment can combine macro data and micro data to jointly determine the traffic demand status of a lane. In this way, the accuracy of the determined traffic demand status of the lane can be improved, and thus the accuracy of the scene type of the target intersection can be improved.

[0208] In some embodiments, the second sample green light period includes multiple sample green light cycles. Step S702 above includes:

[0209] Step (1): For each sample green light cycle, based on the second sample vehicle passage data of that sample green light cycle, determine the number of vehicles released in the second sample lane in each statistical time period of that sample green light cycle, and use this as the number of vehicles released in each statistical time period of that sample green light cycle.

[0210] Step (2): For any statistical time period, calculate the mean number of vehicles released during the green light cycle of each sample within that statistical time period, and obtain the second sample feature representing the change in the number of vehicles released in the second sample lane during each statistical time period of the second sample green light period.

[0211] In this embodiment of the application, the process of obtaining the second sample feature based on the second sample vehicle data is similar to the process of obtaining the second feature to be identified based on the target vehicle data in steps S301-S302 above, and will not be described in detail here.

[0212] In this way, the method of determining the second sample feature of the preset traffic demand state is consistent with the method of determining the second feature to be identified. When calculating the similarity between the second sample feature of the preset traffic demand state and the second feature to be identified in the subsequent calculation, the similarity between the target road segment and the second sample lane in each preset traffic demand state can be accurately reflected, thus ensuring the accuracy of identification.

[0213] In some embodiments, see Figure 8 , Figure 8 This is a flowchart illustrating how to determine the traffic operation status of a target lane based on a third feature to be identified, as provided in an embodiment of this application. Figure 8 middle:

[0214] S801: For each preset traffic operation state, obtain the third sample vehicle passage data of the third sample lane during the third sample green light period when the third sample lane is in the preset traffic operation state.

[0215] S802: Based on the third sample vehicle passage data, obtain the third sample feature representing the probability distribution of vehicle speeds in the third sample lane during the third sample green light period.

[0216] S803: For each preset traffic operation state, the similarity between the third sample feature of the preset traffic operation state and the third feature to be identified is calculated using the second similarity algorithm.

[0217] The second similarity algorithm includes at least one of the following: bulldozing distance algorithm and Mahalanobis distance algorithm.

[0218] S804: The preset traffic operation state with the highest similarity is determined as the traffic operation state of the target lane.

[0219] In this embodiment, a technician can determine, based on practical experience, that the third sample lane is in a certain preset traffic operation state within the third sample time period. Furthermore, vehicle passage data of the third sample lane during the third sample green light period (i.e., third sample vehicle passage data) can be obtained as the third sample feature of the preset traffic operation state. The process of determining the third sample green light period within the third sample time period can refer to the process of determining the target green light period within the control period described above, and will not be repeated here. The length of the third sample time period is the same as the length of the control period.

[0220] For each preset traffic operation state, a second similarity algorithm can be used to obtain the similarity between the third sample feature of that preset traffic operation state and the third feature to be identified. The second similarity algorithm includes at least one of the following: bulldozer distance algorithm, Mahalanobis distance algorithm, and other algorithms used to calculate the similarity between two probability distributions.

[0221] When the second similarity algorithm includes only one of the algorithms mentioned above for calculating the similarity between two probability distributions, the similarity between the third sample feature of the preset traffic operation state and the third feature to be identified can be calculated using the second similarity algorithm.

[0222] When the second similarity algorithm includes multiple algorithms for calculating the similarity between two probability distributions, for each second similarity algorithm, a second similarity between the third sample feature of the preset traffic operation state and the third feature to be identified can be calculated. Furthermore, the similarity between the third sample feature of the preset traffic operation state and the third feature to be identified can be obtained by combining the calculated multiple second similarities. For example, the similarity between the third sample feature of the preset traffic operation state and the third feature to be identified can be obtained by calculating a weighted sum of the calculated multiple second similarities.

[0223] In one implementation, if the preset traffic operation state has multiple third sample features, the similarity between each third sample feature of the preset traffic operation state and the third feature to be identified can be calculated. Based on the average of the multiple similarities, the similarity between the third feature to be identified and the third sample features of the preset traffic operation state can be obtained. In this way, the accuracy of determining the traffic operation state of the target lane can be further improved.

[0224] The greater the similarity between the third feature to be identified and the third sample feature of the preset traffic operation state, the closer the traffic operation state of the target lane is to the preset traffic operation state. Therefore, the preset traffic operation state with the highest similarity can be determined as the traffic operation state of the target lane.

[0225] Based on the above processing, the traffic operation status of the target lane can be determined by calculating the similarity between the third feature to be identified and the third sample feature of each preset traffic operation status, thus ensuring the accuracy of the determined traffic operation status of the target lane.

[0226] In some embodiments, a third feature to be identified can be input into a pre-trained neural network model to obtain the traffic operation status of the target lane. This neural network model is trained based on the third sample feature and its corresponding sample label, where the sample label represents a preset traffic operation status corresponding to the third sample feature. Thus, processing the third feature to be identified using a neural network model to determine the traffic operation status of the target lane can improve the efficiency of intersection scene recognition.

[0227] In addition, in related technologies, the traffic status of a lane can be determined by judging whether the speed of vehicles in the lane exceeds a certain preset threshold. However, since the actual conditions of different roads are different, the preset threshold may also be different, making it difficult to set such a threshold.

[0228] In some embodiments, a traffic operation status sample library can be pre-established to store the third sample features of each of the aforementioned preset traffic operation statuses. That is, steps S801-S804 can be executed based on the third sample features of each preset traffic operation status recorded in the traffic operation status sample library.

[0229] Accordingly, after determining the traffic status of the target lane, the third feature to be identified for that target lane can be recorded in the traffic status sample database, and its corresponding traffic status can be marked for use in the subsequent scene type identification process of other intersections. In this way, the traffic status sample database can be continuously updated, improving the accuracy of determining the traffic status of lanes.

[0230] It is understandable that, compared to the above-mentioned method of determining the traffic status of a target lane based on a neural network model, the method of determining the traffic status of a lane provided in this application embodiment does not require pre-labeling a large amount of sample data, thus reducing the cost of manual labeling.

[0231] In addition, compared with the method of determining the traffic operation status of a lane by using a threshold, the method of determining the traffic operation status of a lane provided in this application embodiment can combine macro data and micro data to jointly determine the traffic operation status of the lane. In this way, the accuracy of the determined traffic operation status of the lane can be improved, and thus the accuracy of the scene type of the target intersection can be improved.

[0232] In some embodiments, step S802 includes:

[0233] Step 1: Based on the third sample vehicle passage data, determine the speed of each vehicle in the third sample lane during the third sample green light period, and use it as the sample speed.

[0234] Step 2: Determine the preset driving speed range to which each sample driving speed belongs.

[0235] Among them, the preset driving speed ranges are adjacent and the length of each preset driving speed range is the same.

[0236] Step 3: For each preset driving speed range, calculate the ratio of the sample driving speed belonging to that preset driving speed range to the total sample driving speed, and obtain the third sample feature.

[0237] In this embodiment of the application, the process of obtaining the third sample feature based on the third sample vehicle data is similar to the process of obtaining the third feature to be identified based on the target vehicle data in steps S401-S403 above, and will not be described in detail here.

[0238] In this way, the method of determining the third sample feature of the preset traffic operation state is consistent with the method of determining the third feature to be identified. When calculating the similarity between the third sample feature of the preset traffic operation state and the third feature to be identified in the subsequent calculation, the similarity between the target road segment and the third sample road segment in each preset traffic operation state can be accurately reflected, thus ensuring the accuracy of identification.

[0239] In some embodiments, the traffic flow discrete state is: discrete, weak car-following, or strong car-following; the traffic demand state is: low traffic demand or high traffic demand; and the traffic operation state is: slow-moving or smooth-flowing. For example... Figure 9 As shown, Figure 9 This is a second flowchart of the intersection scene recognition method provided in the embodiments of this application. Figure 1 Based on this, step S104 includes:

[0240] S1041: If at least one target lane in each direction of the target intersection meets the first screening condition, then the scenario type of the target intersection during the control period is determined to be queuing balance type.

[0241] The first screening condition indicates that the traffic flow discrete state is strong car-following, the traffic demand state is high traffic demand, and the traffic operation state is slow.

[0242] S1042: If there are no target lanes in each direction of the target intersection that meet the first screening condition, and there is at least one target lane that meets the second screening condition, then the scenario type of the target intersection during the control period is determined to be the capacity type.

[0243] The second screening condition indicates that the traffic flow discrete state is strong car-following or weak car-following, the traffic demand state is high traffic demand, and the traffic operation state is smooth.

[0244] S1043: If there are no target lanes in each direction of the target intersection that meet the first screening condition and no target lanes that meet the second screening condition, then the scenario type of the target intersection during the control period is determined to be efficiency improvement type.

[0245] The discrete states of each traffic flow are predetermined based on data patterns and a traffic knowledge model, which is obtained by technicians based on experience. As shown in Figure 10(a), which is a schematic diagram of lanes with different discrete states of traffic flow provided in the embodiments of this application, a gray rectangle in the figure represents a vehicle on a lane.

[0246] For any lane, if the traffic flow of that lane is discrete within a certain time period, it means that most vehicles in the traffic flow are not constrained by the vehicle in front during that time period, and the headway distribution of vehicles leaving the stop line is discrete. As shown in lane 1 of Figure 10(a).

[0247] For any lane, if the traffic flow discrete state of that lane is a weak following state within a certain time period, it means that during that time period, part of the traffic flow is constrained by the vehicle in front, and part is not constrained by the vehicle in front, maintaining a certain headway. As shown in lane 2 of Figure 10(a).

[0248] For any lane, if the traffic flow discrete state of that lane is a strong following relaxation state within a certain time period, it means that during that time period, most vehicles in the traffic flow are constrained by the vehicle in front, and the headway of the vehicles leaving the stop line is relatively concentrated and small. As shown in lane 3 of Figure 10(a).

[0249] Each traffic demand state is predetermined based on data patterns and a traffic knowledge model. As shown in Figures 10(b1) and 10(b2), Figure 10(b1) is a schematic diagram illustrating the change in the number of vehicles released in lanes with low traffic demand according to an embodiment of this application. In Figure 10(b1), the X-axis represents duration, T represents one green light cycle, and the Y-axis represents the number of vehicles released in each statistical time period. Since the traffic demand state of the lane is low, meaning the number of vehicles required to be released in the lane is relatively small, vehicles can be completely released in the first few statistical time periods. Therefore, there are instances of vehicles being released without traffic during the green light cycle. Correspondingly, this manifests as follows: according to the order of the statistical time periods, the number of vehicles released first increases, then tends to stabilize, and finally decreases within each statistical time period.

[0250] Figure 10(b2) is a schematic diagram illustrating the change in the number of vehicles released in a high-traffic-demand lane according to an embodiment of this application. Since the traffic demand state of the lane is high, meaning that a large number of vehicles need to be released, the lane cannot be completely cleared within a single green light cycle. Therefore, there is no empty lane during a green light cycle. Correspondingly, this manifests as follows: according to the chronological order of the statistical time periods, the number of vehicles released first increases, then tends to stabilize, until the end of the green light cycle.

[0251] Each traffic operation state is predetermined based on data patterns and a traffic knowledge model. As shown in Figure 10(c), which is a schematic diagram of roads with different traffic operation states provided in the embodiments of this application, a gray rectangle in the figure represents a vehicle in a lane.

[0252] For any given lane, if the traffic flow is slow within a given time period, it indicates that most vehicles in the traffic flow are traveling at relatively low speeds, generally indicating a congested lane or significant interference within the lane, such as severe interference between pedestrians and motor vehicles. See lane 1 in Figure 10(c).

[0253] For any given lane, if the traffic flow is smooth within a given time period, it means that most vehicles in the traffic flow are traveling at a higher speed, generally occurring in lanes with less traffic or lanes with more traffic but less road interference. See lane 2 in Figure 10(c).

[0254] In this embodiment, when at least one target lane in each direction meets the first screening condition, the scenario type of the target intersection during the control period is preferentially determined to be queuing balancing. If the scenario type of the target intersection during the control period is not queuing balancing (i.e., no target lane meets the first screening condition), and at least one target lane meets the second screening condition, the scenario type of the target intersection during the control period is determined to be capacity-based. Furthermore, if the scenario type of the target intersection during the control period is neither queuing balancing nor capacity-based (i.e., neither the first nor the second screening condition is met), the scenario type of the target intersection during the control period is determined to be efficiency improvement.

[0255] Based on the above processing, when determining the scenario type of the target intersection, for each direction of the target lane, if there is a target lane that meets the first screening condition, the scenario type of the target intersection during the control period can be determined as queuing equilibrium type. When determining the traffic signal control strategy based on the scenario type of the target intersection determined by the traffic status of each target lane, it can be ensured that the needs of each direction of the road segment in the target intersection are met.

[0256] In some embodiments, the intersection scene recognition method provided in this application further includes:

[0257] If the scenario type of the target intersection during the control period is queuing balance, then the first type of traffic signal control strategy will be adopted at the target intersection.

[0258] Among them, the control strategy of the first type of traffic signal is used to shorten the queue length of vehicles at the target intersection.

[0259] If the scenario type of the target intersection during the control period is capacity-based, then the control strategy of the second type of traffic signal will be adopted at the target intersection.

[0260] Among them, the control strategy of the second type of traffic signal is used to improve the vehicle throughput capacity of the target intersection.

[0261] If the scenario type of the target intersection during the control period is efficiency improvement, then the control strategy of the third type of traffic signal will be adopted at the target intersection.

[0262] Among them, the control strategy for the third type of traffic signal is used to reduce the delay time of vehicles at the target intersection.

[0263] In this embodiment, if the scenario type of the target intersection during the control period is queuing equilibrium, that is, at least one target lane in each target road segment of the target intersection has a traffic flow discrete state of strong following, a traffic demand state of high traffic demand, and a traffic operation state of slow movement, in order to avoid excessive queuing of vehicles in at least one target road segment, a first type of traffic signal control strategy can be determined to be adopted at the target intersection. The first type of traffic signal control strategy is used to shorten the queue length of vehicles at the target intersection.

[0264] If the scenario type of the target intersection during the control period is capacity-based, meaning there is at least one target lane with high traffic demand and smooth traffic flow, and there are no target road segments with strong car-following, high traffic demand, and slow traffic flow, then to avoid congestion on any of the target road segments at the target intersection, a second type of traffic signal control strategy can be adopted at the target intersection. This second type of traffic signal control strategy is used to improve the vehicle capacity of the target intersection.

[0265] If the scenario type of the target intersection during the control period is efficiency improvement, it indicates that the traffic conditions of each target lane at the target intersection are relatively good. In this case, in order to reduce the number of vehicle stops or reduce vehicle delay time, it can be determined to adopt the third type of traffic signal control strategy at the target intersection. The third type of traffic signal control strategy is used to reduce vehicle delay time at the target intersection.

[0266] In this way, the appropriate traffic signal control strategy can be adopted at the target intersection based on the scenario type of the target intersection during the control period, and the optimal control strategy can be recommended for the target intersection.

[0267] Figure 11 This is a schematic diagram of a control strategy determination device provided in an embodiment of this application. Figure 11 In this process, the data of the intersection to be controlled (i.e., the target intersection) can be input into the feature extraction and mining module. The data of the target intersection includes: vehicle passage data and signal timing data of the target intersection. The feature extraction and mining module can obtain the headway probability distribution (i.e., the first feature to be identified, representing the probability distribution of headway in the target lane during the target green light period), the periodic traffic flow rate (i.e., the second feature to be identified, representing the change in the number of vehicles released in the target lane during each statistical time period of the target green light period), and the vehicle speed probability distribution (i.e., the third feature to be identified, representing the probability distribution of vehicle speed in the target lane during the target green light period) according to the above step S102.

[0268] The feature extraction and mining module can input the obtained first, second, and third features to be identified into the scene recognition module. The scene recognition module can obtain the first sample features of each preset traffic flow discrete state, the second sample features of each preset traffic demand state, and the third sample features of each preset traffic operation state stored in the scene library construction module. Among them, the preset traffic flow discrete states include: strong car-following, weak car-following, and discrete; the preset traffic demand states include: high traffic demand and low traffic demand; and the preset traffic operation states include: smooth flow and slow flow.

[0269] Accordingly, the scene recognition module can measure the similarity between the first feature to be identified and the first sample features of each preset traffic flow discrete state, obtaining the similarity score between the first feature to be identified and the first sample features of each preset traffic flow discrete state. Based on a preset classification method, the obtained similarity score is classified to determine the traffic flow discrete state corresponding to the first feature to be identified. The preset classification method can be: nearest neighbor mean classification method, or other distance-based classification methods (such as fixed-center clustering method). Similarly, the second feature to be identified is measured for similarity with the second sample features of each preset traffic demand state, obtaining the similarity score between the second feature to be identified and the second sample features of each preset traffic demand state. Based on a preset classification method, the obtained similarity score is classified to determine the traffic demand state corresponding to the second feature to be identified. The third feature to be identified is measured for similarity with the third sample features of each preset traffic operation state, obtaining the similarity score between the third feature to be identified and the third sample features of each preset traffic operation state. Based on a preset classification method, the obtained similarity score is classified to determine the traffic operation state corresponding to the third feature to be identified. In this way, the traffic status of the target lanes in different directions at the target intersection can be obtained, and thus the scene recognition result can be obtained (that is, the scene type of the target intersection during the time period to be controlled is determined according to step S104).

[0270] In addition, after determining the scene type of the target intersection within the control period, the scene library (i.e., the sample library in the above embodiments) can be updated using the first feature to be identified, the second feature to be identified, and the third feature to be identified.

[0271] The intersection scenario output module outputs the scenario type of a given target intersection during the control period. Subsequently, the required traffic signal control strategy can be determined based on the scenario type of the target intersection during the control period.

[0272] In the technical solution of this application, the acquisition, storage, use, processing, transmission, provision and disclosure of vehicle data are all carried out with the user's authorization.

[0273] Based on the same inventive concept, embodiments of this application provide an intersection scene recognition device. See also Figure 12 , Figure 12 This application provides a structural diagram of an intersection scene recognition device, which includes:

[0274] The target vehicle passage data acquisition module 1201 is used to acquire the target vehicle passage data of the target lane during the target green light period in each direction of the target intersection for the target lane in the waiting control time period.

[0275] The feature acquisition module 1202 is used to obtain, based on the target vehicle passing data, a first feature representing the probability distribution of the headway distance in the target lane during the target green light period, a second feature representing the change in the number of vehicles allowed to pass in the target lane during each statistical time period of the target green light period, and a third feature representing the probability distribution of the vehicle speed in the target lane during the target green light period.

[0276] The state determination module 1203 is used to determine the discrete state of traffic flow of the target lane based on the first feature to be identified, the traffic demand state of the target lane based on the second feature to be identified, and the traffic operation state of the target lane based on the third feature to be identified.

[0277] The scenario determination module 1204 is used to determine the scenario type of the target intersection within the time period to be controlled based on the discrete state of traffic flow, traffic demand state, and traffic operation state of the target lanes in each direction.

[0278] In some embodiments, during the control period, the target lane has the highest number of vehicles allowed to pass in each lane of any direction of the road segment at the target intersection.

[0279] In some embodiments, the apparatus further includes:

[0280] The first sample vehicle passage data acquisition module is used to acquire, for each preset traffic flow discrete state, the first sample vehicle passage data of the first sample lane during the first sample green light period when the first sample lane is in the preset traffic flow discrete state, before determining the traffic flow discrete state of the target lane based on the first feature to be identified.

[0281] The first sample feature acquisition module is used to obtain a first sample feature representing the probability distribution of the headway distance in the first sample lane during the first sample green light period based on the first sample vehicle passage data.

[0282] The state determination module 1203 is specifically used for: for each preset traffic flow discrete state, calculating the similarity between the first sample feature of the preset traffic flow discrete state and the first feature to be identified by using a first similarity algorithm; wherein, the first similarity algorithm includes at least one of the following: bulldozing distance algorithm and Mahalanobis distance algorithm; and determining the preset traffic flow discrete state with the highest similarity as the traffic flow discrete state of the target lane.

[0283] In some embodiments, the sample feature acquisition module is specifically configured to: determine, based on the first sample vehicle passage data, the headway of each vehicle at a specified position in the first sample lane during the first sample green light period, as the sample headway; determine the preset headway interval to which each sample headway belongs; wherein, each preset headway interval is adjacent and has the same length; and for each preset headway interval, calculate the ratio of the sample headway belonging to that preset headway interval to all sample headway intervals, to obtain the first sample feature;

[0284] The feature acquisition module 1202 is specifically used for: determining the headway of each vehicle at a specified position in the target lane during the target green light period based on the target vehicle passage data, as the headway to be processed; determining the preset headway interval to which each headway to be processed belongs; and for each preset headway interval, calculating the ratio of the headway to be processed belonging to that preset headway interval to all headways to be processed, thereby obtaining the first feature to be identified.

[0285] In some embodiments, the apparatus further includes:

[0286] The second sample vehicle passage data acquisition module is used to acquire, for each preset traffic demand state, second sample vehicle passage data of the second sample lane during the second sample green light period when the second sample lane is in the preset traffic demand state, before determining the traffic demand state of the target lane based on the second feature to be identified.

[0287] The second sample feature acquisition module is used to obtain second sample features based on the second sample vehicle passage data, representing the change in the number of vehicles allowed to pass through the second sample lane in each statistical time period during the second sample green light period.

[0288] The state determination module 1203 is specifically used for: calculating the similarity between the second sample feature of the preset traffic demand state and the second feature to be identified for each preset traffic demand state; and determining the preset traffic demand state with the highest similarity as the traffic demand state of the target lane.

[0289] In some embodiments, the second sample green light period includes multiple sample green light cycles; the target green light period includes multiple target green light cycles; the second sample feature acquisition module is specifically used for: for each sample green light cycle, based on the second sample vehicle passage data of that sample green light cycle, determining the number of vehicles released in the second sample lane within each statistical time period of that sample green light cycle, as the number of vehicles released in each statistical time period of that sample green light cycle; for any statistical time period, calculating the average number of vehicles released in that statistical time period of each sample green light cycle, to obtain the average number of vehicles released in each statistical time period of the second sample green light period. The second sample feature describes the change in the number of vehicles released in the second sample lane; the feature acquisition module 1202 is specifically used for: for each target green light cycle, based on the target vehicle passage data of the target green light cycle, determining the number of vehicles released in the target lane in each statistical time period of the target green light cycle, as the number of vehicles released in each statistical time period of the target green light cycle; for any statistical time period, calculating the average number of vehicles released in that statistical time period of each target green light cycle, to obtain the second feature representing the change in the number of vehicles released in the target lane in each statistical time period of the target green light period.

[0290] In some embodiments, the apparatus further includes: a third sample vehicle passage data acquisition module, configured to acquire, for each preset traffic operation state, third sample vehicle passage data of the third sample lane during a third sample green light when the third sample lane is in the preset traffic operation state, before determining the traffic operation state of the target lane based on the third feature to be identified; a third sample feature acquisition module, configured to obtain a third sample feature representing the probability distribution of vehicle speeds in the third sample lane during the third sample green light based on the third sample vehicle passage data; the state determination module 1203 is specifically configured to: for each preset traffic operation state, calculate the similarity between the third sample feature of the preset traffic operation state and the third feature to be identified using a second similarity algorithm; wherein the second similarity algorithm includes at least one of the following: bulldozing distance algorithm and Mahalanobis distance algorithm; and determine the preset traffic operation state with the highest similarity as the traffic operation state of the target lane.

[0291] In some embodiments, the third sample feature acquisition module is specifically used to: determine the driving speed of each vehicle in the third sample lane during the third sample green light period based on the third sample vehicle data, as the sample driving speed; determine the preset driving speed interval to which each sample driving speed belongs; wherein, each preset driving speed interval is adjacent and each preset driving speed interval has the same length; for each preset driving speed interval, calculate the ratio of the sample driving speed belonging to that preset driving speed interval to all sample driving speeds, to obtain the third sample feature;

[0292] The feature acquisition module 1202 is specifically used for: determining the speed of each vehicle in the target lane during the target green light period based on the target vehicle data, as the speed to be processed; determining the preset speed range to which each speed to be processed belongs; and for each preset speed range, calculating the ratio of the speed to be processed belonging to that preset speed range among all speeds to be processed, to obtain the third feature to be identified.

[0293] In some embodiments, the traffic flow discrete state is: discrete, weak car-following, or strong car-following; the traffic demand state is: low traffic demand or high traffic demand; the traffic operation state is: slow or smooth; the scenario determination module 1204 is specifically used to: if at least one target lane in each direction of the target intersection meets the first screening condition, then determine the scenario type of the target intersection in the time period to be controlled as queue equilibrium type; wherein, the first screening condition indicates: the traffic flow discrete state is strong car-following, the traffic demand state is high traffic demand, and the traffic operation state is slow; if the target lane in each direction of the target intersection meets the first screening condition, then determine the scenario type of the target intersection in the time period to be controlled as queue equilibrium type. If, among the target lanes, there are no target lanes that meet the first screening condition, and at least one target lane meets the second screening condition, then the scenario type of the target intersection during the control period is determined to be capacity-related. The second screening condition indicates that the traffic flow discrete state is strong or weak car-following, the traffic demand state is high traffic demand, and the traffic operation state is smooth. If, among the target lanes in each direction of the target intersection, there are no target lanes that meet the first screening condition and no target lanes that meet the second screening condition, then the scenario type of the target intersection during the control period is determined to be efficiency-enhancing.

[0294] In some embodiments, the apparatus further includes:

[0295] The first control strategy determination module is used to adopt a first type of traffic signal control strategy at the target intersection if the scenario type of the target intersection during the control period is queue balancing; wherein the first type of traffic signal control strategy is used to shorten the queue length of vehicles at the target intersection. The second control strategy determination module is used to adopt a second type of traffic signal control strategy at the target intersection if the scenario type of the target intersection during the control period is capacity-related; wherein the second type of traffic signal control strategy is used to improve the vehicle capacity of the target intersection. The third control strategy determination module is used to adopt a third type of traffic signal control strategy at the target intersection if the scenario type of the target intersection during the control period is efficiency-improvement; wherein the third type of traffic signal control strategy is used to reduce the delay time of vehicles at the target intersection.

[0296] This application also provides an electronic device, such as... Figure 13 As shown, it includes:

[0297] Memory 1301 is used to store computer programs;

[0298] When the processor 1302 executes the program stored in the memory 1301, it implements the steps of any of the intersection scene recognition methods in the above embodiments.

[0299] Furthermore, the aforementioned electronic device may also include a communication bus and / or a communication interface, with the processor 1302, the communication interface, and the memory 1301 communicating with each other via the communication bus.

[0300] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0301] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0302] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0303] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0304] In another embodiment provided in this application, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, implements the steps of any of the above-described intersection scene recognition methods.

[0305] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the intersection scene recognition methods described above.

[0306] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a solid-state drive (SSD), etc.

[0307] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0308] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments for apparatus, electronic devices, and computer-readable storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0309] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application are included within the scope of protection of this application.

Claims

1. A method for recognizing intersection scenes, characterized in that, The method includes: For each target lane in each direction of the target intersection, obtain the target vehicle passage data of that target lane during the target green light period within the time period to be controlled; Based on the target vehicle passage data, a first feature to be identified is obtained, representing the probability distribution of the headway distance between vehicles in the target lane during the target green light period; a second feature to be identified is obtained, representing the change in the number of vehicles allowed to pass in the target lane during each statistical time period of the target green light period; and a third feature to be identified is obtained, representing the probability distribution of the vehicle speed in the target lane during the target green light period. The traffic flow discrete state of the target lane is determined based on the first feature to be identified, the traffic demand state of the target lane is determined based on the second feature to be identified, and the traffic operation state of the target lane is determined based on the third feature to be identified; the traffic flow discrete state is: discrete, weak car-following, or strong car-following; the traffic demand state is: low traffic demand or high traffic demand; the traffic operation state is: slow or smooth. If at least one target lane in each direction of the target intersection meets the first screening condition, then the scenario type of the target intersection during the control period is determined to be a queuing equilibrium type; wherein, the first screening condition indicates that the traffic flow discrete state is strong car-following, the traffic demand state is high traffic demand, and the traffic operation state is slow. If none of the target lanes in each direction of the target intersection meet the first screening condition, and at least one target lane meets the second screening condition, then the scenario type of the target intersection during the control period is determined to be a capacity type; wherein, the second screening condition indicates that the traffic flow discrete state is strong car-following or weak car-following, the traffic demand state is high traffic demand, and the traffic operation state is smooth. If none of the target lanes in each direction of the target intersection meet the first screening condition and none of the target lanes meet the second screening condition, then the scenario type of the target intersection during the control period is determined to be efficiency improvement type.

2. The method according to claim 1, characterized in that, During the specified control period, the target lane has the highest number of vehicles allowed to pass in any direction of the road segment at the target intersection.

3. The method according to claim 1, characterized in that, Before determining the discrete state of traffic flow for the target lane based on the first feature to be identified, the method further includes: For each preset traffic flow discrete state, the first sample vehicle passage data of the first sample lane during the first sample green light period is obtained when the first sample lane is in the preset traffic flow discrete state. Based on the first sample vehicle passage data, a first sample feature is obtained representing the probability distribution of the headway distance in the first sample lane during the first sample green light period; The step of determining the discrete state of traffic flow for the target lane based on the first feature to be identified includes: For each preset traffic flow discrete state, the similarity between the first sample feature of the preset traffic flow discrete state and the first feature to be identified is calculated using a first similarity algorithm; wherein, the first similarity algorithm includes at least one of the following: bulldozer distance algorithm and Mahalanobis distance algorithm; The preset traffic flow discrete state with the highest similarity is determined as the traffic flow discrete state of the target lane.

4. The method according to claim 3, characterized in that, The first sample feature obtained based on the first sample vehicle passage data, representing the probability distribution of the headway distance in the first sample lane during the first sample green light period, includes: Based on the first sample vehicle passage data, the headway of each vehicle at a specified position in the first sample lane during the first sample green light period is determined as the sample headway. Determine the preset headway time interval to which each sample headway belongs; wherein, each preset headway time interval is adjacent and has the same length; For each preset headway interval, calculate the ratio of the sample headway of the vehicle belonging to that preset headway interval to the total sample headway, and obtain the first sample feature; The first feature to be identified, based on the target vehicle passage data, represents the probability distribution of the headway distance between vehicles in the target lane during the target green light period, including: Based on the target vehicle passage data, the headway of each vehicle at a specified position in the target lane during the target green light period is determined as the headway to be processed. Determine the preset headway interval to which each headway to be processed belongs; For each preset headway time interval, the ratio of the headway time interval to be processed belonging to that preset headway time interval to all headway time intervals to be processed is calculated to obtain the first feature to be identified.

5. The method according to claim 1, characterized in that, Before determining the traffic demand status of the target lane based on the second feature to be identified, the method further includes: For each preset traffic demand state, obtain the second sample vehicle passage data of the second sample lane during the second sample green light when the second sample lane is in that preset traffic demand state. Based on the second sample vehicle passage data, a second sample feature is obtained, representing the change in the number of vehicles allowed to pass through the second sample lane during each statistical time period of the second sample green light period. Determining the traffic demand status of the target lane based on the second feature to be identified includes: For each preset traffic demand state, the similarity between the second sample feature of the preset traffic demand state and the second feature to be identified is calculated; The preset traffic demand state with the highest similarity is determined as the traffic demand state of the target lane.

6. The method according to claim 5, characterized in that, The second sample green light period includes multiple sample green light cycles; the target green light period includes multiple target green light cycles; The second sample feature, derived from the second sample vehicle passage data, represents the change in the number of vehicles allowed to pass through the second sample lanes during each statistical time period of the second sample green light period. This feature includes: For each sample green light cycle, based on the second sample vehicle passage data of that sample green light cycle, the number of vehicles released in the second sample lane during each statistical time period of that sample green light cycle is determined as the number of vehicles released during each statistical time period of that sample green light cycle. For any given statistical time period, the mean number of vehicles allowed to pass during each green light cycle in that statistical time period is calculated to obtain a second sample feature representing the change in the number of vehicles allowed to pass through the second sample lane in each statistical time period during the second sample green light period. The second feature to be identified, derived based on the target vehicle passage data, represents the change in the number of vehicles allowed to pass through the target lane during each statistical time period of the target green light period, including: For each target green light cycle, based on the target vehicle passage data of that target green light cycle, the number of vehicles released in the target lane within each statistical time period of that target green light cycle is determined as the number of vehicles released in each statistical time period of that target green light cycle. For any given statistical time period, the average number of vehicles allowed to pass during that statistical time period for each target green light cycle is calculated, resulting in a second feature to be identified that represents the change in the number of vehicles allowed to pass through the target lane during each statistical time period of the target green light period.

7. The method according to claim 1, characterized in that, Before determining the traffic operation status of the target lane based on the third feature to be identified, the method further includes: For each preset traffic operation state, obtain the third sample vehicle passage data of the third sample lane during the third sample green light period when the third sample lane is in the preset traffic operation state; Based on the third sample vehicle passage data, a third sample feature is obtained that represents the probability distribution of vehicle speeds in the third sample lane during the third sample green light period. Determining the traffic operation status of the target lane based on the third feature to be identified includes: For each preset traffic operation state, a second similarity algorithm is used to calculate the similarity between the third sample feature of the preset traffic operation state and the third feature to be identified; wherein, the second similarity algorithm includes at least one of the following: bulldozer distance algorithm and Mahalanobis distance algorithm; The preset traffic operation state with the highest similarity is determined as the traffic operation state of the target lane.

8. The method according to claim 7, characterized in that, The third sample feature, derived from the third sample vehicle passage data, represents the probability distribution of vehicle speeds in the third sample lane during the third sample green light period, including: Based on the third sample vehicle passage data, the speed of each vehicle in the third sample lane during the third sample green light period is determined as the sample speed. Determine the preset driving speed range to which each sample driving speed belongs; wherein, each preset driving speed range is adjacent and the length of each preset driving speed range is the same; For each preset driving speed range, the ratio of the sample driving speed belonging to that preset driving speed range to the total sample driving speed is calculated to obtain the third sample feature; The third feature to be identified, derived from the target vehicle data, representing the probability distribution of vehicle speeds in the target lane during the target green light period, includes: Based on the target vehicle passage data, the speed of each vehicle in the target lane during the target green light period is determined as the speed to be processed. Determine the preset speed range to which each speed to be processed belongs; For each preset driving speed range, the ratio of the driving speed to be processed belonging to that preset driving speed range to all driving speeds to be processed is calculated to obtain the third feature to be identified.

9. The method according to claim 1, characterized in that, The method further includes: If the scenario type of the target intersection during the control period is queue balancing, then a first type of traffic signal control strategy is adopted at the target intersection; wherein, the first type of traffic signal control strategy is used to shorten the queue length of vehicles at the target intersection; If the scenario type of the target intersection during the control period is capacity-type, then a second type of traffic signal control strategy will be adopted at the target intersection; wherein, the second type of traffic signal control strategy is used to improve the vehicle capacity of the target intersection; If the scenario type of the target intersection during the control period is efficiency improvement type, then a third type of traffic signal control strategy will be adopted at the target intersection; wherein, the third type of traffic signal control strategy is used to reduce the delay time of vehicles at the target intersection.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 1-9.

Images