Multi-lane shunting management and control method, equipment and medium

Abstract

The embodiment of the invention discloses a multi-lane shunting management and control method, equipment and a medium, is applied to the technical field of intelligent traffic, and is used for solving the problem that an existing shunting mode is low in efficiency and low in accuracy. The method comprises the following steps: dividing monitoring sub-regions based on road network data, and determining topological connection data of each sub-region by using historical vehicle driving data; and identifying target traffic characteristics of each sub-region through clustering analysis, and determining a corresponding shunting traffic region according to the target traffic characteristics. And identifying the current congested road section and the congested direction according to the real-time vehicle driving data, and formulating a shunting strategy in combination with the condition of the shunting traffic area. And inputting the passing speed of the current congested road section, the driving data of the shunting traffic area and the shunting strategy into a preset flow prediction model, and outputting and adjusting the shunting strategy according to the model to realize dynamic shunting management and control.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a multi-lane diversion control method, device and medium. Background Art

[0002] With the accelerated advancement of urbanization and the continuous increase in the number of cars, the problem of urban traffic congestion is worsening at an unprecedented rate, becoming a thorny problem that restricts the efficiency of urban operations and the improvement of residents' quality of life. Especially during peak hours in the morning and evening and in busy areas such as commercial centers and transportation hubs, traffic congestion is particularly prominent, which has brought a heavy burden to the daily operation of the city. Traffic congestion not only prolongs the commuting time of citizens and increases travel costs, but also seriously affects the city's logistics transportation and emergency service response speed. Vehicles stranded for a long time in congested sections not only consume a lot of fuel resources, aggravate air pollution, but also greatly reduce the overall operating efficiency of the urban transportation network. In addition, long waiting and poor driving can easily cause driver fatigue and anxiety, increase the risk of traffic accidents, and pose a potential threat to public safety. Therefore, traffic diversion to alleviate urban traffic congestion and improve road capacity is of great significance for daily transportation.

[0003] At present, traditional traffic management methods often rely on manual intervention and static traffic signal control. Since diversion control is based on manual experience, there is a lack of in-depth analysis of road network data, making it difficult to accurately divide monitoring sub-areas and accurately identify the traffic characteristics of each area. It is difficult to adapt to complex traffic flow changes and real-time road conditions, so it is difficult to alleviate traffic congestion during peak hours. In addition, the method that relies on fixed traffic signal control lacks dynamic adjustment capabilities and is difficult to perform dynamic diversion based on actual conditions, making diversion control less efficient. In addition, when formulating diversion strategies, existing diversion control methods often only consider a single path or time factor, lack comprehensive considerations, and are difficult to achieve continuous optimization of diversion control. Summary of the invention

[0004] In order to solve the above technical problems, one or more embodiments of the present specification provide a multi-lane diversion control method, device and medium.

[0005] One or more embodiments of this specification adopt the following technical solutions: One or more embodiments of this specification provide a multi-lane diversion control method, the method comprising: Dividing the current monitoring area based on the road network related data to obtain multiple monitoring sub-areas, and determining the topological connection data corresponding to each monitoring sub-area according to the historical vehicle driving data of each monitoring sub-area; Performing cluster analysis on each of the monitoring sub-areas according to the topological connection data, determining target traffic characteristics of each of the monitoring sub-areas, and determining a diversion traffic area corresponding to each of the monitoring sub-areas based on the target traffic characteristics; Determine the current congested section based on the current vehicle driving data of each monitoring sub-area, and calculate the speed of each direction of multiple lanes in the current congested section, so as to divide and determine the congested direction and the non-congested direction; Determine a diversion traffic area corresponding to the current congested road section, and determine a corresponding diversion strategy according to the traffic speeds in the congested direction and the non-congested direction and the current vehicle driving data of the corresponding diversion traffic area; wherein the diversion strategy includes: a diversion path and a diversion time; The traffic speed of each lane of the current congested road section, the current vehicle driving data of the corresponding diversion traffic area and the diversion strategy are input into the preset traffic prediction model to adjust the diversion strategy based on the output results to perform diversion control.

[0006] Optionally, in one or more embodiments of the present specification, the current monitoring area is divided based on the road network related data to obtain multiple monitoring sub-areas, specifically including: Based on the multi-data interface, obtaining the road network related data of the current monitoring area; wherein the road network related data includes: basic road network data, detector detection data, and road operation data; According to the traffic network structure corresponding to the basic road network data and the preset area boundary conditions, the initial monitoring sub-area of ​​the multi-traffic level corresponding to the current monitoring area is determined; wherein the traffic levels include: intersection level, intercity expressway level, suburban expressway level, and living area road level; the preset area boundary conditions include: urban area level data, urban geographic boundary data, and monitoring area restriction data; According to the location information corresponding to the detector detection data and the operation data, matching is performed with the basic road network data to divide the detector detection data and the operation data into corresponding initial monitoring sub-areas; The internal traffic volume of each of the initial monitoring sub-areas is determined based on the detection data of the detector and the operation data, so as to dynamically adjust the scope of the initial monitoring sub-area according to the internal traffic volume to obtain multiple monitoring sub-areas.

[0007] Optionally, in one or more embodiments of the present specification, determining the topological connection data corresponding to each monitoring sub-area according to the historical vehicle driving data of each monitoring sub-area specifically includes: Acquire historical vehicle driving data of each monitoring sub-area, and identify the entry and exit sections of each monitoring sub-area based on the driving direction data corresponding to each vehicle driving data and the boundary position of each monitoring sub-area; The entry road section and the exit road section are respectively used as traversal start nodes, so as to traverse each road section of the monitoring sub-area based on the traversal start node and the preset traversal direction, obtain an entry road section sequence and an exit road section sequence, and use each road section in the entry road section sequence and the exit road section sequence as a node; wherein the traversal direction is: clockwise or counterclockwise; Determine the adjacent road sections of each of the nodes, and determine the road section type of each of the nodes based on the basic road network data corresponding to the adjacent road sections; wherein the road section type includes: divergence type and confluence type; Determine the vehicle driving frequency and the number of vehicles traveling between the nodes according to the historical vehicle driving data of each node, and determine the connectivity strength of each node according to the vehicle driving frequency, the number of vehicles traveling and the road section type; Based on the connectivity strength corresponding to each of the nodes and the sequence name corresponding to the node, a connectivity path between each of the nodes is determined to obtain topological connection data corresponding to each of the monitoring sub-areas.

[0008] Optionally, in one or more embodiments of the present specification, cluster analysis is performed on each of the monitoring sub-areas according to the topological connection data to determine the target traffic characteristics of each of the monitoring sub-areas, specifically including: Determine the adjacent nodes of each node in the monitoring sub-area based on the topological connection data to construct a neighbor node set corresponding to each node; Sequentially acquiring the neighboring nodes corresponding to each neighboring node in the neighboring node set, so as to determine the common neighboring nodes corresponding to the node and the neighboring nodes of the node based on the intersection of the neighboring node set and the neighboring nodes corresponding to each neighboring node; Constructing a feature vector corresponding to each of the nodes based on the historical vehicle travel data corresponding to the adjacent nodes of each of the nodes and the common neighbor nodes; Perform similarity clustering based on the feature vectors corresponding to each of the nodes to obtain multiple clusters, and extract common features of the nodes in each cluster; The common features of the clusters are summarized and the position distribution corresponding to the clusters is determined, so as to determine the target traffic features of each monitoring sub-area by combining the common features of the clusters and the corresponding position distribution.

[0009] Optionally, in one or more embodiments of the present specification, determining the diversion traffic area corresponding to each of the monitoring sub-areas based on the target traffic characteristics specifically includes: Based on the target traffic characteristics of each monitoring sub-area, identifying the traffic bottleneck data of each monitoring sub-area; wherein the traffic bottleneck data includes: traffic peak time period, congestion node range, and road capacity; Determine the diversion capacity data corresponding to each of the monitoring sub-areas according to the traffic bottleneck data corresponding to each of the monitoring sub-areas; wherein the diversion capacity data includes: diversion time and diversion capacity; Determine, based on the diversion capacity data corresponding to the adjacent monitoring sub-areas of the monitoring sub-area, an optional diversion traffic area corresponding to the diversion time and the traffic peak period; The diversion capacity value of each of the optional diversion traffic areas is determined according to the diversion capacity pair of each of the optional diversion traffic areas, so as to determine the diversion traffic area corresponding to each of the monitoring sub-areas based on the diversion capacity value.

[0010] Optionally, in one or more embodiments of the present specification, determining a diversion traffic area corresponding to the current congested road section, and determining a corresponding diversion strategy according to the traffic speeds in the congested direction and the non-congested direction and the current vehicle driving data of the corresponding diversion traffic area, specifically includes: Determine a plurality of traffic areas to be diverted according to the diverted traffic area corresponding to the monitoring sub-area where the current congested road section is located, and filter the plurality of traffic areas to be diverted according to the incoming road section or the outgoing road section within the preset range of the current congested road section to obtain the diverted traffic area corresponding to the current congested road section; Determine the number of vehicles waiting to enter the non-congested direction based on current vehicle travel data of the entry section and the exit section corresponding to the non-congested direction in the diverted traffic area; Determine the waiting time length of the non-congested direction according to the passing speed of the non-congested direction and the data of the relatively waiting vehicles; Determine the growth rate of the congestion queue in the congestion direction based on the traffic speed in the congestion direction and the current vehicle driving data of the incoming road section and the outgoing road section corresponding to the non-congestion direction in the diverted traffic area; Determine the travel time ratio of the congested direction according to the growth rate of the congested queue, the length of the waiting time and the spatial information of the road section in the non-congested direction; If the travel time ratio is greater than a preset threshold, an alternate route corresponding to the congestion direction in the diversion traffic area is activated.

[0011] Optionally, in one or more embodiments of the present specification, before inputting the traffic speed of each lane of the current congested road section, the current vehicle driving data of the corresponding diversion traffic area, and the diversion strategy into a preset traffic prediction model, the method further includes: Collecting historical vehicle driving data corresponding to the historical diversion strategy of the current monitoring area, and dividing the historical vehicle driving data corresponding to the historical diversion strategy based on a preset ratio to obtain training set data and verification set data; Inputting the training set data into the initial time series model for iterative training to obtain a trained time series model; The trained time series model is evaluated based on the validation set data, so as to adjust the trained time series model according to the evaluation result to obtain a preset traffic prediction model.

[0012] Optionally, in one or more embodiments of the present specification, based on the output result, adjusting the diversion strategy to perform diversion control specifically includes: If it is determined based on the output result that the traffic flow is on a downward trend, determining that the current congested road section maintains the diversion strategy; If it is determined that the traffic flow is on an upward trend, then determining the current travel time ratio corresponding to the current diversion strategy; If the current travel time ratio is less than the preset threshold, the current travel time is gradually increased, and the diversion strategy is adjusted to perform diversion control; If the current travel time ratio is greater than the preset threshold, determining the backup sections corresponding to each exit section in the congested direction; Based on the distance between each of the backup road sections and the congested road section and the traffic flow of each of the backup road sections, the activation order of each of the backup road sections is determined, so as to activate the backup road sections in sequence for traffic diversion.

[0013] One or more embodiments of this specification provide a multi-lane traffic diversion control device, the device comprising: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: execute any of the above-mentioned methods.

[0014] One or more embodiments of the present specification provide a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions are configured to be able to: execute any of the above-described methods.

[0015] At least one of the above technical solutions adopted in the embodiments of this specification can achieve the following beneficial effects: Based on historical and real-time data, traffic characteristics are determined through cluster analysis, diversion strategies are formulated intelligently, and strategies are dynamically adjusted according to the output results of the traffic prediction model, which can more effectively deal with traffic congestion problems and improve road traffic efficiency. By carefully dividing the monitoring area, the refinement of traffic management can be achieved, and congested sections can be quickly identified and responded to, improving traffic diversion efficiency. Comprehensively considering the speed of traffic in the congested and non-congested directions and the vehicle driving data in the diversion traffic area, accurate diversion paths and diversion times can be formulated, which can effectively alleviate the traffic pressure in the congested direction with the help of road space in the non-congested direction, thereby improving the utilization rate of the road. In addition, the optimization method based on the prediction model can more accurately predict future traffic conditions, thereby formulating a more reasonable diversion strategy and achieving more effective diversion control. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art description. Obviously, the drawings described below are only some embodiments recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. In the drawings: Figure 1 A flowchart of a multi-lane traffic diversion control method provided in an embodiment of this specification; Figure 2 A schematic diagram of the structure of a multi-lane traffic diversion control device provided in an embodiment of this specification; Figure 3 A schematic diagram of the structure of a non-volatile storage medium provided in an embodiment of this specification. DETAILED DESCRIPTION

[0017] The embodiments of this specification provide a multi-lane diversion control method, device and medium.

[0018] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0019] like Figure 1 As shown, the present specification embodiment provides a flowchart of a multi-lane diversion control method. Figure 1 It can be seen that in one or more embodiments of this specification, a multi-lane diversion control method includes: S101: Divide the current monitoring area based on road network related data to obtain multiple monitoring sub-areas, and determine the topological connection data corresponding to each monitoring sub-area according to the historical vehicle driving data of each monitoring sub-area.

[0020] In large urban traffic networks, the road network structure is complex and the traffic volume is large, so it is difficult to directly manage and optimize the entire road network. Therefore, in order to decompose the complex traffic control tasks and thus improve the efficiency of diversion control in the embodiments of this specification, the current monitoring area will be divided according to the road network related data to obtain multiple monitoring sub-areas. Then, based on the historical vehicle driving data of each monitoring sub-area, the topological connection data corresponding to each monitoring sub-area is determined. By determining the topological connection data, the traffic connections and mutual influences between the monitoring sub-areas can be obtained, which helps to formulate a more accurate diversion strategy.

[0021] Specifically, in one or more embodiments of the present specification, dividing the current monitoring area based on the road network related data to obtain multiple monitoring sub-areas specifically includes the following process: First, based on the multi-data interface, the road network related data of the current monitoring area is obtained. Among them, it can be understood that the road network related data includes: basic road network data such as road centerline data, road grade data, intersection information, road attribute information, ancillary facilities information, etc.; detector detection data such as traffic flow data, vehicle speed data, vehicle type data, etc.; road operation data such as highway fees, floating bridge fees, bridge fees, etc. Then, according to the traffic network structure corresponding to the basic road network data and the preset area boundary conditions, the initial monitoring sub-area of ​​the multi-traffic level corresponding to the current monitoring area is determined. The traffic levels include: intersection layer, intercity highway layer, suburban express layer, and living area road layer; and the preset area boundary conditions include: urban area level data, urban geographic boundary data, and monitoring area restriction data. Then, according to the location information corresponding to the detector detection data and the operation data, they are matched with the basic road network data, so as to divide the detector detection data and the operation data into the corresponding initial monitoring sub-area. The internal traffic volume of each initial monitoring sub-area is then determined based on the detector detection data and the operation data, so as to dynamically adjust the scope of the initial monitoring sub-area according to the internal traffic volume to obtain multiple monitoring sub-areas.

[0022] In this process, the road network-related data from different sources can be efficiently integrated based on the multi-data interface, which improves the comprehensiveness and accuracy of the data and provides a solid foundation for subsequent analysis and processing. The detector detection data and operation data are matched with the basic road network data according to the location information, and these data can be accurately divided into the corresponding initial monitoring sub-areas. Then, according to the traffic network structure corresponding to the basic road network data and the preset area boundary conditions, the initial monitoring sub-areas of multiple traffic levels corresponding to the current monitoring area can be scientifically determined. This division method fully considers the actual situation and monitoring needs of the road network and improves the pertinence and effectiveness of monitoring. Dividing the traffic level into intersection level, intercity highway level, suburban express level and living area road level, etc., is conducive to targeted monitoring and management of traffic conditions at different levels. In addition, the initial monitoring sub-area is dynamically adjusted according to the changes in traffic flow, so that the monitoring sub-area can be flexibly adjusted according to the actual situation, avoiding the limitations that may be brought about by fixed monitoring areas.

[0023] Further, in one or more embodiments of the present specification, determining the topological connection data corresponding to each monitoring sub-area according to the historical vehicle driving data of each monitoring sub-area specifically includes the following process: First, in order to conduct accurate analysis based on actual traffic conditions and avoid errors caused by subjective judgment, the embodiment of this specification obtains the historical vehicle driving data of each monitoring sub-area, and identifies the entry and exit sections of each monitoring sub-area based on the driving direction data corresponding to each vehicle driving data and the boundary position of each monitoring sub-area. Then, the entry section and the exit section are respectively used as the traversal starting nodes, so as to traverse each section of the monitoring sub-area according to the traversal starting node and the preset traversal direction, obtain the entry section sequence and the exit section sequence, and use each section in the entry section sequence and the exit section sequence as a node. It should be noted that the traversal direction is: clockwise or counterclockwise. Determine the adjacent sections of each node, and then determine the section type of each node according to the basic road network data corresponding to the adjacent sections; wherein the section type includes: diversion type and confluence type. Then, according to the historical vehicle travel data of each node, the vehicle travel frequency and the number of vehicles between each node are determined, so as to determine the connectivity strength of each node according to the vehicle travel frequency, the number of vehicles and the road section type. Based on the connectivity strength corresponding to each node and the sequence name corresponding to the node, the connectivity path between each node is determined to obtain the topological connection data corresponding to each monitoring sub-area.

[0024] For example, in a certain application scenario, the vehicle driving data of the monitoring sub-area in the past period of time is extracted from the traffic monitoring system, including the vehicle's driving direction, timestamp, starting point and end point information. According to the vehicle's driving direction and the boundary position of the monitoring sub-area, it can be determined which road sections are the vehicle's entry sections into the monitoring sub-area and which road sections are the vehicle's exit sections out of the monitoring sub-area. For example, if a vehicle's driving direction is from east to west and its starting point is on the east boundary of the monitoring sub-area, then the road section it is on is the entry section; if a vehicle's driving direction is from west to east and its end point is on the west boundary of the monitoring sub-area, then the road section it is on is the exit section. Then, select the entry section as the traversal start node and set the traversal direction to clockwise. Starting from the entry section, traverse each road section in the monitoring sub-area in a clockwise direction until returning to the exit section. In this way, a sequence of incoming sections is obtained, and a sequence of outgoing sections can be obtained in the same way. In the traversal process, each section can be regarded as a node. For each node, its adjacent sections can be determined according to the basic road network data. According to the connection between adjacent sections, the section type of the node is determined, such as the divergence type (multiple sections merge into one section) or the confluence type (one section divides into multiple sections). According to the historical vehicle driving data, the vehicle driving frequency and the number of vehicles between each node, that is, between adjacent sections, are counted. Combined with the section type and vehicle driving data, a connectivity strength value is assigned to each node, which reflects the size and stability of the traffic flow between nodes. Then, based on the connectivity strength and sequence name of each node, the connectivity path between nodes is constructed. These connectivity paths together constitute the topological connection data of the monitoring sub-area, which describes the connection relationship and traffic flow characteristics between the sections in the monitoring sub-area.

[0025] In the above process of obtaining the topological connection data corresponding to each monitoring sub-area, by setting the traversal start node and the preset traversal direction, all the road sections in the monitoring sub-area can be systematically traversed to construct a clear sequence of entry and exit sections, which not only improves the analysis efficiency, but also ensures the coherence and integrity between the sections. According to the basic road network data corresponding to the adjacent sections, the process can accurately divide the road section types, which helps to more accurately evaluate the mutual influence and potential conflict points between the sections. In addition, the method of evaluating the connectivity strength between each node by combining the road section type and historical vehicle driving data not only considers the impact of traffic flow, but also considers the impact of the road section type on traffic flow, thereby providing a more comprehensive connectivity evaluation. Based on the connectivity strength and sequence name of each node, the process can construct the topological connection data corresponding to the monitoring sub-area, which not only helps to reveal the internal structure of the traffic network, but also provides a scientific basis for subsequent traffic congestion relief, traffic accident prevention and traffic resource optimization.

[0026] S102: performing cluster analysis on each of the monitoring sub-areas according to the topological connection data, determining target traffic characteristics of each of the monitoring sub-areas, and determining a diversion traffic area corresponding to each of the monitoring sub-areas based on the target traffic characteristics.

[0027] In order to determine the diversion traffic area corresponding to each monitoring sub-area, so as to carry out traffic control for the diversion traffic area and alleviate traffic congestion. In the embodiment of this specification, each monitoring sub-area will be clustered according to the topological connection data to determine the target traffic characteristics of each monitoring sub-area. The target traffic characteristics help to reasonably allocate traffic resources and determine the diversion traffic area corresponding to the monitoring sub-area.

[0028] Specifically, in one or more embodiments of the present specification, cluster analysis is performed on each monitoring sub-area according to the topological connection data to determine the target traffic characteristics of each monitoring sub-area, which specifically includes the following process: First, the neighboring nodes of each node in the monitoring sub-area are determined based on the topological connection data, so as to construct the neighboring node set corresponding to each node. Then, the neighboring nodes corresponding to each neighboring node in the neighboring node set are obtained in turn, so as to determine the common neighboring nodes corresponding to the node and the neighboring nodes of the node according to the intersection of the neighboring node set and the neighboring nodes corresponding to each neighboring node. Based on the historical vehicle driving data corresponding to the neighboring nodes and common neighboring nodes of each node, the feature vector corresponding to each node is constructed. Similarity clustering is performed according to the feature vectors corresponding to each node to obtain multiple clusters, and the common features of the nodes in each cluster are extracted. The common features of each cluster are summarized and the location distribution corresponding to each cluster is determined, so as to determine the target traffic features of each monitoring sub-area by combining the common features of each cluster with the corresponding location distribution.

[0029] To make it easier to understand the above, the following example is used to illustrate that, assuming there is a traffic monitoring sub-area, the neighboring nodes of each node are first determined based on the topological connection data. For example, node A is connected to nodes B, C, and D, so the neighbor node set of node A is {B, C, D}. Then, the neighbor nodes of the neighboring nodes in each neighbor node set are obtained in turn, and the intersection is found. Take node A as an example: when the set of adjacent nodes of node B is assumed to be {A, E, F}, the intersection of node B and the neighbor node set of node A {B, C, D} is {B}, but since we are concerned about the intersection of non-self nodes, {B} should actually be excluded, and the intersection with {C, D} in {E, F} is considered, which is assumed to be an empty set; when the set of adjacent nodes of node C is assumed to be {A, G, H}, the intersection of the set of adjacent nodes of node C with {B, C, D} is {A, C}, but {A} is also excluded, and the intersection with {B, D} in {G, H} is considered, which is assumed to be {G}; when the set of adjacent nodes of node D is assumed to be {A, I, J}, the intersection of the set of adjacent nodes of node D with {B, C, D} is {A, D}, excluding {A}, and the intersection with {B, C} in {I, J} is considered, which is assumed to be an empty set. Based on the above information, the common neighbor nodes of node A and adjacent nodes B, C, and D can be determined. Then, based on the historical vehicle driving data corresponding to the adjacent nodes and common neighbor nodes of each node, a feature vector can be constructed. For example, for node A, the following features can be considered: the number of adjacent nodes, such as 3: B, C, and D; the traffic flow with each adjacent node, such as the traffic flow with B is X, the traffic flow with C is Y, and the traffic flow with D is Z; the number of common neighbor nodes and the corresponding traffic flow, such as the traffic flow with C's common neighbor G is W. Combining these features, it constitutes the feature vector of node A. Then, the similarity clustering algorithm can be used to cluster the nodes according to the feature vector. Assume that two clusters are obtained after clustering: cluster 1 and cluster 2. For cluster 1, the common features of the nodes within it can be extracted, such as large traffic flow and being mostly located on urban main roads. Determine the location distribution of cluster 1, such as mainly distributed in the central area of ​​the city. Similarly, for cluster 2, its common features and location distribution can also be extracted. Finally, the target traffic characteristics of each monitoring sub-area are determined by combining the common characteristics and location distribution of each cluster. For example, the target traffic characteristics of monitoring sub-area 1, which includes cluster 1, are: heavy traffic volume, mostly located on main roads, and prone to traffic congestion. The target traffic characteristics of monitoring sub-area 2, which includes cluster 2, are: moderate traffic volume, mostly located on secondary roads, and relatively stable traffic conditions.

[0030] In the above process of determining the target traffic characteristics of each monitoring sub-area, by constructing a set of neighbor nodes and determining common neighbor nodes, the complex relationship between nodes in the traffic network can be deeply explored, revealing the direct and indirect connections between nodes, and further using historical vehicle driving data to construct feature vectors, so that each node not only considers its adjacent nodes, but also considers the influence of common neighbor nodes, thereby more comprehensively reflecting the traffic characteristics of the node. In addition, through the similarity clustering algorithm, the nodes can be divided into multiple clusters based on the similarity of the feature vectors, so that nodes with similar traffic characteristics are classified together, and common features are extracted from the clusters, which can accurately reflect the traffic characteristics of the nodes in the clusters and then determine the target traffic characteristics of each monitoring sub-area, which is helpful for the subsequent formulation of more targeted traffic management strategies and more reasonable control and regulation of traffic diversion.

[0031] Further, in one or more embodiments of the present specification, determining the diversion traffic area corresponding to each monitoring sub-area based on the target traffic characteristics specifically includes the following process: First, according to the target traffic characteristics of each monitoring sub-area obtained by the above process, the traffic bottleneck data of each monitoring sub-area is identified. Among them, it can be understood that the traffic bottleneck data includes: traffic peak time period, congestion node range and road capacity and other data. Then, according to the traffic bottleneck data corresponding to each monitoring sub-area, the diversion capacity data corresponding to each monitoring sub-area is determined. Among them, it should also be noted that the diversion capacity data includes: diversion time and diversion capacity and other data. Then, according to the diversion capacity data corresponding to the adjacent monitoring sub-area of ​​the monitoring sub-area, the optional diversion traffic area corresponding to the diversion time and the traffic peak time period is determined. According to the diversion capacity of each optional diversion traffic area, the diversion capacity value of each optional diversion traffic area is determined, so as to determine the diversion traffic area corresponding to each monitoring sub-area based on the diversion capacity value. By identifying the traffic bottleneck data of each monitoring sub-area, such as traffic peak time period, congestion node range and road capacity, etc., the congestion point in urban traffic can be accurately located. Determining the diversion capacity data of each monitored sub-area based on traffic bottleneck data, including diversion time and diversion capacity, will help to subsequently determine a reliable diversion strategy to effectively guide traffic flow from congested areas to other areas, thereby alleviating traffic pressure.

[0032] S103: Determine the current congested road section according to the current vehicle driving data of each monitoring sub-area, and calculate the travel speeds of multiple lanes in each direction in the current congested road section, so as to divide and determine the congested direction and the non-congested direction.

[0033] In order to determine the congested sections and directions of each monitoring sub-area in the current monitoring area, the current congested section will be determined according to the current vehicle driving data of each monitoring sub-area in the embodiment of this specification, so as to calculate the speed of each direction of multiple lanes in the current congested section, and then determine the congested direction and non-congested direction of multiple lanes according to the speed of each lane and each direction. Congestion is often accompanied by an increase in the risk of traffic accidents. By timely identifying congested sections and directions, it is helpful to carry out targeted diversion for the congested sections and directions, thereby reducing the incidence of traffic accidents.

[0034] S104: Determine a diversion traffic area corresponding to the current congested road section, and determine a corresponding diversion strategy according to the traffic speeds in the congested direction and the non-congested direction and the current vehicle driving data of the corresponding diversion traffic area; wherein the diversion strategy includes: a diversion path and a diversion time.

[0035] Based on the above steps S102-S103, the diversion traffic area corresponding to the current congested road section is determined, and then the corresponding diversion strategy is determined according to the speed of traffic in the congested direction and the non-congested direction and the current vehicle driving data of the corresponding diversion traffic area. Among them, the diversion strategy includes a diversion path and a diversion time. In this process, by comprehensively considering the speed of traffic in the congested direction and the non-congested direction and the current vehicle driving data of the diversion area, the diversion path and the diversion time can be flexibly determined, thereby improving the diversion efficiency while ensuring the road utilization rate.

[0036] Specifically, in one or more embodiments of the present specification, determining a diversion traffic area corresponding to the current congested road section, and determining a corresponding diversion strategy according to the traffic speeds in the congested direction and the non-congested direction and the current vehicle driving data of the corresponding diversion traffic area, specifically includes the following process: First, according to the diversion traffic area corresponding to the monitoring sub-area where the current congested section is located, multiple traffic areas to be diverted are determined, and the multiple traffic areas to be diverted are filtered according to the entry section or exit section within the preset range of the current congested section to obtain the diversion traffic area corresponding to the current congested section. Then, according to the current vehicle driving data of the entry section and exit section corresponding to the non-congested direction in the diversion traffic area, the relative number of vehicles waiting to enter in the non-congested direction is determined. For example: there is a diversion traffic area, which has two main directions: the congested direction and the non-congested direction. Assume that there is only one entry section (section A) and one exit section (section B) corresponding to the non-congested direction. At this time, the vehicle driving data of the entry section is the number of vehicles entering in the current time period, 100 vehicles / hour, and the average vehicle entry speed: 60km / h, and the vehicle driving data of the exit section is the number of vehicles leaving in the current time period, 90 vehicles / hour, and the average vehicle exit speed: 70km / h. At this time, the number of vehicles entering - the number of vehicles leaving = 100 - 90 = 10 vehicles / hour, which means that in the non-congested direction, there are 10 vehicles entering per hour, that is, the number of vehicles entering is greater than the number of vehicles leaving. Then the exit section can accommodate more vehicles leaving, but the entry section may accumulate vehicles due to the slower speed. At this time, the relative number of vehicles waiting to enter can be approximated as the cumulative difference between the entry and exit within a certain period of time. For example, if the observation time is 1 hour, the relative number of vehicles waiting to enter is 10.

[0037] Then, according to the speed of the non-congested direction and the data of the vehicles waiting to enter, the waiting time length in the non-congested direction is determined. For example, in a certain application scenario, the speed of a certain section of road in the non-congested direction is 50km / h. At this time, there are 15 vehicles waiting to enter the section, and the average length of each vehicle is 5 meters. Then, the length of the road occupied by the waiting vehicles can be determined as follows: the length of the road occupied by the waiting vehicles = the number of vehicles × the average length of each vehicle = 15×5 = 75 meters. Then calculate the time required for the vehicle to pass through the section. Since the speed is 50km / h, it needs to be converted into meters / second for calculation: 50km / h = 50×1000 meters / 3600 seconds ≈ 13.89 meters / second, so the time required for the vehicle to pass 75 meters is: required time = road length ÷ speed = 75÷13.89≈5.39 seconds. Get the vehicle start interval. If the time interval between vehicles starting to pass the road section is 1 second, then the time required for 15 vehicles to pass can be determined as: total time = number of vehicles × (passing time of each vehicle + start interval) = 15 × (5.39 + 1) = 15 × 6.39 = 95.85 seconds. The waiting time length in the non-congested direction can be determined by statistically combining this time with the preset safety buffer time. It can be understood that the waiting time length is the time window to ensure that vehicles in the non-congested direction can pass smoothly without causing new congestion.

[0038] Then, the traffic speed in the congested direction and the current vehicle driving data of the incoming and outgoing sections corresponding to the non-congested direction in the diverted traffic area are used to determine the growth rate of the congested queue in the congested direction, so as to determine the travel time ratio in the congested direction according to the growth rate of the congested queue, the waiting time length and the spatial information of the section in the non-congested direction. If the travel time ratio is greater than the preset threshold, the backup route corresponding to the congested direction in the diverted traffic area is activated.

[0039] In this process, the effectiveness and pertinence of the diversion strategy are ensured by analyzing multiple traffic areas to be diverted and filtering based on the data of the current congested road section. By accurately calculating the relative number of vehicles waiting to enter the non-congested direction and the length of waiting time, the road section space in the non-congested direction can be used more effectively, and according to the growth rate of the congested queue, the travel time ratio in the congested direction is adjusted in time to ensure the balance of traffic flow.

[0040] S105: Input the traffic speed of each lane of the current congested road section, the current vehicle driving data of the corresponding diversion traffic area, and the diversion strategy into a preset traffic prediction model, so as to adjust the diversion strategy based on the output result to perform diversion control.

[0041] After the diversion strategy is obtained through the above steps for diversion, in order to be able to adjust the diversion strategy in real time based on the traffic conditions for diversion control. In the embodiment of this specification, the speed of each lane in the current congested section, the current vehicle driving data of the corresponding diversion traffic area and the diversion strategy are input into the preset traffic prediction model, so as to adjust the diversion strategy for diversion control according to the output results. In this process, the traffic prediction model can output the prediction results in real time according to the input data. These results reflect the traffic flow trend in the future period of time. According to the prediction results, the diversion strategy can be dynamically adjusted to adapt to changes in traffic conditions and ensure the effectiveness and flexibility of the diversion control measures.

[0042] Specifically, in one or more embodiments of the present specification, before the traffic speed of each lane of the current congested road section, the current vehicle driving data of the corresponding diversion traffic area, and the diversion strategy are input into the preset traffic prediction model, the method further includes the following process: First, collect the historical vehicle driving data corresponding to the historical diversion strategy of the current monitoring area, and divide the historical vehicle driving data corresponding to the historical diversion strategy based on the preset ratio to obtain training set data and validation set data. Input the training set data into the initial time series model for iterative training to obtain a trained time series model. Then evaluate the trained time series model based on the validation set data, and adjust the trained time series model according to the evaluation results to obtain a preset traffic prediction model. In this process, by using large-scale historical vehicle driving data as a training set, the time series model can learn the historical patterns and trends of traffic flow. The method of using training set data and validation set data to train and adjust the time series model to obtain a preset traffic prediction model improves prediction accuracy.

[0043] Furthermore, in one or more embodiments of the present specification, based on the output result, the diversion strategy is adjusted to perform diversion control, which specifically includes the following process: If the traffic flow is determined to be on a downward trend according to the output results, that is, the current diversion strategy can alleviate the congestion of the congested section, then it can be determined that the current congested section can continue to maintain the diversion strategy. If the traffic flow is determined to be on an upward trend, that is, the current diversion strategy cannot alleviate the congestion of the congested section, then the current travel time ratio corresponding to the current diversion strategy is determined. If the current travel time ratio is less than the preset threshold, then the current travel time is gradually increased, and the diversion strategy is adjusted to perform diversion control. If the current travel time ratio is greater than the preset threshold, then the spare sections corresponding to each exit section in the congested direction are determined. According to the distance between each spare section and the congested section and the traffic flow of each spare section, the activation order of each spare section is determined, so that the spare sections are activated in turn for diversion. By dynamically adjusting the diversion strategy and the activation order of the spare sections, the process can make more effective use of road resources, not only alleviate the traffic pressure on the congested section, but also improve the overall traffic efficiency and reduce the waiting time of vehicles.

[0044] like Figure 2 As shown, the present specification embodiment provides a schematic diagram of the structure of a multi-lane diversion control device. Figure 2 It can be seen that in one or more embodiments of this specification, a multi-lane diversion control device includes: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to: execute any of the above-mentioned methods.

[0045] like Figure 3 As shown, the present specification provides a schematic diagram of the structure of a non-volatile storage medium. Figure 3 It can be seen that in one or more embodiments of the present specification, a non-volatile storage medium stores computer executable instructions 301, and the computer executable instructions 301 can: execute any of the methods described above.

[0046] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0047] The above specific embodiments of the present specification are described. In some cases, the actions or steps recorded in the specification can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0048] The above is only one or more embodiments of this specification and is not intended to limit this specification. For those skilled in the art, one or more embodiments of this specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of one or more embodiments of this specification should be included in the scope of this specification.

Claims

1. A multi-lane diversion control method, characterized in that: The method comprises: Dividing the current monitoring area based on the road network related data to obtain multiple monitoring sub-areas, and determining the topological connection data corresponding to each monitoring sub-area according to the historical vehicle driving data of each monitoring sub-area; Performing cluster analysis on each of the monitoring sub-areas according to the topological connection data, determining target traffic characteristics of each of the monitoring sub-areas, and determining a diversion traffic area corresponding to each of the monitoring sub-areas based on the target traffic characteristics; Determine the current congested section based on the current vehicle driving data of each monitoring sub-area, and calculate the speed of each direction of multiple lanes in the current congested section, so as to divide and determine the congested direction and the non-congested direction; Determine a diversion traffic area corresponding to the current congested road section, and determine a corresponding diversion strategy according to the traffic speeds in the congested direction and the non-congested direction and the current vehicle driving data of the corresponding diversion traffic area; wherein the diversion strategy includes: a diversion path and a diversion time; The traffic speed of each lane of the current congested road section, the current vehicle driving data of the corresponding diversion traffic area and the diversion strategy are input into the preset traffic prediction model to adjust the diversion strategy based on the output results to perform diversion control.

2. A multi-lane diversion control method according to claim 1, characterized in that: The method of dividing the current monitoring area based on the road network related data to obtain multiple monitoring sub-areas specifically includes: Based on the multi-data interface, obtaining the road network related data of the current monitoring area; wherein the road network related data includes: basic road network data, detector detection data, and road operation data; According to the traffic network structure corresponding to the basic road network data and the preset area boundary conditions, the initial monitoring sub-area of ​​the multi-traffic level corresponding to the current monitoring area is determined; wherein the traffic levels include: intersection level, intercity expressway level, suburban expressway level, and living area road level; the preset area boundary conditions include: urban area level data, urban geographic boundary data, and monitoring area restriction data; According to the location information corresponding to the detector detection data and the operation data, matching is performed with the basic road network data to divide the detector detection data and the operation data into corresponding initial monitoring sub-areas; The internal traffic volume of each of the initial monitoring sub-areas is determined based on the detection data of the detector and the operation data, so as to dynamically adjust the scope of the initial monitoring sub-area according to the internal traffic volume to obtain multiple monitoring sub-areas.

3. A multi-lane diversion control method according to claim 1, characterized in that: Determining the topological connection data corresponding to each of the monitoring sub-areas according to the historical vehicle driving data of each of the monitoring sub-areas specifically includes: Acquire historical vehicle driving data of each monitoring sub-area, and identify the entry and exit sections of each monitoring sub-area based on the driving direction data corresponding to each vehicle driving data and the boundary position of each monitoring sub-area; The entry road section and the exit road section are respectively used as traversal start nodes, so as to traverse each road section of the monitoring sub-area based on the traversal start node and the preset traversal direction, obtain an entry road section sequence and an exit road section sequence, and use each road section in the entry road section sequence and the exit road section sequence as a node; wherein the traversal direction is: clockwise or counterclockwise; Determine the adjacent road sections of each of the nodes, and determine the road section type of each of the nodes based on the basic road network data corresponding to the adjacent road sections; wherein the road section type includes: divergence type and confluence type; Determine the vehicle driving frequency and the number of vehicles traveling between the nodes according to the historical vehicle driving data of each node, and determine the connectivity strength of each node according to the vehicle driving frequency, the number of vehicles traveling and the road section type; Based on the connectivity strength corresponding to each of the nodes and the sequence name corresponding to the node, a connectivity path between each of the nodes is determined to obtain topological connection data corresponding to each of the monitoring sub-areas.

4. A multi-lane diversion control method according to claim 3, characterized in that: Performing cluster analysis on each of the monitoring sub-areas according to the topological connection data to determine the target traffic characteristics of each of the monitoring sub-areas specifically includes: Determine the adjacent nodes of each node in the monitoring sub-area based on the topological connection data to construct a neighbor node set corresponding to each node; Sequentially acquiring the neighboring nodes corresponding to each neighboring node in the neighboring node set, so as to determine the common neighboring nodes corresponding to the node and the neighboring nodes of the node based on the intersection of the neighboring node set and the neighboring nodes corresponding to each neighboring node; Constructing a feature vector corresponding to each of the nodes based on the historical vehicle travel data corresponding to the adjacent nodes of each of the nodes and the common neighbor nodes; Perform similarity clustering based on the feature vectors corresponding to each of the nodes to obtain multiple clusters, and extract common features of the nodes in each cluster; The common features of the clusters are summarized and the position distribution corresponding to the clusters is determined, so as to determine the target traffic features of each monitoring sub-area by combining the common features of the clusters and the corresponding position distribution.

5. A multi-lane diversion control method according to claim 1, characterized in that: Determining the diversion traffic area corresponding to each of the monitoring sub-areas based on the target traffic characteristics specifically includes: Based on the target traffic characteristics of each monitoring sub-area, identifying the traffic bottleneck data of each monitoring sub-area; wherein the traffic bottleneck data includes: traffic peak time period, congestion node range, and road capacity; Determine the diversion capacity data corresponding to each of the monitoring sub-areas according to the traffic bottleneck data corresponding to each of the monitoring sub-areas; wherein the diversion capacity data includes: diversion time and diversion capacity; Determine, based on the diversion capacity data corresponding to the adjacent monitoring sub-areas of the monitoring sub-area, an optional diversion traffic area corresponding to the diversion time and the traffic peak period; The diversion capacity value of each of the optional diversion traffic areas is determined according to the diversion capacity pair of each of the optional diversion traffic areas, so as to determine the diversion traffic area corresponding to each of the monitoring sub-areas based on the diversion capacity value.

6. A multi-lane diversion control method according to claim 5, characterized in that: Determining a diversion traffic area corresponding to the current congested road section, and determining a corresponding diversion strategy according to the traffic speeds in the congested direction and the non-congested direction and the current vehicle driving data of the corresponding diversion traffic area, specifically includes: Determine a plurality of traffic areas to be diverted according to the diverted traffic area corresponding to the monitoring sub-area where the current congested road section is located, and filter the plurality of traffic areas to be diverted according to the incoming road section or the outgoing road section within the preset range of the current congested road section to obtain the diverted traffic area corresponding to the current congested road section; Determine the number of vehicles waiting to enter the non-congested direction based on current vehicle travel data of the entry section and the exit section corresponding to the non-congested direction in the diverted traffic area; Determine the waiting time length of the non-congested direction according to the passing speed of the non-congested direction and the data of the relatively waiting vehicles; Determine the growth rate of the congestion queue in the congestion direction based on the traffic speed in the congestion direction and the current vehicle driving data of the incoming road section and the outgoing road section corresponding to the non-congestion direction in the diverted traffic area; Determine the travel time ratio of the congested direction according to the growth rate of the congested queue, the length of the waiting time and the spatial information of the road section in the non-congested direction; If the travel time ratio is greater than a preset threshold, an alternate route corresponding to the congestion direction in the diversion traffic area is activated.

7. A multi-lane diversion control method according to claim 1, characterized in that: Before inputting the traffic speed of each lane of the current congested road section, the current vehicle driving data of the corresponding diversion traffic area and the diversion strategy into the preset traffic flow prediction model, the method further includes: Collecting historical vehicle driving data corresponding to the historical diversion strategy of the current monitoring area, and dividing the historical vehicle driving data corresponding to the historical diversion strategy based on a preset ratio to obtain training set data and verification set data; Inputting the training set data into the initial time series model for iterative training to obtain a trained time series model; The trained time series model is evaluated based on the validation set data, so as to adjust the trained time series model according to the evaluation result to obtain a preset traffic prediction model.

8. A multi-lane diversion control method according to claim 6, characterized in that: Based on the output result, the diversion strategy is adjusted to perform diversion control, specifically including: If it is determined based on the output result that the traffic flow is on a downward trend, determining that the current congested road section maintains the diversion strategy; If it is determined that the traffic flow is on an upward trend, then determining the current travel time ratio corresponding to the current diversion strategy; If the current travel time ratio is less than the preset threshold, the current travel time is gradually increased, and the diversion strategy is adjusted to perform diversion control; If the current travel time ratio is greater than the preset threshold, determining the backup sections corresponding to each exit section in the congested direction; Based on the distance between each of the backup road sections and the congested road section and the traffic flow of each of the backup road sections, the activation order of each of the backup road sections is determined, so as to activate the backup road sections in sequence for traffic diversion.

9. A multi-lane diversion control device, characterized in that: The device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can: execute any of the methods described in claims 1-8.

10. A non-volatile storage medium storing computer executable instructions, characterized in that: The computer executable instructions can: execute the method described in any one of claims 1 to 8.

Images