CN117253364B - Traffic jam event extraction and situation fusion method and system - Google Patents
Traffic jam event extraction and situation fusion method and system Download PDFInfo
- Publication number
- CN117253364B CN117253364B CN202311516071.2A CN202311516071A CN117253364B CN 117253364 B CN117253364 B CN 117253364B CN 202311516071 A CN202311516071 A CN 202311516071A CN 117253364 B CN117253364 B CN 117253364B
- Authority
- CN
- China
- Prior art keywords
- congestion
- time
- event
- subgraph
- mileage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000605 extraction Methods 0.000 title claims abstract description 21
- 238000007500 overflow downdraw method Methods 0.000 title claims description 7
- 238000000034 method Methods 0.000 claims abstract description 26
- 238000013507 mapping Methods 0.000 claims abstract description 13
- 230000004927 fusion Effects 0.000 claims description 13
- 238000010276 construction Methods 0.000 claims description 8
- 238000010845 search algorithm Methods 0.000 claims description 3
- 208000027744 congestion Diseases 0.000 description 194
- 238000004458 analytical method Methods 0.000 description 6
- 239000000284 extract Substances 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 238000013467 fragmentation Methods 0.000 description 2
- 238000006062 fragmentation reaction Methods 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 101100377706 Escherichia phage T5 A2.2 gene Proteins 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to a method and a system for extracting traffic jam events and fusing situation, which are used for extracting jam features from jam event data from different sources and constructing a set of jam events; constructing a road network topological graph by taking the mileage piles as nodes and the adjacency relations of the mileage piles as arcs, and mapping longitude and latitude coordinates of the congestion position to the mileage piles; constructing topology subgraphs time by time based on a mileage pile topology network by taking a time slice as a unit, wherein each subgraph represents a certain local congestion state of a road network; merging the subgraphs according to node intersection conditions according to the time slicing sequence, so that each subgraph represents a certain partial one-time complete congestion event of the road network; slicing each event subgraph according to a preset time window to obtain situation time sequence data of the congestion event. The technical scheme of the invention can be used for carrying out real-time modeling and extraction on traffic jam events of the expressway network in real time, locating the jammed road section and the jam degree at the time more accurately, and grasping the time change rule of the jam.
Description
Technical Field
The invention belongs to the field of intelligent traffic informatization, and relates to a traffic jam event extraction and situation fusion algorithm and system based on highway network topological graph calculation.
Background
Along with the continuous acceleration of the urban process, road traffic is increasingly congested, and the traffic congestion becomes one of the biggest problems of urban traffic operation. At present, most internet map service providers realize the perception of congestion events in a road network by providing floating GPS information acquired during travel services such as navigation for drivers and passengers, and acquire the congestion state of a certain road section at a certain time point through aggregation analysis, so that congestion related data services are formed again.
However, due to the limitation of the configuration quantity of the floating cars, the observation range and the precision thereof, the road network congestion state cannot be observed completely, so that the reported congestion event data lacks space-time continuity; meanwhile, the workload of manually extracting, processing and analyzing the congestion data is huge in the face of huge data in the national road network range, and a tool is needed to assist in processing and analyzing; in addition, each service provider provides inconsistent caliber defined for the congestion event, and the aim of fusing multi-party congestion data to improve the observation precision is difficult to achieve. Therefore, the current analysis work of the congestion status of the highway network generally carries out event extraction analysis by a GIS visual congestion thermodynamic diagram mode, but the method can only find the static coverage range of the congestion event and cannot describe the congestion event from the dynamic angle of time-space.
Disclosure of Invention
The invention aims to overcome the problem that a large amount of high-speed traffic congestion data lacks space-time continuity and the problem of fusion complexity among multi-source data, and provides a traffic congestion event extraction and situation fusion method and system by utilizing a highway network topology graph modeling method so as to output independent and complete congestion events.
In order to achieve the technical purpose, the invention adopts the following technical scheme:
a traffic jam event extraction and situation fusion method comprises the following steps:
s1, extracting congestion characteristics from congestion event data from different sources, and constructing a set of congestion events; the congestion characteristics include congestion location, congestion time, and congestion level;
s2, constructing a road network topological graph by taking the mileage piles as nodes and the adjacency relations of the mileage piles as arcs, mapping longitude and latitude coordinates of the congestion position to the mileage piles, and constructing a set of congestion events attached to the road network of the mileage piles;
s3, constructing topological subgraphs time by time based on the set constructed in the S2, determining nodes forming a subgraph based on connectivity of the nodes for mileage stake nodes in each time slice, adding the nodes and arcs in the subgraph according to the node sequence, and giving weight attributes to the arcs based on the congestion degree level, wherein each subgraph represents a certain local congestion state of a road network;
s4, merging the subgraphs according to the node intersection condition according to the time slicing sequence, so that each subgraph represents a certain partial one-time complete congestion event of the road network;
s5, slicing each event subgraph according to a preset time window to obtain situation time sequence data of the congestion event.
As a preferred embodiment, in S1, the congestion location is spatial route information provided by a linetring object, and when mapping, the congestion location is mapped to a mileage pile based on a nearest neighbor search algorithm.
In a preferred embodiment, in S2, when the road network topology map is constructed, a pile number of the mileage pile plus the downlink identifier is used as a node number.
In a preferred embodiment, in the step S3, when the topology subgraph is constructed, for the case that the same congestion event subgraph is not connected, the route of the milepost node planned with the shortest route is returned to the non-connected milepost node.
In the step S3, when the topology subgraph is constructed, a trace-back identifier is set for the subgraph, and is used for tracing back congestion data of each server constituting the subgraph through a mapping relationship.
As a preferred embodiment, in the step S4, a first time window is set, and the sub-graph reported by the congestion-free data exceeding the first time window is regarded as an event sub-graph representing a complete congestion event; and merging the subgraphs with the non-congestion data reporting time not exceeding the first time window into a new subgraph according to the node intersection condition until the non-congestion data reporting time of the subgraph exceeds the first time window.
As a preferred embodiment, the step S5 further includes, in slice data, if an arc is assigned to different congestion degree levels by different source servers, weighting and fusing the congestion degree level values.
As a preferred embodiment, the congestion duration and the congestion frequency duty ratio are acquired according to the congestion data provided by the source service provider of the congestion level data, and the weight of each congestion level is distributed based on the congestion duration and the congestion frequency duty ratio;
the congestion duration refers to the time duty ratio corresponding to each congestion degree grade event;
and the congestion frequency duty ratio refers to the frequency duty ratio corresponding to the congestion degree level event, and is calculated according to the record number reported by the data.
As a preferred embodiment, the congestion duration and the congestion frequency are weighted to obtain the weight of the congestion level, and the weight of the congestion duration is higher than the weight of the congestion frequency.
Another object of the present invention is to provide a traffic congestion event extraction and situation fusion system, including:
the congestion feature extraction module is used for extracting congestion features from congestion event data from different sources and constructing a set of congestion events; the congestion characteristics include congestion location, congestion time, and congestion level;
the road network construction module of the mileage pile uses mileage piles as nodes, the adjacency relationship of the mileage piles is used as arcs to construct a road network topological graph, longitude and latitude coordinates of the congestion position are mapped to the mileage piles, and a set of congestion events after the mileage pile road network is attached is constructed;
the sub-graph creation module is used for constructing a topological sub-graph time by taking time slices as units, determining nodes forming one sub-graph based on connectivity of the nodes for mileage stake nodes in each time slice, adding the nodes and arcs in the sub-graph according to the node sequence, and giving weight attributes to the arcs based on the congestion degree level, so that each sub-graph represents a certain local congestion state of a road network; merging the subgraphs according to node intersection conditions according to the time slicing sequence, so that each subgraph represents a certain partial one-time complete congestion event of the road network;
and the situation time sequence data output module is used for slicing each event subgraph according to a preset time window to obtain situation time sequence data of the congestion event.
As a preferred embodiment, the situation time sequence data output module includes a data fusion unit, and if an arc of slice data is given different congestion degree grades by different source servers, the congestion degree grade values are weighted and fused.
As a preferred embodiment, acquiring a congestion duration and a congestion frequency duty ratio according to congestion data provided by a source service provider of the congestion level grade data, and assigning weights based on the congestion duration and the congestion frequency duty ratio;
the congestion duration refers to the time duty ratio corresponding to each congestion degree grade event;
and the congestion frequency duty ratio refers to the frequency duty ratio corresponding to the congestion degree level event, and is calculated according to the record number reported by the data.
As a preferred embodiment, the congestion duration and the congestion frequency are weighted to obtain the weight of the congestion level, and the weight of the congestion duration is higher than the weight of the congestion frequency.
The invention takes graph network modeling of traffic jam events as a core solution idea, namely mapping the high-speed traffic jam events of each time segment into a high-speed mileage stake interval, modeling the graph network, continuously adding new jam events into the graph network (adding edges) by setting a time window, and removing edges meeting certain conditions (once complete jam events), thereby realizing real-time extraction and fusion of the jam events.
The invention has the following characteristics: 1) Providing key feature definition of congestion data; 2) The network dynamic modeling of the large-range graph can be performed in real time according to the discrete data of the congestion event; 3) Supporting multi-source traffic jam event data fusion; 4) And supporting situation fusion of traffic jam events.
The method and the system can establish a unified traffic congestion monitoring system by integrating high-speed congestion event data sources of multiple parties so as to overcome the limitation of a single data source. The scheme of the invention can model and extract traffic jam events of the expressway network in real time, more accurately position the jammed road section and the jam degree at the time, and grasp the time change rule of the jam. The method can extract the high-accuracy complete congestion event result and analyze the congestion event, can find the root of congestion formation based on the traffic situation analysis of the fusion data, and provides targeted traffic management countermeasures. The technical scheme fills the gap of data fragmentation in the current traffic management field, is beneficial to the cooperation of all departments to deal with urban traffic jams, improves the management level and the management capability, and provides better quality service for the traffic travel public.
Detailed Description
The method of the invention receives a plurality of high-speed traffic jam event records and outputs differentiated traffic jam events, and comprises three steps: a1, extracting congestion event characteristics and constructing a road network; a2, road network attachment of a congestion event and event subgraph construction; a3 Congestion event situation fusion method specifically comprises the following steps:
a1 Congestion event feature extraction and road network construction process
The method comprises the following steps: 1) Normalized extraction of congestion events is convenient for carrying out event alignment and fusion on traffic congestion data of various sources; 2) Road network is constructed based on the topological structure of the highway network mileage pile, so that the congestion event in the subsequent step is conveniently attached to the road network.
A1.1 Congestion event feature extraction
Because the accuracy and frequency of road condition acquisition equipment, the distribution position of the equipment, detection objects and the like are different, congestion event data from different sources can show a certain difference in the result. For example, the congestion data provided by the flight is more in hundred degrees and the coverage area is wider, but the content of the congestion data field is less (the downlink direction of the road number and the congestion position is lacking), and the hundred-degree congestion data can provide enhanced fields such as the congestion distance, the duration and the like besides the necessary basic information.
Under the condition of different data quality, in order to ensure the feasibility of fusion of the multi-source congestion events, the characteristic extraction of the congestion events is needed. Considering the application of the present post-method, congestion events can be extracted as several key features:
congestion location: position information in a LineStng format, which represents a route range covered by congestion, and coordinates are unified by using WGS84 coordinates;
time: the congestion event occurrence time monitored by the equipment is accurate to the minute;
congestion conditions: the congestion level is indicated, 1 indicates clear, 2 indicates slow travel, 3 indicates congestion, and 4 indicates severe congestion.
Based on the characteristics, a congestion event extraction algorithm can be constructed, and the rest of enhancement data can be used as refinement data in the congestion event analysis process, and the specific steps can be seen in step A3.
A1.2 Highway network construction method
According to different observation objects, road networks have various construction forms, and the road network directed graph is constructed by using topological structure relations based on mileage piles. The mileage piles are used as road network nodes because the mileage piles have higher density (the mileage piles are spaced by 100 meters and the distance between the portal frames is different from hundreds to thousands of meters) compared with portal facilities, and meanwhile, the mileage piles are used as the road network building nodes when judging the congestion event, which are not required to reach the meter level or the ten meter level, and are used as road infrastructure, and are more close to the service demands compared with the geohash gridding mode and other processing methods.
The mileage stake marks are formed by an absolute position and a relative position together, for example, G310K 24+600 is expressed on the highway 310, and the highway is positioned 24.6 km away from the highway starting point with the direction of the highway starting point as positive. In addition, note that the mileage stake marks on both sides of the road at the same position are consistent, so as to be able to be distinguished into different nodes when constructing a topology network, and additionally add uplink and downlink identifications. The scheme defines that the number sequence of the mileage piles is increased to be in the uplink direction (the code is 1), and conversely, the number sequence of the mileage piles is increased to be in the downlink direction (the code is 0). For example: g310 At K24+600, the uplink mileage stake node is numbered as G310 K24+600-1, and the downlink mileage stake node is numbered as G310 K24+600-0.
Based on the processed mileage stake node numbers, constructing a road network topological graph, wherein the road network topological graph is defined as follows:
: directed graph G is a binary set of ordered pairs of some elements in a non-empty finite set V and E;
for the node set of graph G, each element in VIs one of the graphs GNodes, i.e. each node represents a mileage stake;
for the arc set of graph G, each element in EIs marked asOne of the graphs GTo the point ofThe arc set is constructed based on the adjacent relation of the prior mileage piles, namely, the arc set only comprises the topological relation between every two adjacent mileage piles;
is a mapping, and any two mileage stake nodes are givenAndand returning to the mileage stake node path planned by the shortest path, and sequencing the mileage stake node paths in sequence.
A2 Road network attachment of congestion events and event subgraph construction
The congestion event location information is spatial route information provided by a LineStng object, and in order to meet modeling requirements of the congestion event, the location of the congestion event location information is firstly required to be mapped onto a mileage pile topology network. On the basis, with continuous updating of the congestion event data, superposition of the congestion events is realized. Dividing this step into a 2.1) road network attachment of congestion events; a2.2 A congestion event topology subgraph is generated and two works are carried out.
A2.1 Road network attachment for congestion events
The LineStng object contains a series of road (longitude and latitude) coordinate points, and in the process of road network attachment, the invention adopts a nearest neighbor search algorithm to finish the mapping from line data to a network map. The algorithm has general processing means and is not the focus of the present invention, and is not described in detail here. The attach procedure for congestion events is defined as follows:
: t represents the set of all congestion events,expressing the r-th congestion data reported by the s service provider at the t moment;
: τ represents a piece of congestion event data reported, and the contents in each bracket are road condition data provided by s service provider (e.g.) Where d represents the level of congestion level,representing a corresponding longitude and latitude coordinate position, and sequencing according to a route sequence;
is a mapping, and the longitude and latitude coordinates are mapped to corresponding mileage piles through a nearest neighbor algorithm.
Traversing τ byMapping longitude and latitude coordinates to a mileage pile to obtain:
whereinIs attached with longitude and latitude coordinatesIs a mileage stake node;
is congestion data after attachment to the milepost.
The following work was carried out based on the results after the attachment of the mileage stake.
A2.2 Event subgraph construction and event extraction
The above steps are extracting an event of a time segment in the congestion event, and to construct a complete primary congestion event, congestion data under continuous time segments needs to be processed. Because the event occurrence position is mostly only on a section of road on the expressway network, if the whole road network topology is constructed, a lot of extra computing resources are consumed when the whole network congestion event is identified. Therefore, the invention extracts events by constructing topology subgraphs (namely road network local structures) time by time based on the mileage pile topology network (the congestion data set after the mileage piles are attached). The method comprises the following steps:
s1: acquisition time slice t 1 The congestion data (in minutes, less than minutes, in decimal places) is provided that the current congestion data provider is a and b, and then the congestion slice data are respectively:
s2: because the mileage pile is attached by the nearest neighbor algorithm, there is a situation that two longitude and latitude positioning points before and after the mileage pile are attached to the same mileage pile, and when constructing the subgraph, there may be a situation that the subgraphs of the same congestion event are not communicated. This step takes this into account, forAnd (3) withIs performed by the node of (1)Mapping to obtain the complemented congestion slice dataAnd (3) with；
S3: adding nodes and arcs to the mileage pile node v in each time slice according to the node sequence to obtain a graph，Is thatAnd (3) withIs formed by the mileage stake nodes of the utility model,and the congestion degree grade forms the weight between every two adjacent mileage pile nodes, namely, the weight of the arc is assigned with 1, 2, 3 or 4 according to the congestion degree grade.
And when a new congestion mileage pile node which is not communicated with the mileage pile node in the subgraph appears, the congestion mileage pile node represents a new congestion event, namely, the congestion mileage pile node belongs to the new subgraph.
The congestion degree observed by different service providers on the same road section may not be consistent, i.e. one arc may contain a plurality of weight values, and then index fusion is performed through weighting calculation, so that the problem is solved.
S4: after the steps S1 to S3 are completed, the graphThere will be multiple sub-graphs, each representing a different congestion event occurring on the full road network, each sub-graph being distinguished and marked asWhereinIs a sub-graph of each of the sub-graphs,representing compositionCongestion data of each service provider of the subgraph can find corresponding congestion data through the mapping relation，t s Is the start time of subgraph creation, t e Is the end time of the congestion event, defaults to one minute later (i.e., t) when the subgraph is first created e =t s +1 because the service provider data reporting time interval is one to two minutes).
Considering that the timeliness of the reported data obtained in practical application is in the order of minutes, and meanwhile, the time resolution is too large and different congestion events can not be distinguished, a reasonable time window W is required to be set, if no congestion report exists in the exceeding time window, the congestion event is considered to be ended, namely, a complete certain sub-graph is considered to be extracted, and the invention is set to W=10 (minutes) by default.
S5: after each time slice extracts the subgraph, judging the ending time of the congestion event of each subgraph, if the congestion event is not reported for more than 10 minutes, namely, no new subgraph is added for 10 minutes continuously, then considering that one congestion event is ended, and archiving the subgraph into an R graph set, wherein the R graph set is expressed asAt the same time shift out the sub-graphThe method comprises the steps of carrying out a first treatment on the surface of the The set in the R graph is the set of subgraphs representing a complete congestion event; for subgraphs with no congestion reporting time not exceeding 10 minutes, the congestion event is consideredStill continuing, retaining the subgraph;
s6: for each time slice (at the start of slice time t s To be accurate) inSubgraph withinJudging whether the nodes of the subgraph have intersections, if so, merging the intersections into one subgraph to update the subgraph, and using t s Minimum starting time, t e The maximum is the end time. Such as three sub-graphsCan be combined into sub-graphHere, whereI in (a) represents the number of the sub-graph, j represents the time slice, i.e. the sub-graph of which the j-th time slice starts to be constructed.
And (3) repeating the steps S1 to S6, and continuously adding, updating and moving out the subgraph to complete the complete extraction of each congestion event.
A3 Congestion event situation fusion method
The step A2 is extraction of time and place from beginning to end of a congestion event, and the emphasis is on the integrity of the congestion event, so that the problems of discretization and fragmentation of congestion event data are solved. However, the above-described work results can only support related traffic analysis such as the number of congestion times, the congestion range, and the like. The method can provide fine guidance and early warning for management and control of traffic police, road companies and the like when a certain area of the road network starts to be gradually congested, when the highest congestion level is reached, how the congestion in the area spreads and conducts along with time and when the congestion dissipates.
The method comprises the following steps of:
s1: considering the fluctuation conduction range of the congestion event, a sliding time window w=5 # -is setMinute), from t s Initially, slicing the congestion event time period according to w;
s2: acquiring a sub-graph in RSlicing according to the time window to obtain time slices；
S3: sequentially taking time slices and obtainingIn a time range ofSubgraph of (a)And pass throughRetrospective tracingIs recorded as the congestion data of；
S4: due toIn the method, the congestion degrees observed by different service providers of every two adjacent mileage piles are possibly inconsistent, and the congestion degrees are fused to obtain the fused congestion degree grade as follows:
wherein d c Represents the level, theta, of congestion observed by different service providers c The weight assigned to the congestion degree level is assigned according to the congestion duration and the congestion frequency duty ratio, and is defined as follows:
each congestion level eventThe corresponding time duty cycle:，the accumulated time is counted according to the congestion degree level;
the frequency duty cycle corresponding to each congestion level event:，the number of topological edges is counted according to the congestion degree level, namely the record number of data reporting;
for example, for every two adjacent mileage piles, the congestion degree level observed by the service provider 1 is 3, the congestion degree level observed by the service provider 2 is 4, the congestion duration and the congestion frequency of the congestion degree levels 3 and 4 are calculated based on the congestion data, and finally the weight theta of the congestion degree level 3 is obtained 3 And a weight θ of congestion degree level 4 4 The congestion degree grade after fusion is。
The frequency of the congestion event is easily influenced by the reported data frequency of each service provider and the equipment condition, while the duration of the congestion event is relatively more robust, so that the two factors are weighted by considering the ratio of 7:3, and finally the weighting parameters are obtained：
Repeating the steps S1-S4 to finally obtain situation time sequence data of one congestion event, namely. Based on this, congestion events can be implementedIs a trace back and deduction of the above.
Claims (10)
1. The traffic jam event extraction and situation fusion method is characterized by comprising the following steps of:
s1, extracting congestion characteristics from congestion event data from different sources, and constructing a set of congestion events; the congestion characteristics include congestion location, congestion time, and congestion level;
s2, constructing a road network topological graph by taking the mileage piles as nodes and the adjacency relations of the mileage piles as arcs, mapping longitude and latitude coordinates of the congestion position to the mileage piles, and constructing a set of congestion events attached to the road network of the mileage piles;
s3, constructing topological subgraphs time by time based on the set constructed in the S2, determining nodes forming a subgraph based on connectivity of the nodes for mileage stake nodes in each time slice, adding the nodes and arcs in the subgraph according to the node sequence, and giving weight attributes to the arcs based on the congestion degree level, wherein each subgraph represents a certain local congestion state of a road network;
s4, merging the subgraphs according to the node intersection condition according to the time slicing sequence, so that each subgraph represents a certain partial one-time complete congestion event of the road network;
s5, slicing each event subgraph according to a preset time window to obtain situation time sequence data of the congestion event.
2. The method of claim 1, wherein in S1, the congestion location is spatial route information provided in a linetrig object, and the congestion location is mapped to a milepost based on a nearest neighbor search algorithm when mapped.
3. The method according to claim 1, wherein in S2, when constructing the road network topology map, a pile number of the mileage pile plus a downlink identifier is used as a node number.
4. The method according to claim 1, wherein in S3, when constructing the topology subgraph, for the case that the same congestion event subgraph is not connected, the mileage stake node path planned with the shortest path is returned to the mileage stake node that is not connected.
5. The method according to claim 1, wherein in S3, when constructing the topology subgraph, a trace-back identifier is set for the subgraph, and is used for tracing back congestion data of each service provider constituting the subgraph through the mapping relationship.
6. The method according to claim 1, wherein in S4, a first time window is set, and the sub-graph reported by the congestion-free data exceeding the first time window is regarded as an event sub-graph representing a complete congestion event; and merging the subgraphs with the non-congestion data reporting time not exceeding the first time window into a new subgraph according to the node intersection condition until the non-congestion data reporting time of the subgraph exceeds the first time window.
7. The method according to claim 1 or 5, wherein S5 further comprises weighting and fusing the congestion level values in slice data if an arc is assigned different congestion level by different source servers.
8. The method of claim 7, wherein the congestion duration and the congestion frequency duty cycle are obtained from congestion data provided by a source server of congestion level data, and wherein the weight of each congestion level is assigned based on the congestion duration and the congestion frequency duty cycle;
the congestion duration refers to the time duty ratio corresponding to each congestion degree grade event;
and the congestion frequency duty ratio refers to the frequency duty ratio corresponding to the congestion degree level event, and is calculated according to the record number reported by the data.
9. The method of claim 8 wherein the weight of the congestion duration and the congestion frequency are weighted to obtain a weight of the congestion level, the weight of the congestion duration being higher than the weight of the congestion frequency.
10. A traffic congestion event extraction and situation fusion system, comprising:
the congestion feature extraction module is used for extracting congestion features from congestion event data from different sources and constructing a set of congestion events; the congestion characteristics include congestion location, congestion time, and congestion level;
the road network construction module of the mileage pile uses mileage piles as nodes, the adjacency relationship of the mileage piles is used as arcs to construct a road network topological graph, longitude and latitude coordinates of the congestion position are mapped to the mileage piles, and a set of congestion events after the mileage pile road network is attached is constructed;
the sub-graph creation module is used for constructing a topological sub-graph time by taking time slices as units, determining nodes forming one sub-graph based on connectivity of the nodes for mileage stake nodes in each time slice, adding the nodes and arcs in the sub-graph according to the node sequence, and giving weight attributes to the arcs based on the congestion degree level, so that each sub-graph represents a certain local congestion state of a road network; merging the subgraphs according to node intersection conditions according to the time slicing sequence, so that each subgraph represents a certain partial one-time complete congestion event of the road network;
and the situation time sequence data output module is used for slicing each event subgraph according to a preset time window to obtain situation time sequence data of the congestion event.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311516071.2A CN117253364B (en) | 2023-11-15 | 2023-11-15 | Traffic jam event extraction and situation fusion method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311516071.2A CN117253364B (en) | 2023-11-15 | 2023-11-15 | Traffic jam event extraction and situation fusion method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117253364A CN117253364A (en) | 2023-12-19 |
| CN117253364B true CN117253364B (en) | 2024-01-26 |
Family
ID=89137194
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311516071.2A Active CN117253364B (en) | 2023-11-15 | 2023-11-15 | Traffic jam event extraction and situation fusion method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117253364B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104809879A (en) * | 2015-05-14 | 2015-07-29 | 重庆大学 | Expressway road traffic state estimation method based on dynamic Bayesian network |
| CN108765939A (en) * | 2018-05-11 | 2018-11-06 | 贵阳信息技术研究院（中科院软件所贵阳分部） | Dynamic traffic jam index calculation method based on clustering algorithm |
| CN108898833A (en) * | 2018-07-11 | 2018-11-27 | 贵州交通信息与应急指挥中心 | A kind of high speed congestion space-time analysis method |
| CN110675629A (en) * | 2019-10-08 | 2020-01-10 | 苏交科集团股份有限公司 | Big data-based highway congestion prediction and active prevention and control method |
| CN111540199A (en) * | 2020-04-21 | 2020-08-14 | 浙江省交通规划设计研究院有限公司 | High-speed traffic flow prediction method based on multi-mode fusion and graph attention machine mechanism |
| CN115457764A (en) * | 2022-08-24 | 2022-12-09 | 华南理工大学 | Road section traffic density estimation method, device and medium based on vehicle track data |
| CN115547084A (en) * | 2021-06-30 | 2022-12-30 | 阿里巴巴新加坡控股有限公司 | Road traffic condition prediction method and device, electronic equipment and computer program product |
| CN115641721A (en) * | 2022-12-06 | 2023-01-24 | 南京感动科技有限公司 | Multi-source traffic flow index fusion and road condition calculation method based on fusion index |
| CN115880909A (en) * | 2022-12-15 | 2023-03-31 | 北京电信规划设计院有限公司 | Highway congestion identification method based on floating car data |
| CN116596380A (en) * | 2023-05-16 | 2023-08-15 | 浙江高信技术股份有限公司 | Optimization determination method, platform, equipment and medium for expressway construction organization scheme and management and control scheme |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070002866A1 (en) * | 2005-06-29 | 2007-01-04 | Denso Corporation | Receive power priority flooding in mobile ad hoc networks |
| US10922965B2 (en) * | 2018-03-07 | 2021-02-16 | Here Global B.V. | Method, apparatus, and system for detecting a merge lane traffic jam |
-
2023
- 2023-11-15 CN CN202311516071.2A patent/CN117253364B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104809879A (en) * | 2015-05-14 | 2015-07-29 | 重庆大学 | Expressway road traffic state estimation method based on dynamic Bayesian network |
| CN108765939A (en) * | 2018-05-11 | 2018-11-06 | 贵阳信息技术研究院（中科院软件所贵阳分部） | Dynamic traffic jam index calculation method based on clustering algorithm |
| CN108898833A (en) * | 2018-07-11 | 2018-11-27 | 贵州交通信息与应急指挥中心 | A kind of high speed congestion space-time analysis method |
| CN110675629A (en) * | 2019-10-08 | 2020-01-10 | 苏交科集团股份有限公司 | Big data-based highway congestion prediction and active prevention and control method |
| CN111540199A (en) * | 2020-04-21 | 2020-08-14 | 浙江省交通规划设计研究院有限公司 | High-speed traffic flow prediction method based on multi-mode fusion and graph attention machine mechanism |
| CN115547084A (en) * | 2021-06-30 | 2022-12-30 | 阿里巴巴新加坡控股有限公司 | Road traffic condition prediction method and device, electronic equipment and computer program product |
| CN115457764A (en) * | 2022-08-24 | 2022-12-09 | 华南理工大学 | Road section traffic density estimation method, device and medium based on vehicle track data |
| CN115641721A (en) * | 2022-12-06 | 2023-01-24 | 南京感动科技有限公司 | Multi-source traffic flow index fusion and road condition calculation method based on fusion index |
| CN115880909A (en) * | 2022-12-15 | 2023-03-31 | 北京电信规划设计院有限公司 | Highway congestion identification method based on floating car data |
| CN116596380A (en) * | 2023-05-16 | 2023-08-15 | 浙江高信技术股份有限公司 | Optimization determination method, platform, equipment and medium for expressway construction organization scheme and management and control scheme |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117253364A (en) | 2023-12-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110111575B (en) | Urban traffic flow network analysis method based on complex network theory | |
| US10984652B2 (en) | Method and system for modeling and processing vehicular traffic data and information and applying thereof | |
| CN110796337B (en) | System for evaluating service accessibility of urban bus stop | |
| CN109697221A (en) | Method for digging, device, storage medium and the electronic equipment of track rule | |
| JP2018538627A (en) | Method and apparatus for obtaining route heat of a traffic road | |
| WO2022083166A1 (en) | Method and system for reconstructing vehicle's driving trajectory when checkpoint data is missing | |
| Kong et al. | RMGen: A tri-layer vehicular trajectory data generation model exploring urban region division and mobility pattern | |
| CN112991743B (en) | Real-time traffic risk AI prediction method based on driving path and system thereof | |
| CN114298493A (en) | Road operation monitoring system, method, terminal and storage medium | |
| CN117253364B (en) | Traffic jam event extraction and situation fusion method and system | |
| Liu et al. | Determination of routing velocity with GPS floating car data and webGIS-based instantaneous traffic information dissemination | |
| CN112288591A (en) | Three-dimensional management and control platform system based on laser point cloud map | |
| Habtie et al. | Cellular network based real-time urban road traffic state estimation framework using neural network model estimation | |
| CN109166310A (en) | Road trip time estimation method based on LBS and conventional traffic road condition data | |
| CN113284030B (en) | Urban traffic network community division method | |
| Padrón et al. | Realistic traffic model for urban environments based on induction loop data | |
| CN111640303A (en) | City commuting path identification method and equipment | |
| Arregui et al. | Impact of the road network configuration on map‐matching algorithms for FCD in urban environments | |
| Celuch et al. | Data Fusion for Public Transport Traffic Data Analysis: A Case Study | |
| Joubert et al. | Development of a map-matching algorithm for dynamic-sampling-rate GPS signals to determine vehicle routes on a MATSim network | |
| CN113190769B (en) | Communication characteristic data determining method, device, electronic equipment and storage medium | |
| CN113990064B (en) | Traffic data acquisition method, system and computer readable storage medium | |
| CN114626964B (en) | New energy automobile monitoring information cross-region sharing method | |
| Rahman et al. | Understanding real time traffic characteristics of urban zones using GPS data: A computational study on Dhaka City | |
| Liu et al. | Analysis and visualization of traffic conditions of road network by route bus probe data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |