CN112614349B - Traffic condition prediction method and system based on big data - Google Patents
Traffic condition prediction method and system based on big data Download PDFInfo
- Publication number
- CN112614349B CN112614349B CN202011625743.XA CN202011625743A CN112614349B CN 112614349 B CN112614349 B CN 112614349B CN 202011625743 A CN202011625743 A CN 202011625743A CN 112614349 B CN112614349 B CN 112614349B
- Authority
- CN
- China
- Prior art keywords
- data
- road
- prediction
- prediction result
- time
- 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
Images
Classifications
-
- 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
- G08G1/0125—Traffic data processing
- G08G1/0133—Traffic data processing for classifying traffic situation
-
- 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
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
- G08G1/0112—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from the vehicle, e.g. floating car data [FCD]
Abstract
A traffic condition prediction method based on big data comprises the following steps: s1, acquiring basic data, wherein the basic data comprises map data and vehicle track data; s2, adding a time tag in the basic data to obtain analysis sample data; s3, generating a preliminary prediction result based on the analysis sample data; s4, acquiring dynamic data, wherein the dynamic data comprises traffic interruption data; s5, optimizing the preliminary prediction result by using the dynamic data to obtain an optimized prediction result; and S6, carrying out error correction on the optimized prediction result to obtain a final prediction result. The invention provides a traffic condition prediction method and system based on big data, which have high accuracy of prediction results and are simple and feasible.
Description
Technical Field
The invention relates to the technical field of traffic prediction, in particular to a traffic condition prediction method and system based on big data.
Background
With the development of urbanization, the contradiction between traffic infrastructure and automobile holding capacity is more severe, the congestion problem is more serious, and economic loss, travel time consumption and environmental pollution are inevitably caused. The treatment of traffic jam is firstly prevented, the traffic state change trend in a short time is predicted according to the existing traffic state of a road, and the possible jam phenomenon is early warned; and then, information platforms such as traffic broadcast and microblog are used for sending out early warning, leading vehicles to reasonably select running routes and strengthening order management so as to avoid congestion or relieve congestion degree. Therefore, how to establish a long-acting model to perform timely early warning on traffic jam is a research hotspot of optimizing an urban intelligent traffic system. In the prior art, a plurality of traffic condition prediction methods are available, which can realize prediction by using complex algorithms such as machine learning algorithms or deep learning algorithms based on traffic flow, and the methods generally have the problem of difficult realization.
Disclosure of Invention
In order to solve the defects in the prior art, the invention provides a traffic condition prediction method and system based on big data, and the prediction result has high accuracy and is simple and easy to implement.
In order to achieve the purpose, the invention adopts the specific scheme that: a traffic condition prediction method based on big data comprises the following steps:
s1, acquiring basic data, wherein the basic data comprises map data and vehicle track data;
s2, adding a time tag in the basic data to obtain analysis sample data;
s3, generating a preliminary prediction result based on the analysis sample data;
s4, acquiring dynamic data, wherein the dynamic data comprises traffic interruption data;
s5, optimizing the preliminary prediction result by using the dynamic data to obtain an optimized prediction result;
and S6, carrying out error correction on the optimized prediction result to obtain a final prediction result.
As a further optimization of the traffic condition prediction method based on big data: in S1, the map data is composed of a plurality of road sub-data, the road sub-data including a road starting point, a road ending point, and a road capacity;
the method for acquiring the vehicle track data comprises the following steps:
s11, obtaining the reading authority of the vehicle position information;
s12, acquiring a plurality of original track data of the vehicle in a plurality of sample periods, wherein the original track data consists of a plurality of road subdata; s13, extracting all road subdata from all original track data, and giving a position weight to each road subdata to obtain position sample data;
and S14, combining all the position sample data into vehicle track data.
As a further optimization of the traffic condition prediction method based on big data: the specific method of S2 includes:
s21, dividing twenty-four hours into a plurality of time segments on average;
s22, adding one or more time slices in each position sample data;
s23, aggregating all time slices in the same position sample data to obtain time labels;
and S24, taking the map data and the vehicle track data added with the time tags as analysis sample data.
As a further optimization of the traffic condition prediction method based on big data: the specific method of S3 includes:
s31, calculating an increment reference value based on the success rate of the reading authority of the vehicle position information acquired in the S11;
s32, carrying out area division on the map to obtain a plurality of prediction areas, and mapping the map data into the prediction areas;
s33, generating a plurality of increment adjusting values corresponding to the prediction areas one by one;
s34, fusing all the incremental adjustment values with the incremental reference values one by one to obtain a plurality of incremental actual values;
s35, generating road load reference data according to the vehicle track data;
s36, adding the road load reference data and the increment actual value to obtain road load initial prediction data;
and S37, integrating the initial prediction data of all road loads into a preliminary prediction result.
As a further optimization of the traffic condition prediction method based on big data: the specific method of S35 includes:
s351, constructing a time axis, wherein the time axis averagely divides twenty-four hours into a plurality of time elements, and the length of each time element is smaller than that of each time fragment;
s352, mapping the time labels to a time axis;
s353, calculating the number of all vehicle track data related to the road sub-data in one time element, and recording as a capacity analysis value;
and S354, calculating road load reference data according to the capacity analysis value and the road capacity in the road sub data.
As a further optimization of the traffic condition prediction method based on big data: in S4, the traffic interruption data includes road sub-data and interruption reason data, the interruption reason data is accident sub-data and/or construction sub-data, and the dynamic data further includes weather data.
As a further optimization of the traffic condition prediction method based on big data: the specific method of S5 includes:
s51, generating a load optimization value based on the traffic interruption data;
s52, fusing the load optimization value and the road load initial prediction data to obtain road load optimization prediction data;
s53, replacing the road load initial prediction data in the preliminary prediction result with road load optimization prediction data;
s54, generating a load batch processing value based on the weather data;
and S55, carrying out batch processing on the road load initial prediction data and the road load optimization prediction data in the preliminary prediction result by using the load batch processing value to obtain an optimization prediction result.
As a further optimization of the traffic condition prediction method based on big data: the specific method of S6 includes:
s61, acquiring historical congestion data based on the road sub-data;
s62, generating an error correction value based on the historical congestion data;
and S63, carrying out error correction on the optimized prediction result by using the error correction value to obtain a final prediction result.
A big data based traffic condition prediction system for performing the above big data based traffic condition prediction method, the system comprising:
the data transmission mechanism is used for acquiring basic data and dynamic data and outputting a final prediction result;
the data processing mechanism is used for generating analysis sample data, a preliminary prediction result, an optimized prediction result and a final prediction result;
and the data storage mechanism is used for storing the basic data and the final prediction result.
Has the advantages that: the method obtains the number of vehicles appearing on the road at the target moment based on the track of the vehicles, judges whether congestion occurs by comparing the number with the road capacity, and comprehensively considers influence factors such as traffic accidents, construction, weather and the like, thereby ensuring the accuracy of the prediction result.
Drawings
FIG. 1 is a flow chart of the present invention;
fig. 2 is an illustration of a ground plan in an embodiment.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, a traffic condition prediction method based on big data includes S1 to S6.
And S1, acquiring basic data, wherein the basic data comprises map data and vehicle track data.
In S1, the map data is composed of a plurality of road sub-data including a road start point, a road end point, and a road capacity. The road sub-data may be represented as [ c ]i,cj,capij]Wherein c isiAs a starting point of the road, cjFor road end point, capijIs the road capacity. The specific method for acquiring the road subdata comprises the following steps: firstly, obtaining a map image, then extracting all roads from the map image and representing the roads as a reticular line graph, calling all cross points and end points in the reticular line graph as nodes and forming a node set, then setting a node mark c for each node, and then utilizing the node marks corresponding to the starting point and the end point of the roadTo identify roads, e.g. Rij:[ci,cj]Thereafter, the road capacity of each road is calculated
Wherein k isijNumber of lanes of road,/ijThe length of the road, m is the standard berth length, and finally the road capacity capijWith the sign R of the roadij:[ci,cj]And combining the road sub-data. In the invention, the road capacity is the maximum number of vehicles which can exist simultaneously on the road, and in practical situations, the lane width can meet the traffic demands of various vehicles, and meanwhile, the upper part of the road is an open environment without considering the vehicle height, so the road capacity only needs to consider the relation between the vehicle length and the road length. In the calculation, a lane is divided into a plurality of segments on average, and only one vehicle exists in each segment, the length of the segment should be the length of the vehicle plus the safe distance between the vehicle and the front vehicle. When congestion occurs, the distance between the vehicles is reduced, namely the distance between two adjacent vehicles is obviously smaller than the normal safe distance, and the length of the segment is also reduced correspondingly. However, the distance between the vehicles is not too small, otherwise, the vehicles cannot move smoothly, for example, when an accident happens in the front, if the distance between the vehicles and the front vehicle is too small, the vehicles cannot smoothly bypass the accident vehicle, so the invention calculates the road capacity by using the standard berth length, and the road capacity reaches the maximum value under the condition that the vehicles can normally run. Since the main traffic vehicle of an urban road is a passenger car, the standard parking space length m can be selected to be the length of a small standard parking space, for example 6000 mm. Of course, vehicles with vehicle length exceeding the standard parking space length, such as buses, exist on the road, but the number of such vehicles is small, and the introduction of the vehicle length of such vehicles into the calculation process of the road capacity only causes the reduction of the calculation result of the road capacity, so that the vehicles are ignored in the invention, and only passenger cars are used as calculation bases.
At S1, the method of acquiring vehicle trajectory data includes S11 to S14.
And S11, acquiring the reading authority of the vehicle position information. The data source of the vehicle position information may be a vehicle navigation system or a device such as a smartphone of the driver, and in consideration of sensitive information that the position information belongs to an individual, the read authority of the vehicle position information is first acquired, and if the acquisition is successful, S12 is executed, and if the acquisition is unsuccessful, the operation is terminated.
And S12, acquiring a plurality of original track data of the vehicle in a plurality of sample periods, wherein the original track data is composed of a plurality of road sub-data. In urban traffic, vehicles can be simply divided into two categories, namely a home vehicle and a commercial vehicle, wherein the home vehicle is mainly used for daily commuting, the track of the home vehicle is relatively fixed, and the track of a bus in the commercial vehicle is also fixed due to the fixed line, and on the other hand, the commuting time and the commercial time of the bus are also fixed, so that the home vehicle and the bus can be considered to always appear on a fixed road at a fixed time. It should be noted that the track of the bus should be unique, but the domestic vehicles are different, and although the track of the domestic vehicle is relatively fixed, the track is not unique, and there may be multiple tracks, for example, in fig. 2, when the domestic vehicle Car isiWhen the starting point of the commuting route is A and the end point is B, the selectable track is { [ c { [ C ]1,c2]，[c2,c4]，[c4,c5]And { [ c ]1,c2]，[c2,c3]，[c3,c6]Two kinds. Considering that the daily commute period is in most cases one day, the factor in this embodiment is also that one sample period is set to one day, i.e. 24 hours, in other embodiments of the invention, other time lengths may be selected according to actual requirements.
And S13, extracting all the road sub-data from all the original track data, and giving a position weight to each road sub-data to obtain position sample data. For a home vehicle, although there may be multiple raw trajectory data for each, the probability of selecting each trajectory during actual travel is different, for example, at G1And G2Two original trajectoriesIn the data, G1Shorter path of (G)2Greater road width, greater number of lanes, greater road capacity, then CariThere will be a higher probability according to G1Move but when G1The vehicle may select G when congestion occurs2To avoid congestion. The raw trajectory data may first be given a trajectory weight α, e.g. may be G1Given a track weight of 2, G2And giving track weight 1, wherein the probability of selecting the original track data by the vehicle can be represented by giving the original track data track weight. On the basis, the track weight alpha can be directly given to the road subdata in the original track data, namely G1In (c)1,c2]、[c2,c4]And [ c)4,c5]All track weights of 2, at [ c ]4,c5]In, [ c ]1,c2]、[c2,c3]And [ c)3,c6]Are all 1, and then the track weights corresponding to the same road sub-data are added to obtain the position weight, e.g. [ c ]1,c2]Has a position weight of 3, [ c ]2,c4]And [ c)4,c5]Has a position weight of 2, [ c ]2,c3]And [ c)3,c6]Is 1. Since the road capacity is not required to be used when generating the vehicle trajectory data, only the road sign is used here.
And S14, combining all the position sample data into vehicle track data. The vehicle track data is composed of a plurality of road sub-data with track weight, which can be expressed by multi-tuple, such as CariThe vehicle track data of (1) is GJi＝{3[c1,c2],2[c2,c4],2[c4,c5],1[c2,c3],1[c3,c6]}。
And S2, adding a time tag in the basic data to obtain analysis sample data. Specifically, the time stamp is added to the vehicle trajectory data. The time tag is used to identify the time when the vehicle is traveling according to the vehicle trajectory data.
Specific methods of S2 include S21 to S24.
S21, equally dividing twenty-four hours into a plurality of time segments. The more the number of the time segments is, the higher the accuracy of the time labels is, but the more the subsequent processing process becomes complicated, in practical application, the appropriate number of the time segments can be selected according to the requirement, specifically, if the local vehicle holding quantity is larger, or the road traffic capacity is weaker, the number of the time segments can be increased when congestion is more likely to occur, or conversely, if the local vehicle holding quantity is smaller, or the road traffic capacity is stronger, the number of the time segments can be reduced when congestion is less likely to occur, so as to reduce the complexity of the method. For example, in this embodiment, 24 hours can be divided equally into 480 time slices, each time slice being 3 minutes in length, and the time slices are denoted as tpP is more than or equal to 001 and less than or equal to 480, and p is a three-digit integer.
And S22, adding one or more time slices in each position sample data. Because the vehicle needs time in the process of passing through the road, the required time length is positively correlated with the road length, the addition of the time segment in the position sample data can represent the traveling time of the vehicle on the road corresponding to the position sample data, and the more vehicles on the same road in the same time period, the easier the vehicle approaches the road capacity, and the more congestion is easy to occur. The position sample data after the time slice is added can be represented as [ c ]i,cj,tptp+1......]. Since the travel time of the vehicle on the road is not necessarily an integer multiple of the time segment, a rounding manner is adopted when adding the time segment, for example, when the travel time of the vehicle on one road is 2 minutes and the travel time on the next road is 4 minutes, one time segment can be added to the position sample data corresponding to two roads respectively. It should be noted that at least one time slice should be included in one position sample data to ensure the continuity of the vehicle trajectory data.
And S23, aggregating all time slices in the same position sample data to obtain the time labels. When a position sample dataThe inclusion of multiple time slices can make recording and processing difficult, and therefore the time slices need to be aggregated into a time tag to reduce the complexity of the subsequent process. The time stamp after aggregation may be denoted tpqAnd p is more than or equal to 001 and less than or equal to 480,001 and less than or equal to 480, p is less than or equal to q, and q is a three-bit integer, wherein p is less than q when only one time segment is added to the position sample data, and p is less than q when a plurality of time segments are added to the position sample data. Still with CariFor example, the vehicle trajectory data GJiAfter the time tag is added, is represented as
GJi＝{3[c1,c2,t160160],2[c2,c4,t160162],2[c4,c5,t162165],1[c2,c3,t160166],1[c3,c6,t166170]}。
And S24, taking the map data and the vehicle track data added with the time tags as analysis sample data.
And S3, generating a preliminary prediction result based on the analysis sample data. Specific methods of S3 include S31 to S37.
And S31, calculating an increment reference value based on the success rate of the reading authority of the vehicle position information acquired in the S11. Since the vehicle trajectory data, which is one of the analysis sample data, can be obtained gradually after the vehicle is authorized, if the vehicle trajectory data is not authorized, the vehicle trajectory data of the vehicle is lost, which directly affects the accuracy of the prediction result, and in order to improve the accuracy of the prediction result, the interference of the vehicle which does not obtain the reading right needs to be considered. The method for calculating the increment reference value comprises the following steps
Beta is the success rate of the reading authority for obtaining the vehicle position information, and because beta satisfies 0-1, lambda satisfies 0.24-0.4.
And S32, dividing the map into a plurality of prediction areas, and mapping the map data into the prediction areas. In a city, the probability of congestion in different areas is different, meanwhile, the attractiveness of different areas to people is different, and in daily commutes, people can go home and go to entertainment or consumption after work, so that for areas including superstores or entertainment centers, vehicles from other areas should be considered in addition to vehicles passing through normal commutes in the areas, and therefore, the prediction areas need to be divided, so that the traffic condition possibly occurring in each prediction area is accurately evaluated, and the accuracy of the overall prediction result is further improved.
And S33, generating a plurality of increment adjusting values corresponding to the prediction areas one by one. The prediction area may be divided into a plurality of categories, and the incremental adjustment values corresponding to different categories are different, for example, the prediction area may be divided into a central business area, a residential area, an industrial area, a transition area between urban and rural areas, and a suburban and rural area, and the incremental adjustment values corresponding to five categories of prediction areas are gradually decreased. The incremental adjustment was recorded as μ, and μ < 0.3.
And S34, fusing all the incremental adjustment values with the incremental reference values one by one to obtain a plurality of incremental actual values. The specific fusion method is
And S35, generating road load reference data according to the vehicle track data. Specific methods of S35 include S351 to S534.
S351, constructing a time axis, wherein the time axis averagely divides twenty-four hours into a plurality of time elements, and the length of each time element is smaller than that of each time clip. Because the time slices are used for forming vehicle track data corresponding to the vehicles, the action objects of the time slices are the vehicles, the time elements are used for generating traffic prediction results, the action objects of the time slices are traffic conditions of the whole city, the precision requirement on the time elements is higher than that on the time slices, the length of the time elements is smaller than that of the time slices, and the number of the corresponding time elements is larger than that of the time slices.
And S352, mapping the time labels on a time axis. Because the length of a time slice is greater than the length of a time bin, a time tag will correspond to multiple time bins when the time tag is mapped onto the time axis.
And S353, calculating the number of all vehicle track data related to the road sub-data in one time element, and recording the number as a capacity analysis value. At the same time, the more the number of vehicle track data associated with the road sub-data is, the more vehicles that may appear on the road corresponding to the road sub-data are, the more congestion is likely to appear, so the number of all vehicle track data associated with the road sub-data in one time element is recorded as a capacity analysis value, and the capacity analysis value is recorded as RF.
And S354, calculating road load reference data according to the capacity analysis value and the road capacity in the road sub data. The specific calculation method of the road load reference data is
And S36, adding the road load reference data and the increment actual value to obtain initial road load prediction data. That is, the road load initial prediction data BDS is BDJ + θ. Because the incremental actual value includes the incremental adjustment value and the incremental reference value, namely two influence factors including the type of the predicted area and the possible interference generated by the vehicle which fails to acquire the position information reading authority, the initial predicted data of the road load should be larger than the reference data of the road load, when the BDS is larger than or equal to 1, the number of vehicles which may appear on the road is not smaller than the road capacity, so that the road will be blocked, and when the BDS is smaller, the larger the gap between the vehicles which may appear on the road and the road capacity is, the more difficult the congestion is.
And S37, integrating the initial prediction data of all road loads into a preliminary prediction result.
And S4, acquiring dynamic data, wherein the dynamic data comprises traffic interruption data. In S4, the traffic interruption data includes road sub-data and interruption reason data, the interruption reason data is accident sub-data and/or construction sub-data, and the dynamic data further includes weather data. In addition to the two factors of influence of the type of the predicted area and the possible interference caused by the vehicle which fails to acquire the position information reading authority, in the actual situation, more situations which may cause the occurrence of congestion occur, wherein traffic interruption and severe weather are the two most common factors, so the invention further considers the two factors to further optimize the initial road load prediction data. Specifically, if severe weather, such as rain and snow weather, occurs, the congestion probability is increased on most roads, and traffic interruption only causes the probability of congestion on some roads to be increased greatly, where the traffic sub-data causes the probability of congestion to be increased greatly in a short time, and the traffic sub-data causes the probability of congestion to be increased greatly in a long time.
And S5, optimizing the preliminary prediction result by using the dynamic data to obtain an optimized prediction result. Specific methods of S5 include S51 to S55.
And S51, generating a load optimization value based on the traffic interruption data. When the traffic interruption data is the accident subdata, an accident factor A is generated, when the traffic interruption data is the construction subdata, a construction factor C is generated, and the load optimization value bd is A + C, wherein if the accident subdata does not exist, A is 0, and if the construction subdata does not exist, C is 0.
And S52, fusing the load optimization value and the road load initial prediction data to obtain road load optimization prediction data. Since the traffic interruption data corresponds to the road and has no direct relation with other roads, the load optimization value bd may be added to the road load preliminary prediction data corresponding to the preliminary prediction result.
And S53, replacing the road load initial prediction data in the preliminary prediction result with road load optimization prediction data.
And S54, generating a load batch processing value based on the weather data. Because the influence of severe weather on traffic is all-around, and the influence acts on each road, when severe weather occurs, all data in the preliminary prediction result need to be subjected to uniform batch processing, and the load batch processing value is recorded as W.
And S55, carrying out batch processing on the road load initial prediction data and the road load optimization prediction data in the preliminary prediction result by using the load batch processing value to obtain an optimization prediction result. Specifically, the load batch processing value W is added to all the initial road load prediction data and the optimized road load prediction data to obtain the road load processing prediction results, and then all the road load processing prediction results are integrated into the optimized prediction results.
And S6, carrying out error correction on the optimized prediction result to obtain a final prediction result. Specific methods of S6 include S61 to S63. The optimized prediction result covers various factors which can cause congestion, such as vehicle factors, area factors, accident factors, construction factors, weather factors and the like, but errors still exist, mainly the deviation possibly existing between the theoretical prediction result and the actual result, and in order to eliminate the deviation and further improve the accuracy of the prediction result, the error correction needs to be carried out on the optimized prediction result again.
And S61, acquiring historical congestion data based on the road sub-data. The historical congestion data corresponds to the road, and after the historical congestion data is obtained, error correction can be performed on the initial prediction data of the road load or the optimized prediction data of the road load corresponding to the road in the optimized prediction result.
And S62, generating an error correction value based on the historical congestion data. The historical congestion data can also be acquired in a plurality of sample periods, and an error correction value F is generated by using the times of the historical congestion data appearing on the road in the plurality of sample periods, wherein the size of the error correction value F is inversely related to the times of the historical congestion data. Since the congestion corresponds to the road, the error correction value F also corresponds one-to-one to the road.
And S63, carrying out error correction on the optimized prediction result by using the error correction value to obtain a final prediction result. Specifically, the error correction value F and the corresponding road load processing result in the optimization prediction result are superposed to obtain a road load correction prediction result, and then the road load correction prediction result is used for replacing the road load processing prediction result in the optimization prediction result to obtain a final prediction result.
The system comprises a data transmission mechanism, a data processing mechanism and a data storage mechanism.
And the data transmission mechanism is used for acquiring the basic data and the dynamic data and outputting a final prediction result.
And the data processing mechanism is used for generating analysis sample data, a preliminary prediction result, an optimized prediction result and a final prediction result.
And the data storage mechanism is used for storing the basic data and the final prediction result.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (5)
1. A traffic condition prediction method based on big data is characterized by comprising the following steps:
s1, acquiring basic data, wherein the basic data comprises map data and vehicle track data;
in S1, the map data is composed of a plurality of road sub-data, the road sub-data including a road starting point, a road ending point, and a road capacity;
the method for acquiring the vehicle track data comprises the following steps:
s11, obtaining the reading authority of the vehicle position information;
s12, acquiring a plurality of original track data of the vehicle in a plurality of sample periods, wherein the original track data consists of a plurality of road subdata;
s13, extracting all road subdata from all original track data, and giving a position weight to each road subdata to obtain position sample data;
s14, combining all position sample data into vehicle track data;
s2, adding a time tag in the basic data to obtain analysis sample data;
s3, generating a preliminary prediction result based on the analysis sample data;
the specific method of S3 includes:
s31, calculating an increment reference value based on the success rate of the reading authority of the vehicle position information acquired in the S11;
s32, carrying out area division on the map to obtain a plurality of prediction areas, and mapping the map data into the prediction areas;
s33, generating a plurality of increment adjusting values corresponding to the prediction areas one by one;
s34, fusing all the incremental adjustment values with the incremental reference values one by one to obtain a plurality of incremental actual values;
s35, generating road load reference data according to the vehicle track data;
s36, adding the road load reference data and the increment actual value to obtain road load initial prediction data;
s37, integrating the initial prediction data of all road loads into a preliminary prediction result;
s4, acquiring dynamic data, wherein the dynamic data comprises traffic interruption data;
s5, optimizing the preliminary prediction result by using the dynamic data to obtain an optimized prediction result;
the specific method of S5 includes:
s51, generating a load optimization value based on the traffic interruption data;
s52, fusing the load optimization value and the road load initial prediction data to obtain road load optimization prediction data;
s53, replacing the road load initial prediction data in the preliminary prediction result with road load optimization prediction data;
s54, generating a load batch processing value based on the weather data;
s55, carrying out batch processing on the road load initial prediction data and the road load optimization prediction data in the preliminary prediction result by using the load batch processing value to obtain an optimization prediction result;
s6, carrying out error correction on the optimized prediction result to obtain a final prediction result;
the specific method of S6 includes:
s61, acquiring historical congestion data based on the road sub-data;
s62, generating an error correction value based on the historical congestion data;
and S63, carrying out error correction on the optimized prediction result by using the error correction value to obtain a final prediction result.
2. The big data-based traffic condition prediction method according to claim 1, wherein the specific method of S2 comprises:
s21, dividing twenty-four hours into a plurality of time segments on average;
s22, adding one or more time slices in each position sample data;
s23, aggregating all time slices in the same position sample data to obtain time labels;
and S24, taking the map data and the vehicle track data added with the time tags as analysis sample data.
3. The big-data-based traffic condition prediction method according to claim 2, wherein the specific method of S35 comprises:
s351, constructing a time axis, wherein the time axis averagely divides twenty-four hours into a plurality of time elements, and the length of each time element is smaller than that of each time fragment;
s352, mapping the time labels to a time axis;
s353, calculating the number of all vehicle track data related to the road sub-data in one time element, and recording as a capacity analysis value;
and S354, calculating road load reference data according to the capacity analysis value and the road capacity in the road sub data.
4. The big-data-based traffic condition prediction method of claim 3, wherein in S4, the traffic interruption data comprises road sub-data and interruption reason data, the interruption reason data is accident sub-data and/or construction sub-data, and the dynamic data further comprises weather data.
5. A big-data based traffic condition prediction system for performing the big-data based traffic condition prediction method according to claim 1, the system comprising:
the data transmission mechanism is used for acquiring basic data and dynamic data and outputting a final prediction result;
the data processing mechanism is used for generating analysis sample data, a preliminary prediction result, an optimized prediction result and a final prediction result;
and the data storage mechanism is used for storing the basic data and the final prediction result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011625743.XA CN112614349B (en) | 2020-12-31 | 2020-12-31 | Traffic condition prediction method and system based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011625743.XA CN112614349B (en) | 2020-12-31 | 2020-12-31 | Traffic condition prediction method and system based on big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112614349A CN112614349A (en) | 2021-04-06 |
| CN112614349B true CN112614349B (en) | 2021-09-03 |
Family
ID=75252983
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011625743.XA Active CN112614349B (en) | 2020-12-31 | 2020-12-31 | Traffic condition prediction method and system based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112614349B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1614356A (en) * | 2003-11-04 | 2005-05-11 | 日本先锋公司 | Traffic-condition notifying device, its system, its method, its program and recording medium storing the program |
| CN101438334A (en) * | 2006-03-03 | 2009-05-20 | 因瑞克斯有限公司 | Dynamic time series prediction of future traffic conditions |
| CN109377752A (en) * | 2018-10-19 | 2019-02-22 | 桂林电子科技大学 | Short-term traffic flow variation prediction method, apparatus, computer equipment and storage medium |
| CN109544911A (en) * | 2018-10-30 | 2019-03-29 | 中山大学 | A kind of city road net traffic state prediction technique based on LSTM-CNN |
| CN110570660A (en) * | 2019-11-06 | 2019-12-13 | 深圳市城市交通规划设计研究中心有限公司 | real-time online traffic simulation system and method |
-
2020
- 2020-12-31 CN CN202011625743.XA patent/CN112614349B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1614356A (en) * | 2003-11-04 | 2005-05-11 | 日本先锋公司 | Traffic-condition notifying device, its system, its method, its program and recording medium storing the program |
| CN101438334A (en) * | 2006-03-03 | 2009-05-20 | 因瑞克斯有限公司 | Dynamic time series prediction of future traffic conditions |
| CN109377752A (en) * | 2018-10-19 | 2019-02-22 | 桂林电子科技大学 | Short-term traffic flow variation prediction method, apparatus, computer equipment and storage medium |
| CN109544911A (en) * | 2018-10-30 | 2019-03-29 | 中山大学 | A kind of city road net traffic state prediction technique based on LSTM-CNN |
| CN110570660A (en) * | 2019-11-06 | 2019-12-13 | 深圳市城市交通规划设计研究中心有限公司 | real-time online traffic simulation system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112614349A (en) | 2021-04-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109927709B (en) | Vehicle driving route working condition determining method, energy management method and system | |
| US20190171227A1 (en) | Optimization of mission efficiency through platoon opportunity assessment | |
| Sims et al. | The Sydney coordinated adaptive traffic (SCAT) system philosophy and benefits | |
| CN108848460B (en) | Man-vehicle association method based on RFID and GPS data | |
| Pinjari et al. | Highway capacity impacts of autonomous vehicles: An assessment | |
| WO2014154614A1 (en) | Backend for driver assistance systems | |
| CN110807926B (en) | Road impedance prediction method and device based on hybrid traffic | |
| CN110472999B (en) | Passenger flow mode analysis method and device based on subway and shared bicycle data | |
| CN103632547B (en) | The lower link travel time prediction system of moving bottleneck impact and implementation method | |
| CN103177562A (en) | Method and device for obtaining information of traffic condition prediction | |
| CN109661692B (en) | Traffic incident prediction method and device and terminal equipment | |
| CN112382098A (en) | Traffic jam detection method and device, electronic equipment and storage medium | |
| CN105046983A (en) | Traffic flow prediction system and method based on vehicle-road cooperation | |
| CN112785835A (en) | Method and device for acquiring road condition information and vehicle-mounted device | |
| CN110379159A (en) | Trip chain extracting method based on motor vehicle electronic mark data | |
| CN112926768A (en) | Ground road lane-level traffic flow prediction method based on space-time attention mechanism | |
| CN112614349B (en) | Traffic condition prediction method and system based on big data | |
| CN110556014A (en) | intelligent bus dispatching platform system | |
| CN113724497A (en) | Method and device for predicting real-time traffic flow of target road | |
| CN111767644B (en) | Method for estimating actual traffic capacity of expressway road section by considering speed limit influence of single tunnel | |
| DE102012210454A1 (en) | Method for providing data for electronic horizon for e.g. navigation apparatus of motor car, involves merging parameters from group of vehicles to data packet for electronic horizon when group of vehicles, comprises two other vehicles | |
| CN112833884B (en) | Navigation method and navigation system based on big data | |
| CN107657814B (en) | Road condition information generation method and device | |
| CN114495519B (en) | Intelligent visual display system for expressway | |
| CN115204755A (en) | Service area access rate measuring method and device, electronic equipment and readable storage medium |
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 |