Trajectory and meteorological data fusion analysis method and system

Abstract

The application provides a trajectory and meteorological data fusion analysis method and system. Trajectory data and meteorological data are fused based on a road network topology structure, the most reliable meteorological data with the highest probability is selected in the case of multi-source meteorological data, and accurate average speed analysis and correlation analysis are carried out based on road section segmentation, thereby providing data support for order distribution and path planning of business personnel such as express delivery personnel or take-out personnel.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of data analysis, and particularly relates to a trajectory and meteorological data fusion analysis method and system. BACKGROUND

[0002] The characteristics of urban traffic are different due to the differences in population size, main industry, geographical location, etc. of different cities, but they have similar main features, i.e. the focus of urban traffic is passenger transport, commuting is the main traffic demand during the traffic peak period, the size of urban traffic volume is directly related to the traffic service level and traffic policy and traffic management of each city, etc. With the continuous increase of the number of motor vehicles in China, the problem of urban traffic congestion is becoming more and more serious. Micro-level traffic management and control is an effective means to solve the problem of traffic congestion, and how to identify the frequently occurring road sections and the frequently occurring time of traffic congestion is the basis of management and control.

[0003] And the road network is a complex system, the driving state of the vehicle on the road is restricted by weather, road conditions, vehicle state and other factors, and shows a certain correlation in the time and space scale. Time series, a group of data arranged in a certain order according to time, such as the traffic volume of the previous period and the traffic volume of the previous several periods, which are arranged in a group of regular traffic volume data sequences.

[0004] With the improvement of the Internet and the requirement of life convenience, domestic e-commerce and take-out industry develop rapidly, and the analysis of road navigation trajectory data is crucial for e-commerce and delivery personnel. At present, e-commerce and take-out platforms collect real-time trajectory data of delivery personnel, generating a large amount of trajectory data, but the analysis of these data is not accurate enough, and cannot provide strong support for business.

[0005] Therefore, a new data analysis method is needed to provide more accurate and objective data navigation support for e-commerce, take-out and other platforms.

[0006] The above statements are only used to provide background technical information related to the application, and the content described in this part is not prior art to other parts of the application unless indicated herein. SUMMARY

[0007] The application provides a trajectory and meteorological data fusion analysis method and system, which fuses trajectory data and meteorological data based on the topology of the road network, selects the meteorological data with the highest credible probability in the case of multiple source meteorological data, and performs accurate average speed analysis and correlation analysis based on road section segmentation, thereby providing data support for order distribution and path planning of business personnel such as express delivery personnel or delivery personnel.

[0008] In addition, the application also achieves: 1) through the road section level secondary index, replacing the coarse-grained area statistics, improving the road network index efficiency, and realizing the accurate analysis of the road network; 2) through the credible probability model to optimize the selection accuracy of meteorological data, solving the conflict problem of multi-source meteorological data; 3) through the fusion of the influence of meteorology on the distribution speed, supporting the path optimization decision.

[0009] In the specific take-out scenario, the application realizes the accurate analysis of the take-out employee trajectory, and the fusion and correlation analysis of the meteorological data and time data by matching the take-out employee trajectory based on the road section and fusing the surrounding meteorological station data. The algorithm can be distributed for calculation and can process massive trajectory data, so as to realize the accurate analysis of the influence of the average speed of the take-out employee on different road sections under the meteorological condition, and provide data support for the subsequent take-out employee path planning and overall scheduling scheme under different meteorological conditions.

[0010] According to a first aspect of the embodiments of the application, a trajectory and meteorological data fusion analysis method is provided, including the following steps:

[0011] The target area is gridded based on the road network topology and a secondary index is established, and the road section is processed and associated with the road network, the meteorological station of the target area is matched and associated with the road section, and the road network grid index, the road section association library and the meteorological station association library are obtained;

[0012] According to the road network grid index and the road section association library, the target personnel trajectory data is matched with the road network road section, the trajectory point-road section association relationship and the offset distance are obtained, and the road section passing event is generated;

[0013] According to the meteorological station association library, the meteorological station data with high credible probability is selected and fused with the passing event to obtain meteorological fusion data;

[0014] Based on the meteorological fusion data, the correlation between the road section speed and the meteorological factor is analyzed by the Pearson coefficient.

[0015] In some embodiments of the application, the target area is gridded based on the road network topology and a secondary index is established, and the road section is processed and associated with the road network, including:

[0016] The target area road network is segmented according to the preset length, a secondary index containing the road section geometric center point coordinates and the offset distance is established;

[0017] Based on the kilometer-level grid division, the cross association relationship between the grid and the road section is established, the grid and the cross relationship road section are indexed, and the road network grid index and the road section association library are obtained.

[0018] In some embodiments of the application, the meteorological station of the target area is matched and associated with the road section, including:

[0019] The correlation degree k of the weather station and the road section is calculated according to the position of the weather station and the effective radius, and the formula is:

[0020] ;

[0021] ;

[0022] If r or , then k = 0;

[0023] If r and , then + ;

[0024] wherein and are the distances from the weather station to the two ends of the road section; r is the effective radius of the weather station;

[0025] According to the correlation degree k of the weather station and the road section, the road section is associated with at least one weather station, and a weather station association library is obtained.

[0026] In some embodiments of the present application, the target personnel trajectory data is matched with the road network section according to the road network grid index and the road section association library, the trajectory point-road section association relationship and the offset distance are obtained, and the road section passing event is generated, including:

[0027] The target personnel trajectory data is obtained through the Kafka message queue;

[0028] The trajectory data is searched according to the trajectory point, the road network grid index and the road section association library, matched to the road section and the trajectory point offset distance is calculated;

[0029] According to the trajectory point offset distance, the trajectory point event of the adjacent road section end point is improved, and a series of trajectory point events of the target personnel trajectory data are determined; the trajectory point event includes the trajectory point position and the trajectory time;

[0030] Through a series of trajectory point arrays, the passing event corresponding to the target personnel trajectory data is generated, including the road section ID, the entering time and the leaving time.

[0031] In some embodiments of the present application, the trajectory point offset distance is calculated, including:

[0032] Suppose the trajectory point is a(x0, y0), and the center line points of the two road sections closest to the grid where the trajectory point a(x0, y0) is located are p1(x1, y1) and p2(x2, y2) respectively;

[0033] The offset distance of the trajectory point a(x0, y0) is calculated by the perpendicular line method , the formula is:

[0034] ;

[0035] t < 0, ;

[0036] t > 0, ;

[0037] wherein, is the reference point offset.

[0038] In some embodiments of the present application, the weather station data with high credibility probability is selected and fused with the path event, including:

[0039] For the path event time t0, the credibility probability pᵢ of the associated weather station data is calculated:

[0040] ;

[0041] ;

[0042] wherein, ω1, ω2 are weight coefficients; m is the effective weather time;

[0043] The weather station data with the highest credibility probability of the associated weather station is selected and fused with the path event to obtain the path event including weather data.

[0044] In some embodiments of the present application, after generating the path event of the road section, it further includes:

[0045] If there are n path events in the entire road section passing time range, the average speed v of the road section is calculated based on the path event, and the formula is:

[0046] V = ;

[0047] wherein, t1 is the entering time, and t2 is the ending time.

[0048] In some embodiments of the present application, the correlation between the road section speed and the weather factor is analyzed by Pearson coefficient, including:

[0049] For different road sections, there are N historical data; the weather factors include temperature, precipitation and wind speed;

[0050] The correlation coefficient r between the average speed and the weather factor is calculated, and the formula is:

[0051] r = Cov(X, Y) / (σX * σY);

[0052] Cov(X, Y) = ∑[(x - x̅) * (y - y̅)] / N;

[0053] wherein X is the average speed; Y is the meteorological factor; σX and σY are the standard deviations of X and Y; x̅ is the average of the plurality of average speed records; y̅ is the average of the plurality of meteorological factor records;

[0054] When r approaches 1, it indicates a positive correlation, and when r approaches -1, it indicates a negative correlation.

[0055] According to a second aspect of the embodiments of the present application, a trajectory and meteorological data fusion analysis system is provided, comprising:

[0056] A road network preprocessing module is configured to grid the target area based on the road network topology and establish a secondary index, and perform segmentation processing and road network association on the road segments, match and associate the meteorological stations of the target area with the road segments, and obtain a road network grid index, a road segment association library, and a meteorological station association library.

[0057] A road network matching module is configured to match the target personnel trajectory data with the road network segments according to the road network grid index and the road segment association library, obtain a trajectory point-road segment association relationship and an offset distance, and generate a road segment passage event.

[0058] A data fusion module is configured to select meteorological station data with high credibility probability from the meteorological station association library and fuse the selected meteorological station data with the passage event to obtain meteorological fusion data.

[0059] An association analysis module is configured to analyze the correlation between the road segment speed and the meteorological factor based on the meteorological fusion data by using the Pearson coefficient.

[0060] According to a third aspect of the embodiments of the present application, a trajectory and meteorological data fusion analysis device is provided, comprising a storage unit configured to store executable instructions, and a processing unit configured to connect with the storage unit to execute the executable instructions to complete the trajectory and meteorological data fusion analysis method.

[0061] According to a fourth aspect of the embodiments of the present application, a computer readable storage medium is provided, which stores a computer program; the computer program is executed by a processor to implement the trajectory and meteorological data fusion analysis method.

[0062] The trajectory and meteorological data fusion analysis method and system of the present application can realize accurate analysis of trajectories, fusion and association analysis of meteorological data and time data by matching the trajectories of delivery personnel based on road segments and fusing the data of surrounding meteorological stations. The algorithm can be distributed for computing and can process massive trajectory data to provide data support for order allocation and path planning of personnel such as delivery personnel or express delivery personnel.

[0063] In addition, this application also achieves: 1) improving the efficiency of road network indexing by replacing coarse-grained regional statistics with road segment-level secondary indexes, and realizing accurate analysis of the road network; 2) optimizing the accuracy of meteorological data selection through a reliable probability model, and solving the problem of conflict between multi-source meteorological data; 3) accurately supporting route optimization decisions by integrating the impact of meteorology on delivery speed. Attached Figure Description

[0064] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0065] Figure 1 The diagram illustrates the steps of a trajectory and meteorological data fusion analysis method according to an embodiment of this application.

[0066] Figure 2 The diagram shows a schematic representation of the road network preprocessing steps according to an embodiment of this application;

[0067] Figure 3 The diagram shows a schematic flowchart of road network preprocessing according to an embodiment of this application;

[0068] Figure 4 The diagram illustrates the steps of road network matching according to an embodiment of this application;

[0069] Figure 5 The diagram shows a schematic diagram of the road network matching process according to an embodiment of this application;

[0070] Figure 6 The flowchart of the generation path event according to an embodiment of this application is shown;

[0071] Figure 7 The diagram illustrates the data fusion steps according to an embodiment of this application;

[0072] Figure 8 The diagram shows a schematic of data fusion according to an embodiment of this application;

[0073] Figure 9 The diagram illustrates the steps of association analysis according to an embodiment of this application;

[0074] Figure 10 The diagram shows a schematic representation of a trajectory and meteorological data fusion analysis system according to an embodiment of this application.

[0075] Figure 11 The diagram shows a schematic of the structure of a trajectory and meteorological data fusion analysis device according to an embodiment of this application. Detailed Implementation

[0076] Regarding this application, with the advancement of the internet and the increasing demand for convenience in daily life, the domestic e-commerce and food delivery industries have experienced explosive growth. The analysis of road navigation trajectory data is crucial for e-commerce platforms and delivery drivers. Currently, both e-commerce and food delivery platforms collect delivery driver trajectory data in real time, generating massive amounts of trajectory data. However, current analysis of this data is mostly limited to regional and hotspot statistics, lacking precision and failing to integrate with meteorological data for comprehensive business support.

[0077] This invention proposes a method and system for fusion analysis of trajectory and meteorological data. By matching the trajectory of delivery riders based on road segments and integrating data from surrounding meteorological stations, it achieves accurate analysis of delivery rider trajectories and fusion and correlation analysis with meteorological and time data. The algorithm can perform distributed computing and process massive trajectory data, providing data support for order dispatching and route planning for delivery riders.

[0078] This process includes: gridding the target area based on road network topology and establishing a secondary index; segmenting road segments and associating them with the road network; matching and associating meteorological stations in the target area with road segments to obtain a road network grid index, a road segment association database, and a meteorological station association database; matching target personnel trajectory data with road network segments based on the road network grid index and road segment association database to obtain trajectory point-road segment association relationships and offset distances, generating road segment transit events; fusing meteorological station data with high reliability probability with transit events based on the meteorological station association database to obtain meteorological fusion data; and analyzing the correlation between road segment speed and meteorological factors using the Pearson coefficient based on the meteorological fusion data.

[0079] This application also achieves the following: 1) Improving road network indexing efficiency and enabling precise road network analysis by replacing coarse-grained regional statistics with a road segment-level secondary index; 2) Optimizing the accuracy of meteorological data selection through a reliable probability model, resolving the conflict problem of multi-source meteorological data; 3) Supporting route optimization decisions by integrating the impact of meteorology on delivery speed. This addresses the shortcomings of current trajectory analysis methods, which lack precision and fail to integrate with meteorological data to provide strong support for business operations.

[0080] To make the technical solutions and advantages of the embodiments of this application clearer, the exemplary embodiments of this application will be described in further detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not an exhaustive list of all embodiments. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other.

[0081] Example 1

[0082] Figure 1 The diagram illustrates the steps of a trajectory and meteorological data fusion analysis method according to an embodiment of this application.

[0083] like Figure 1 As shown, the trajectory and meteorological data fusion analysis method of this application includes the following steps:

[0084] S1: Based on the road network topology, the target area is gridded and a secondary index is established. The road segments are segmented and associated with the road network. The meteorological stations in the target area are matched and associated with the road segments to obtain the road network grid index, the road segment association library and the meteorological station association library.

[0085] S2: Based on the road network grid index and road segment association library, match the target personnel trajectory data with the road network segments to obtain the trajectory point-road segment association relationship and offset distance, and generate road segment path events;

[0086] S3: Based on the meteorological station association database, select meteorological station data with high reliability probability and fuse them with the associated events to obtain meteorological fusion data;

[0087] S4: Based on meteorological fusion data, the correlation between road segment speed and meteorological factors is analyzed using the Pearson coefficient.

[0088] This application embodiment improves road network indexing efficiency by replacing coarse-grained regional statistics with a road segment-level two-level index, achieving accurate road network analysis; and accurately supports route optimization decisions by integrating the impact of weather on delivery speed.

[0089] In addition, this application achieves accurate analysis of delivery rider trajectories and integration with meteorological and temporal data by matching delivery rider trajectories based on road segments and integrating data from surrounding weather stations. The algorithm can perform distributed computing and process massive amounts of trajectory data, thereby enabling accurate analysis of the impact of average delivery rider speed on different road segments under meteorological conditions. This provides data support for subsequent delivery rider route planning and overall scheduling schemes under different weather conditions.

[0090] First, data collection is conducted, including the collection of trajectory data and meteorological data. This application focuses on real-time data, using a KAFKA message queue to connect real-time trajectory data and meteorological data from the business platform and the meteorological platform. On the food delivery platform, trajectory data includes the delivery person's ID, location, and timestamp, while meteorological data includes temperature, humidity, precipitation, wind speed, wind direction, and timestamp.

[0091] Next, road network preprocessing is performed based on the road network topology, the region is gridded, and road segments are segmented. Based on this, road segment matching and association are performed on the meteorological stations.

[0092] Figure 2 The diagram shows a schematic representation of the road network preprocessing steps according to an embodiment of this application;

[0093] likeFigure 2 As shown, when preprocessing the road network in S1, the target area is grid-divided based on the road network topology, a secondary index is established, and the road segments are segmented and associated with the road network, including the following steps:

[0094] S11: Segment the road network of the target area according to a preset length, and establish a secondary index including the coordinates of the geometric center point of the road segment and the offset distance;

[0095] S12: Based on the grid division, establish the cross-association relationship between the grid and the road segment, index the road segments with the grid and cross relationship, and obtain the road network grid index and the road segment association library.

[0096] When preprocessing the road network in S1, it also includes matching and associating the weather stations in the target area, and the steps are as follows:

[0097] S13: Calculate the association degree k between it and the road segment according to the position and effective radius of the weather station. The formula is:

[0098] ;

[0099] ;

[0100] If > r or , then k = 0;

[0101] If < r and , then + ;

[0102] Where and are the distances from the weather station to both ends of the road segment; r is the effective radius of the weather station;

[0103] S14: According to the association degree k between the weather station and the road segment, associate at least one weather station with the road segment to obtain the weather station association library.

[0104] Figure 3 shows the flow principle diagram of the road network preprocessing according to the embodiment of the present application;

[0105] As Figure 3 shown, after the initialization of the road network topology data, first perform regional grid processing; then segment the road network, divide and associate the grid with longer road segments, for example, 300 meters as a road segment, establish a secondary index of the road network, and on this basis, perform kilometer-level grid processing and index the road segments with the grid and cross relationship.

[0106] The original topology of the road network is (V, E), where V is the intersection geometry and E is the set of roads. After segmentation, Eij is the j-th segment of road Ei. The road segment is represented by the set of center points (p1, p2... pn), where the coordinates of p1 are (x1, y1), and so on.

[0107] The grid Gmn represents the grid within the region, where m and n are grid numbers. The top-leftmost grid number is 0,0, and the top-left corner coordinates are (x0, y0), and so on. Typically, a grid size of 1 km x 1 km is used. Road segments intersecting with grids will be associated with each other. The process for confirming intersection relationships is as follows:

[0108] If a certain road segment has a center point O(Xi, Yi) that satisfies:

[0109] Xi > x0 + 1000*(m-1), and Xi <x0+1000*m;

[0110] Yi > y0 + 1000*(n-1), and Yi <y0+1000*n;

[0111] If the road segment intersects with grid Gm,n, then during data preprocessing, all road segments associated with the grid will be linked and the linking relationships will be stored.

[0112] During data preprocessing, the offset distance of the center point p of each road segment relative to the starting point of the road segment is calculated simultaneously. The formula is as follows:

[0113] ;

[0114] Then, the meteorological station road segment matching and association are performed.

[0115] Meteorological station data includes static and dynamic data. Static data includes the location of the meteorological station, its effective spatial range, and its effective time range. Dynamic data includes meteorological data such as temperature, humidity, precipitation, wind speed, wind direction, station number, and release time.

[0116] Meteorological stations are matched based on their location and effective spatial range, and virtual meteorological stations are established for each road segment. Therefore, one road segment can correspond to multiple virtual meteorological stations. In the subsequent fusion process, a reliability probability judgment will be made, and a reliable meteorological data source will be selected for data fusion.

[0117] The relationship between weather stations and road sections is determined as follows:

[0118] Let the coordinates of the weather station be ( , The effective radius of the data is r, and the coordinates of the starting and ending points of the road segment are respectively ( , ), ( , ), the calculation formula for the correlation degree k between the meteorological station and a certain flight segment is as follows:

[0119] ;

[0120]

[0121] If > r or , then k = 0;

[0122] If < r and , then + ;

[0123] After matching, a road segment can be associated with multiple meteorological stations with different k values.

[0124] Figure 4 shows a schematic diagram of the steps of road network matching according to an embodiment of the present application;

[0125] If Figure 4 as shown, when the road network matching module is in S2, according to the road network grid index and the road segment association library, the target personnel trajectory data is matched with the road network segments to obtain the trajectory point-road segment association relationship and the offset distance, and a road segment passing event is generated, including the following steps:

[0126] S21: Search for the road network grid index and the road segment association library according to the trajectory points of the trajectory data, match the road segments and calculate the trajectory point offset distance; obtain the target personnel trajectory data through the Kafka message queue;

[0127] S22: Improve the trajectory point events at the endpoints of adjacent road segments according to the trajectory point offset distance, and determine a series of trajectory point events of the target personnel trajectory data; the trajectory point events include the trajectory point position and the trajectory time;

[0128] Preferably, the interpolation method is used in the embodiment of the present application to calculate the specific time of entering and leaving the road segment to improve the trajectory point events at the endpoints of adjacent road segments.

[0129] S23: Generate a passing event corresponding to the target personnel trajectory data that includes the road segment ID, the entry time, and the departure time through a series of trajectory point arrays.

[0130] Among them, the specific steps for S21 to calculate the trajectory point offset distance include the following:

[0131] S211: Let the trajectory point be a(x0, y0), and the center line points of the two road segments closest to the grid where the trajectory point a(x0, y0) is located are p1(x1, y1) and p2(x2, y2) respectively;

[0132] S212: Calculate the offset distance of trajectory point a(x0,y0) using the perpendicular line method. The formula is:

[0133] ;

[0134] When t<0, ;

[0135] When t>0, ;

[0136] in, This is the offset of the reference point.

[0137] Figure 5 The diagram shows a schematic diagram of the road network matching process according to an embodiment of this application;

[0138] In practice, based on the real-time trajectory of the deliveryman, the road segment is matched through the trajectory point a(x0,y0). First, the grid where the trajectory point is located is searched to obtain its associated road segment. The centerline points p1(x1,y1) and p2(x2,y2) of the two closest road segments are obtained. The offset distance of a is obtained through the perpendicular line as follows, and the corresponding association relationship is stored.

[0139] like Figure 5 As shown, the processing flow is as follows: Receive the real-time trajectory point a(x0,y0) of the delivery person as input; Grid indexing: Call the pre-established gridded road network index system; Grid positioning: Determine the grid area to which the trajectory point belongs based on its coordinates; Path acquisition: Retrieve all candidate paths (segments) associated with the grid; Key point positioning: Find the two closest centerline reference points (p1, p2) on the candidate path; Offset calculation: Calculate the vertical offset distance from the trajectory point to the path using a geometric algorithm; Data storage: Save the mapping relationship between the trajectory point and the road segment, as well as the offset data.

[0140] In practice, route events are generated based on route segment matching. The route event includes the following fields: deliveryman number, route ID, route length, start time, and end time.

[0141] Figure 6 The flowchart of the generation path event according to an embodiment of this application is shown;

[0142] like Figure 6As shown, the process includes: 1) Start, perform initialization processing; 2) Match trajectory points to road segments, matching the current trajectory point to the corresponding road segment (e.g., using a spatial matching algorithm); 3) Move to the next trajectory point, read the next trajectory point in the trajectory sequence and continue processing; 4) Determine if the road segment has changed, check if the road segment matched by the current trajectory point is the same as the road segment of the previous trajectory point. If they are the same road segment, continue road segment matching and process the next trajectory point; if they are not the same road segment, trigger the event generation process; 5) Event generation process: When the road segment changes, calculate the departure time from the road segment, record the time of departure from the previous road segment (usually using the timestamp of the previous trajectory point); write the previous path event, generate a complete "departure from road segment" event record and store it; generate a new path event, create an "entry into road segment" event record for the current road segment; 6) Repeat the above steps in a loop until all trajectory points have been processed ("trajectory loop ends"), and terminate the process.

[0143] In the event generation process, when recording the time of "leaving the road segment" or "entering the road segment", because the time interval between trajectory points is relatively long, the deliveryman usually has already passed the starting point or the end point when entering the next road segment. Therefore, it is necessary to calculate the time of passing the starting point and the end point separately.

[0144] The embodiments of this application preferably use interpolation to calculate the specific times of entering and leaving the road segment.

[0145] Let the previous trajectory point be (a1, t1), the next trajectory point be (a2, t2), and the road segment boundary point be a. Then the formula for the time t between the end of the previous road segment and the start of the new road segment is as follows:

[0146] Based on the aforementioned offset distance calculation formula, the offset distance l1 corresponding to the starting point of the road segment a1 and the offset distance l2 corresponding to the starting point of the road segment a2 are obtained.

[0147] The distance from the starting point of the road segment to the road segment boundary point a is l. If points a1, a2 and a match different road segments, the distances are concatenated.

[0148] The journey time to the road segment boundary point a is obtained. for:

[0149] ;

[0150] The above calculations record the specific times when delivery drivers travel through the starting and ending points of a route. This time is more complete and accurate than the time reported in the trajectory report, enabling more precise trajectory analysis.

[0151] Figure 7 The diagram illustrates the data fusion steps according to an embodiment of this application;

[0152] like Figure 7As shown, in S3, meteorological station data with high reliability are selected and fused with the pathway events. This specifically includes the following steps:

[0153] S31: For the event time t0, calculate the reliability probability pᵢ of the associated weather station data:

[0154] ;

[0155] ;

[0156] Where ω1 and ω2 are weighting coefficients; m is the effective meteorological time.

[0157] S32: Select the meteorological station data with the highest reliability probability of the associated meteorological station and fuse it with the path events to obtain the path events including meteorological data.

[0158] In the preferred implementation, after generating the road segment transit event, the average speed of the road segment is also calculated. The delivery rider trajectory characteristics mainly involve analyzing the delivery speed of delivery riders on different road segments. The specific process of calculating the average speed of a delivery rider through a road segment is as follows:

[0159] If a road segment has n transit events within the entire time frame, and the transit event array is (segment ID, segment length l, entry time t1, end time t2), then the average speed v of the road segment can be calculated based on these transit events using the following formula:

[0160] V = .

[0161] Figure 8 The diagram shows a schematic of data fusion according to an embodiment of this application;

[0162] Delivery speed is greatly affected by the weather, and different road sections may be within the coverage of multiple weather stations. Different weather stations release data at different frequencies. Therefore, it is necessary to calculate the reliability probability of weather data from different sources based on the weather station correlation calculated in the preprocessing and the release time of the specific weather data, and select the weather data with the highest reliability probability for fusion.

[0163] like Figure 8As shown, the processing flow is as follows: 1) Start, initialize the data fusion process; 2) Loop through each path event, traversing each stored path event (such as entering / leaving a road segment); 3) Find the associated meteorological stations for the road segment, querying the associated meteorological monitoring stations based on the road segment where the current path event is located; 4) Determine the number of associated meteorological stations, check the number of associated meteorological stations, if it is not a single meteorological station, directly obtain the latest data of that meteorological station; if it is multiple meteorological stations, obtain the latest data of all associated meteorological stations, and then calculate the meteorological data with the highest confidence probability (selecting the optimal data through a weighting algorithm); 5) Data fusion, binding the selected / calculated meteorological data (such as temperature, precipitation) with the current path event; 6) Loop through and repeat the above steps until all path events are processed; 7) End and complete the data fusion process.

[0164] The process of calculating the credible probability is as follows:

[0165] First, calculate the Euclidean distance di between different data sources, and then calculate the confidence probability pi based on the Euclidean distance.

[0166] Suppose the delivery driver travels through a certain road segment for t0 days. This road segment is associated with two weather stations, with correlation degrees of t0 and t0 respectively. , Within the effective meteorological time range m (generally 1 hour), there are multiple dynamic data points, namely (meteorological station 1, and (weather station 2, If the distance is 1, then the corresponding Euclidean distance di is calculated using the formula:

[0167] ;

[0168] The formula for calculating the credible probability pi is:

[0169] ;

[0170] in and The weight configuration can be adjusted according to different scenarios to prioritize time and space. The default value is 0.5 for both.

[0171] If there are multiple dynamic data of the same type, the one with the larger Euclidean distance is selected and fused with the traffic flow feature data. After merging the different data, the final data format is (start time t1, end time t2, road segment index i, traffic flow speed v, temperature t, precipitation r, wind direction w, wind speed).

[0172] Finally, based on the final fused data, correlation analysis can be performed to analyze the factors influencing waterway traffic flow. The correlation analysis uses the Pearson coefficient formula to calculate the relationships between different factors.

[0173] Figure 9 The diagram illustrates the steps of association analysis according to an embodiment of this application;

[0174] like Figure 9 As shown, in the correlation analysis in S4, the correlation between road segment speed and meteorological factors is analyzed using the Pearson coefficient, including the following steps:

[0175] S41: For different road sections, there are N historical data points; meteorological factors include temperature, precipitation, and wind speed.

[0176] S42: Calculate the correlation coefficient r between average speed and meteorological factors, using the following formula:

[0177] r = Cov(X, Y) / (σX * σY);

[0178] Cov (X,Y)= Σ[(x-x̅) * (y-y̅)] / N;

[0179] Where X is the average speed; Y is the meteorological factor; σX and σY are the standard deviations of X and Y, respectively; x̅ is the average of multiple average speed records; and y̅ is the average of multiple meteorological factor records.

[0180] When r is close to 1, it indicates a positive correlation; when r is close to -1, it indicates a negative correlation.

[0181] The trajectory and meteorological data fusion analysis method of this application is adopted. By fusing trajectory data and meteorological data based on the road network topology, the meteorological data with the highest reliability probability is selected in the case of multi-source meteorological data. Accurate average speed analysis and correlation analysis are performed based on road segment, providing data support for deliverymen such as couriers or food deliverymen for order dispatch and route planning.

[0182] In addition, this application also achieves: 1) improving the efficiency of road network indexing by replacing coarse-grained regional statistics with road segment-level secondary indexes, and realizing accurate analysis of the road network; 2) optimizing the accuracy of meteorological data selection through a reliable probability model, and solving the problem of conflict between multi-source meteorological data; 3) supporting route optimization decision-making by integrating the impact of meteorology on delivery speed.

[0183] Example 2

[0184] This embodiment provides a trajectory and meteorological data fusion analysis system. For details not disclosed in this embodiment, please refer to the specific implementation of trajectory and meteorological data fusion analysis methods in other embodiments.

[0185] Figure 10The diagram shows a schematic representation of a trajectory and meteorological data fusion analysis system according to an embodiment of this application.

[0186] like Figure 10 As shown, an embodiment of this application provides a trajectory and meteorological data fusion analysis system, which includes:

[0187] The road network preprocessing module 10 is used to grid the target area based on the road network topology and establish a secondary index, and to perform segmentation processing and road network association on road segments, and to match and associate the meteorological stations in the target area with road segments to obtain the road network grid index, road segment association library and meteorological station association library;

[0188] The road network matching module 20 is used to match the target person trajectory data with the road network segments based on the road network grid index and the road segment association library, to obtain the trajectory point-road segment association relationship and offset distance, and to generate road segment path events;

[0189] The data fusion module 30 is used to select meteorological station data with high reliability and fuse them with the path events according to the meteorological station association database to obtain meteorological fusion data;

[0190] The correlation analysis module 40 is used to analyze the correlation between road segment speed and meteorological factors based on meteorological fusion data and the Pearson coefficient.

[0191] In practical implementation, this technical solution consists of a trajectory and meteorological data acquisition module, a road network preprocessing module, a road network matching module, a feature extraction module, a data fusion module, and a correlation analysis module.

[0192] The road network matching and data fusion modules can be processed in a distributed manner using computing tools such as Spark, with the Spark engine used for road network matching and data fusion. Other components utilize Kafka message queues to obtain trajectory and meteorological data streams for real-time data access.

[0193] The correlation analysis module outputs the Pearson correlation coefficient r between road segment speed and meteorological factors.

[0194] The distributed computing architecture of this application supports Spark distributed processing through modular design (road network matching, data fusion, etc.), thereby improving the throughput of massive data.

[0195] Therefore, the trajectory and meteorological data fusion analysis system of this application is adopted. By fusing trajectory data and meteorological data based on the road network topology, the meteorological data with the highest reliability probability is selected in the case of multi-source meteorological data. Based on road segmentation, accurate average speed analysis and correlation analysis are performed to provide data support for deliverymen such as couriers or food deliverymen for order dispatching and route planning.

[0196] Example 3

[0197] This embodiment provides a trajectory and meteorological data fusion analysis device. For details not disclosed in the trajectory and meteorological data fusion analysis device of this embodiment, please refer to the specific implementation content of the trajectory and meteorological data fusion analysis method or system in other embodiments.

[0198] Figure 11 The diagram shows a schematic of the structure of a trajectory and meteorological data fusion analysis device 400 according to an embodiment of this application.

[0199] like Figure 11 As shown, the trajectory and meteorological data fusion analysis device 400 includes: a storage unit 402 for storing executable instructions; and a processing unit 401 for connecting to the storage unit 402 to execute the executable instructions to complete the trajectory and meteorological data fusion analysis method.

[0200] Those skilled in the art will understand that the schematic diagram is merely an example of the trajectory and meteorological data fusion analysis device 400 and does not constitute a limitation on the trajectory and meteorological data fusion analysis device 400. It may include more or fewer components than shown in the diagram, or combine certain components, or different components. For example, the trajectory and meteorological data fusion analysis device 400 may also include input and output devices, network access devices, buses, etc.

[0201] The processing unit 401 (Central Processing Unit, CPU) can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the processing unit 401 can be any conventional processor. The processing unit 401 is the control center of the trajectory and meteorological data fusion analysis device 400, connecting all parts of the device through various interfaces and lines.

[0202] Storage unit 402 can be used to store computer-readable instructions. Processing unit 401 implements various functions of trajectory and meteorological data fusion analysis device 400 by running or executing computer-readable instructions or modules stored in storage unit 402 and calling data stored in storage unit 402. Storage unit 402 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of trajectory and meteorological data fusion analysis device 400, etc. In addition, storage unit 402 may include hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, read-only memory (ROM), random access memory (RAM), or other non-volatile / volatile storage devices.

[0203] If the integrated module of the trajectory and meteorological data fusion analysis device 400 is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by instructing related hardware through computer-readable instructions. The computer-readable instructions can be stored in a computer-readable storage medium, and when executed by a processor, the computer-readable instructions can implement the steps of the various method embodiments described above.

[0204] Example 4

[0205] This embodiment provides a computer-readable storage medium on which a computer program is stored; the computer program is executed by a processor to implement the trajectory and meteorological data fusion analysis method in other embodiments.

[0206] Using the trajectory and meteorological data fusion analysis device and computer-readable storage medium of this application, and employing the trajectory and meteorological data fusion analysis method of this application, trajectory data and meteorological data are fused based on the road network topology. In the case of multiple sources of meteorological data, the meteorological data with the highest reliability probability is selected, and accurate average speed analysis and correlation analysis are performed based on road segmentation. This provides data support for delivery personnel such as couriers or food delivery workers in order dispatching and route planning.

[0207] In addition, this application also achieves: 1) improving the efficiency of road network indexing by replacing coarse-grained regional statistics with road segment-level secondary indexes, and realizing accurate analysis of the road network; 2) optimizing the accuracy of meteorological data selection through a reliable probability model, and solving the problem of conflict between multi-source meteorological data; 3) supporting route optimization decision-making by integrating the impact of meteorology on delivery speed.

[0208] Those skilled in the art will understand that the terminology used in this invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular forms “a,” “the,” and “the” as used in this invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0209] It should be understood that although the terms first, second, third, etc., may be used in this invention to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first information may also be referred to as second information without departing from the scope of this invention, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to a determination."

[0210] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0211] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for fusion analysis of trajectory and meteorological data, characterized in that, Includes the following steps: The target area is gridded based on the road network topology and a secondary index is established. Road segments are segmented and associated with the road network. Meteorological stations in the target area are matched and associated with road segments to obtain a road network grid index, a road segment association database, and a meteorological station association database. Based on the road network grid index and road segment association library, the target personnel trajectory data is matched with the road network segments to obtain the trajectory point-road segment association relationship and offset distance, and to generate road segment path events; Based on the meteorological station association database, meteorological station data with high reliability are selected and fused with the path events to obtain meteorological fusion data; Based on meteorological fusion data, the correlation between road segment speed and meteorological factors was analyzed using the Pearson coefficient. The step of gridding the target area based on the road network topology and establishing a secondary index, and segmenting road segments and associating them with the road network, includes: The target area road network is divided into segments according to a preset length, and a secondary index containing the coordinates of the geometric center point of each segment and its offset distance is established. Based on grid division, the cross-association relationship between grids and road segments is established, and the grids and road segments with cross-association relationships are indexed to obtain the road network grid index and road segment association library; The step of matching target personnel trajectory data with road network segments based on the road network grid index and road segment association library to obtain trajectory point-road segment association and offset distance, and generating road segment path events, includes: The trajectory data is searched according to the trajectory points, the road network grid index and the road segment association library are searched, the road segments are matched and the trajectory point offset distance is calculated; Based on the trajectory point offset distance, the trajectory point events of adjacent road segment endpoints are improved, and a series of trajectory point events of the target person's trajectory data are determined; the trajectory point events include trajectory point location and trajectory time. Through the series of trajectory point events, a path event corresponding to the target person's trajectory data, including the road segment ID, entry time, and exit time, is generated.

2. The trajectory and meteorological data fusion analysis method according to claim 1, characterized in that, The process of matching and associating weather stations in the target area by road segment includes: The correlation degree k between the weather station and the road segment is calculated based on the weather station's location and effective radius, using the following formula: ； ； like > r or ,but =0; like < r and ,but + ; in and denoted as , where is the distance from the weather station to both ends of the road segment; 'r' is the effective radius of the weather station; , ( ) represents the coordinates of the weather station; , ), ( , These are the coordinates of the starting and ending points of the road segment, respectively. Based on the correlation degree k between meteorological stations and road segments, each road segment is associated with at least one meteorological station to obtain a meteorological station association database.

3. The trajectory and meteorological data fusion analysis method according to claim 1, characterized in that, The calculation of trajectory point offset distance includes: Let the trajectory point be a(x0,y0), and the centerline points of the two road segments closest to the trajectory point a(x0,y0) in the grid are p1(x1,y1) and p2(x2,y2), respectively; The offset distance of trajectory point a(x0,y0) is calculated using the perpendicular line method. The formula is: ; When t<0, ; When t>0, ; in, This is the offset of the reference point.

4. The trajectory and meteorological data fusion analysis method according to claim 1, characterized in that, The process of fusing meteorological station data with high reliability probability with the pathway events includes: Regarding the timing of events along the path Calculate the reliability probability of data from associated meteorological stations. : ； ； in, , is the weighting coefficient; m is the effective meteorological time. The meteorological station data with the highest reliability among the associated meteorological stations is selected and fused with the pathway events to obtain pathway events that include meteorological data.

5. The trajectory and meteorological data fusion analysis method according to claim 1, characterized in that, Following the generation of the road segment path event, the following is also included: If there are n path events within the entire time range of a road segment, the average speed v of the road segment can be calculated based on these path events using the following formula: V = ； Where t1 is the entry time and t2 is the end time.

6. The trajectory and meteorological data fusion analysis method according to claim 1, characterized in that, The analysis of the correlation between road segment speed and meteorological factors using the Pearson coefficient includes: For different road sections, N historical data points are provided; the meteorological factors include temperature, precipitation, and wind speed. The correlation coefficient r between average speed and meteorological factors is calculated using the following formula: r = Cov(X, Y) / (σX * σY); Cov (X,Y)= Σ[( ) * ( )] / N; Where X is the average speed; Y is the meteorological factor; σX and σY are the standard deviations of X and Y, respectively; The average of multiple average speed records; This represents the average value of multiple meteorological factor records. When r is close to 1, it indicates a positive correlation; when r is close to -1, it indicates a negative correlation.

7. A trajectory and meteorological data fusion analysis system, characterized in that, include: The road network preprocessing module is used to grid the target area based on the road network topology and establish a secondary index, and to segment and associate road segments. It also performs road segment matching and association of meteorological stations in the target area to obtain the road network grid index, road segment association library, and meteorological station association library. The step of gridding the target area based on the road network topology and establishing a secondary index, and segmenting road segments and associating them with the road network, includes: The target area road network is divided into segments according to a preset length, and a secondary index containing the coordinates of the geometric center point of each segment and its offset distance is established. Based on grid division, the cross-association relationship between grids and road segments is established, and the grids and road segments with cross-association relationships are indexed to obtain the road network grid index and road segment association library; The road network matching module is used to match the target person trajectory data with the road network segments based on the road network grid index and the road segment association library, to obtain the trajectory point-road segment association relationship and offset distance, and generate road segment path events; The step of matching target personnel trajectory data with road network segments based on the road network grid index and road segment association library to obtain trajectory point-road segment association and offset distance, and generating road segment path events, includes: The trajectory data is searched according to the trajectory points, the road network grid index and the road segment association library are searched, the road segments are matched and the trajectory point offset distance is calculated; Based on the trajectory point offset distance, the trajectory point events of adjacent road segment endpoints are improved, and a series of trajectory point events of the target person's trajectory data are determined; the trajectory point events include trajectory point location and trajectory time. Through the series of trajectory point events, a path event corresponding to the target person's trajectory data, including the road segment ID, entry time, and exit time, is generated; The data fusion module is used to select meteorological station data with high reliability based on the meteorological station association database and fuse it with the path events to obtain meteorological fusion data; The correlation analysis module is used to analyze the correlation between road segment speed and meteorological factors based on meteorological fusion data and the Pearson coefficient.

8. A device for fusing and analyzing trajectory and meteorological data. Its features are, Includes: a storage unit for storing executable instructions; And a processing unit, for connecting to the storage unit to execute executable instructions to complete the trajectory and meteorological data fusion analysis method as described in any one of claims 1-5.

Images