Intercity comprehensive transportation track space-time discrete labeled generation method and system

Abstract

The invention discloses an inter-city comprehensive traffic travel trajectory space-time discrete labeled generation method and system. Simulated discrete travel trajectory data which is similar to actual travel trajectory data and provided with labels can be generated. Comprising the following steps: carrying out layering and gridding processing on infrastructures corresponding to different transportation modes; storing infrastructure data of all grids, and generating a complete grid base map data file; generating a single or complex traffic mode travel continuous track according to the grid base map; and simulating an actual travel trajectory data generation rule, and generating discrete travel trajectory data with labels. The method comprises a grid division and infrastructure mapping module, an infrastructure data storage module, a continuous travel trajectory generation module and a discrete travel trajectory data generation module. According to the method, by simulating the generation rule of the actual travel trajectory data, the generation of the discrete travel trajectory data with the labels is realized, and a new solution is provided for the problem of insufficient data with the labels in the travel trajectory data related research.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent transportation engineering technology, and specifically relates to a method and system for generating space-time discrete labeled inter-city integrated transportation travel trajectories. Background Art

[0002] As the foundation of the construction of a national comprehensive three-dimensional transportation network, understanding the travel patterns of residents within and between cities plays an important role in transportation planning, control and management. By analyzing and summarizing the travel patterns of residents, we can accurately identify the hot spots and main directions of traffic demand, provide scientific decision-making basis for relevant departments of traffic management, and realize scientific planning of the transportation network. However, the research area is too large, the residents' travel modes are diverse, and there are many routes to choose from, which makes the acquisition of residents' trajectory data a major problem, bringing great challenges to the study of residents' travel patterns.

[0003] Resident travel survey research has always been a widely studied field. Traditionally, the collection of resident travel information mainly relies on questionnaires and other forms, which can obtain relatively extensive and comprehensive information; however, there are serious problems in the efficiency and quality of data collection, such as low collection rate, waste of manpower and material resources, and inability to guarantee the validity of the data obtained. With the development of mobile communication technology, active data such as mobile phone communication data and GPS positioning data have become new data sources for some scholars to conduct travel research. This type of data has a wide coverage, a large user group, and a high research value, but this type of data still has limitations such as being able to be generated actively and requiring specific software.

[0004] Unlike mobile phone communication data and GPS positioning data, mobile phone signaling data that records the activities between mobile phones and communication base stations is passively generated and has a more comprehensive data volume. With the widespread popularity of mobile phones today, many scholars have begun to use mobile phone signaling data to conduct research in the field of resident travel, such as matching residents' travel trajectories or identifying travel modes. However, actual travel trajectory data such as mobile phone signaling does not include information on residents' travel methods. Although it can be manually labeled, the high cost has become a major problem that limits the application of actual travel trajectory data.

[0005] Considering the high cost of manually annotating travel trajectory data, we hope to explore a method to generate labeled simulated discrete travel trajectory data that can replace actual travel trajectory data, so that the actual travel trajectory data can be fully utilized without incurring excessive costs. Summary of the invention

[0006] In order to solve the above technical problems, the present invention proposes a method and system for generating spatiotemporal discrete labeled inter-city comprehensive transportation travel trajectories, so as to generate simulated discrete travel trajectory data with similar characteristics to actual travel trajectory data and with labels. To a certain extent, it can replace the actual travel trajectory data for use, making up for the high cost problem caused by the lack of labels or manual labeling of actual travel trajectory data.

[0007] To achieve the above object, the technical solution adopted by the present invention is:

[0008] The method for generating space-time discrete labeled inter-city comprehensive transportation travel trajectories includes the following steps:

[0009] Step S10, obtaining relevant geographic information data of the study area, dividing the study area into grids through ArcGis software, and realizing map mapping of various infrastructures;

[0010] Step S20, storing various infrastructure data and generating a complete grid base map of the study area;

[0011] Step S30, selecting a travel starting point and generating a continuous travel trajectory of a single traffic trip or a complex traffic trip;

[0012] Step S40, simulating the generation rules of actual travel trajectory data to generate simulated discrete travel trajectory data with labels.

[0013] As a further improvement of the present invention, step S10 specifically includes:

[0014] Step S11, the relevant geographic information data of the research area includes administrative division information and transportation infrastructure information that various transportation modes rely on, such as high-speed rail / conventional rail, expressways, national and provincial roads, and railway station and expressway toll station information. The geographic information data are all .shp files;

[0015] Step S12, selecting a suitable size to divide the study area into grids by ArcGis software, and numbering the grids as 1, 2, ..., n, where n is the number of grids;

[0016] Step S13, all traffic infrastructures are mapped to grids respectively through ArcGis software. If there is any infrastructure in the grid, it is stored as 1, otherwise it is stored as 0.

[0017] As a further improvement of the present invention, the step S20 specifically includes:

[0018] Step S21, storing own grid information;

[0019] Arrange different infrastructures in a certain order, and the corresponding infrastructure information in each grid is a string of numbers with a length equal to the number of infrastructure types and a value of 1 or 0. In order to reduce storage space, the infrastructure information of the grid is converted from a binary string of 1 / 0 to a decimal number. At this point, all the infrastructure information in each grid is stored as a decimal number.

[0020] Step S22, storing surrounding grid information;

[0021] When the infrastructure information of each grid is complete, in order to generate a continuous travel trajectory to achieve the continuity of travel trajectory data, the infrastructure information of the eight surrounding grids is retrieved and stored. The information finally stored by each grid includes the infrastructure information of itself and the eight surrounding grids, in the form of [num_i_1, num_i_2, num_i_3, num_i_4, num_i_5, num_i_6, num_i_7, num_i_8, num_i_9], where i is the number of the grid, and num_i_1-num_i_9 are the decimal numbers of the infrastructure information of the grids above left, above, above right, to the left, itself, to the right, below left, below, and below right of grid i.

[0022] Step S23, grid number conversion.

[0023] In order to facilitate the retrieval and verification of the mapping of different infrastructures, the grid numbers are converted into the form of (x_i, y_i) to indicate the row and column status of the grid, and finally a base map data dictionary file of the form {(x_1, y_1): [num_1_1, …, num_1_9], …, {(x_n, y_n): [num_n_1, …, num_n_9]} is formed.

[0024] As a further improvement of the present invention, the step S30 specifically includes:

[0025] Step S31, selecting a trajectory generation starting point;

[0026] Under the condition that all infrastructure information is complete in step S20, for different transportation modes, the starting point grid of the transportation mode is randomly selected according to its actual situation.

[0027] Step S32, generating a single traffic continuous travel trajectory;

[0028] For a certain mode of transportation, after obtaining the starting grid in step S31, the infrastructure conditions of the eight surrounding grids are obtained through the depth-first search method, and the next grid of the trajectory is randomly selected from the grids that allow this mode of transportation to pass, and the cumulative estimated trajectory length is calculated. From the grid data, it is statistically obtained that the distribution of the road length of each mode of transportation in the grid satisfies the normal distribution with a mean of 1 and a variance of 0.1. The grid corresponding to the trajectory length is executed according to this rule, and so on. When the expected distance of the travel of this mode of transportation is reached or there is no next trajectory point around, the search stops, the trajectory generation ends, and the grids that have been passed will not be selected repeatedly to exclude special cases such as circling. If the trajectory generation ends because there is no next trajectory point around, the trajectory is discarded and a new trajectory is generated.

[0029] Step S33, generating a complex traffic continuous travel trajectory;

[0030] An example of generating trajectories for traveling by two modes of transportation is given. The method for generating trajectories for traveling by more than two modes of transportation is similar. The first half and the second half of the trajectory for traveling by two modes of transportation are different modes of transportation. After obtaining the starting point by the first half of the travel mode, the infrastructure of the eight grids around it is obtained by the depth-first search method. The next grid of the trajectory is randomly selected from the grids that allow the first half of the travel mode to pass, and so on, until the expected distance of the first half of the travel mode is reached. After that, the mode conversion is performed. When selecting the next grid, the grids around that allow the first half of the travel mode to be converted to the second half of the travel mode will be searched. One of them will be randomly selected as the next grid of the trajectory. After that, the depth-first search method will be continued to search and obtain the remaining trajectory until the length of the second half of the trajectory meets the requirement. The search ends. If there is no mode conversion grid when the mode conversion is performed or the length of the second half of the trajectory does not meet the requirement because the next grid cannot be found, the trajectory will be discarded and regenerated until the condition is met.

[0031] As a further improvement of the present invention, the step S40 specifically includes:

[0032] Step S41, obtaining a travel track dot grid;

[0033] First, the starting point of the continuous trajectory is set as the starting point of the discrete travel trajectory data. Secondly, the travel speeds of different transportation modes are different. The travel distance of the transportation mode in this time period is calculated according to the operating speed of the corresponding transportation mode and the time interval of the trajectory data. The operating speed of the transportation mode is obtained by simulating the actual situation according to the actual acceleration / deceleration. By counting the time interval of the actual travel trajectory data, its distribution satisfies the normal distribution with a mean of 11 and a variance of 2. Then, the number of grid passes in this time period is obtained according to the road length distribution, so as to obtain the grid dot corresponding to the next discrete travel trajectory data, and record the corresponding cumulative time and transportation mode, and so on, until the end of the travel trajectory. If it is a multi-transportation mode travel trajectory with a mode conversion point, the mode conversion grid will be retained in the travel trajectory dot grid sequence.

[0034] Step S42, generating discrete travel trajectory data.

[0035] Each travel trajectory grid is assigned a number 1, 2, ..., m, where m is the number of travel trajectory grids; and the timestamp is calculated based on the accumulated time; in addition, since there is a deviation between the location of the actual travel trajectory data and the actual driving trajectory of the traveler, spatial disturbance is added to each travel trajectory grid, and x_i, y_i are converted to locx_i, locy_i, making the generated trajectory data more realistic. Thus, labeled simulated discrete travel trajectory data has been obtained, including: ID, timestamp time, longitude locx, latitude locy and corresponding transportation mode mode.

[0036] The present invention provides a system for generating a space-time discrete labelled inter-city integrated transportation travel trajectory, comprising a grid division and infrastructure mapping module, an infrastructure data storage module, a continuous travel trajectory generation module and a discrete travel trajectory data generation module, wherein:

[0037] The grid division and infrastructure mapping module is used to grid the study area and map the corresponding infrastructure to the corresponding grid, which is the base map data source used for subsequent trajectory generation;

[0038] The infrastructure data storage module is used to reasonably and efficiently store information of different transportation infrastructures to facilitate grid search;

[0039] The continuous travel trajectory generation module is used to generate travel trajectories for single-mode travel or multiple-mode travel, and is the basis for generating discrete travel trajectory data;

[0040] The discrete travel trajectory data generation module is used to generate and store simulated discrete travel trajectory data from the continuous travel trajectory according to the actual travel trajectory data dotting rule.

[0041] Beneficial effect: The inter-city integrated transportation travel trajectory spatiotemporal discrete labeled generation method and system proposed in the present invention can generate simulated discrete travel trajectory data with labels that can replace actual trajectory data, so that the actual trajectory data can be fully utilized without incurring excessively high costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a flow chart of the present invention;

[0043] Figure 2 is an architectural diagram of the grid partitioning and infrastructure mapping of the present invention;

[0044] Figure 3 is an architectural diagram of infrastructure data storage of the present invention;

[0045] Figure 4 It is a schematic diagram of the continuous travel trajectory generation of the present invention;

[0046] Figure 5 It is a schematic diagram of the architecture of discrete travel trajectory data generation of the present invention;

[0047] Figure 6 is a grid division and infrastructure mapping result diagram of the study area in an embodiment of the present invention;

[0048] Figure 7 It is a format display of single traffic discrete travel trajectory data generated in an embodiment of the present invention;

[0049] Figure 8 It is a graphic display of single traffic discrete travel trajectory data generated in an embodiment of the present invention;

[0050] Fig. 9 It is a format display of complex traffic discrete travel trajectory data generated in an embodiment of the present invention;

[0051] Fig.10 It is a graphic representation of complex traffic discrete travel trajectory data generated in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments:

[0053] The present invention is applicable to any area of ​​any size. As long as there is relevant administrative division information and various infrastructure information, corresponding simulated discrete travel trajectory data with labels can be generated. In the following embodiment, taking some areas of Jiangsu Province as an example, single traffic discrete travel trajectory data of expressways, national and provincial roads, high-speed railways, and ordinary railways are generated, as well as complex traffic discrete travel trajectory data of high-speed railways turning into national and provincial roads, ordinary railways turning into national and provincial roads, and expressways turning into national and provincial roads.

[0054] Example

[0055] like Figure 1 As shown, the inter-city integrated transportation travel trajectory time-space discrete label generation method and system of this embodiment include the following steps:

[0056] Step S10, obtaining relevant geographic information data of some areas of Jiangsu Province, dividing the study area into grids through ArcGis software, and realizing map mapping of various infrastructures;

[0057] Step S20, storing various infrastructure data and generating a complete grid base map of the study area;

[0058] Step S30, selecting a travel starting point and generating a continuous travel trajectory of a single traffic trip or a complex traffic trip;

[0059] Step S40, simulating the generation rules of actual travel trajectory data to generate simulated discrete travel trajectory data with labels.

[0060] like Figure 2 As shown, step S10 specifically includes:

[0061] Step S11, the relevant geographic information data of some areas of Jiangsu Province include administrative division information and transportation infrastructure information that various transportation modes rely on, such as high-speed rail / conventional rail, expressway, national and provincial roads and other road information, as well as railway station, expressway toll station and other site information, and the geographic information data are all .shp files;

[0062] Step S12, using ArcGis software to select a suitable size to grid the study area, and number the grids as 1, 2, ..., n, where n is the number of grids. In this embodiment, the length and width of the study area are gridded based on 1 km, and the number of grids n after division is 118664, the number of rows is 326, and the number of columns is 364;

[0063] Step S13, using ArcGis software to map all traffic infrastructure to grids. If there is infrastructure, it is stored as 1, otherwise it is stored as 0. The grid division and infrastructure mapping results of the study area are shown in Figure 6。

[0064] like Figure 3 As shown, step S20 specifically includes:

[0065] Step S21, storing own grid information;

[0066] Arrange different infrastructures in a certain order, and the corresponding infrastructure information in each grid is a string of numbers with a length equal to the number of infrastructure types and a value of 1 or 0. In order to reduce storage space, the infrastructure information of the grid is converted from a binary string of 1 / 0 to a decimal number. At this point, all the infrastructure information in each grid is stored as a decimal number.

[0067] Step S22, storing surrounding grid information;

[0068] When the infrastructure information of each grid is complete, in order to generate a continuous travel trajectory to achieve the continuity of travel trajectory data, the infrastructure information of the eight surrounding grids is retrieved and stored. The information finally stored by each grid includes the infrastructure information of itself and the eight surrounding grids, in the form of [num_i_1, num_i_2, num_i_3, num_i_4, num_i_5, num_i_6, num_i_7, num_i_8, num_i_9], where i is the number of the grid, and num_i_1-num_i_9 are the decimal numbers of the infrastructure information of the grids above left, above, above right, to the left, itself, to the right, below left, below, and below right of grid i.

[0069] Step S23, grid number conversion. In order to facilitate the retrieval and verification of the mapping of different infrastructures, the grid number is converted into the form of (x_i, y_i) to indicate the row and column of the grid. Finally, a base map data dictionary file of the form {(x_1, y_1): [num_1_1, ..., num_1_9], ..., {(x_n, y_n): [num_n_1, ..., num_n_9]} is formed.

[0070] like Figure 4 As shown, step S30 specifically includes:

[0071] Step S31, selecting a starting point for trajectory generation;

[0072] The starting points of the trajectories of different modes of transportation are usually in different places, such as the starting point of high-speed rail / conventional rail is the railway station, the starting point of expressway is the expressway toll station, and the starting point of national and provincial roads does not have a relatively fixed place. Therefore, under the condition of obtaining complete infrastructure information in step S20, for different modes of transportation, the starting point grid of the transportation mode is randomly selected according to its actual situation.

[0073] Step S32, generating a single traffic continuous travel trajectory;

[0074] For a certain mode of transportation, after obtaining the starting grid in step S31, the infrastructure of the eight surrounding grids is obtained through the depth-first search method, and the next grid of the trajectory is randomly selected from the grids that allow this mode of transportation to pass, and the cumulative estimated length of the trajectory is calculated. From the grid data, the distribution of the road length (unit: kilometers) of each mode of transportation in the grid satisfies the normal distribution with a mean of 1 and a variance of 0.1. The length of the trajectory corresponding to the grid is executed according to this rule. Similarly, the search stops when the expected distance of the mode of transportation is reached or there is no next trajectory point around, and the trajectory is generated. The track is finished, and the grids that have been passed will not be selected repeatedly to exclude special cases such as circles. Because there is no If the next trajectory point ends, the trajectory is discarded and a new trajectory is generated.

[0075] Step S33, generating a complex traffic continuous travel trajectory.

[0076] In the actual travel process, a travel trajectory often includes multiple travel modes. For this situation, this method gives an example of a trajectory generation method for completing travel through two modes of transportation. The trajectory generation method for completing travel through more than two modes of transportation is similar. The first half and the second half of the trajectory for completing travel by two modes of transportation are different modes of transportation. After obtaining the starting point by the first half of the travel mode, the infrastructure of the eight grids around it is obtained by the depth-first search method, and the next grid of the trajectory is randomly selected from the grids that allow the first half of the travel mode to pass. And so on, until the expected distance of the first half of the transportation mode is reached. After that, the mode conversion is performed, and when selecting the next grid, the grids around that allow the first half of the transportation mode to be converted to the second half of the transportation mode will be searched, and one of them will be randomly selected as the next grid of the trajectory. After that, the remaining trajectory is searched and obtained by the depth-first search method until the length of the second half of the trajectory meets the requirements, and the search ends. If there is no mode conversion grid when performing mode conversion or the length of the second half of the trajectory does not meet the requirements because the next grid cannot be found, the trajectory is discarded and regenerated until the conditions are met.

[0077] like Figure 5 As shown, step S40 specifically includes:

[0078] Step S41, obtaining a travel track dot grid;

[0079] First, the starting point of the continuous trajectory is set as the starting point of the discrete travel trajectory data. Secondly, the travel speeds of different transportation modes are different. The travel distance of the transportation mode in this time period is calculated according to the operating speed of the corresponding transportation mode and the time interval of the trajectory data. The operating speed of the transportation mode is obtained by simulating the actual situation according to the actual acceleration / deceleration; by counting the time interval of the actual travel trajectory data (unit: minutes), its distribution satisfies the normal distribution with a mean of 11 and a variance of 2. Then, according to the road length distribution, the number of grid passes in this time period is obtained, so as to obtain the grid points corresponding to the next discrete travel trajectory data, and record the corresponding cumulative time and transportation mode. And so on, until the end of the travel trajectory. If it is a multi-transportation mode travel trajectory with a mode conversion point, the mode conversion grid will be retained in the travel trajectory dot grid sequence.

[0080] Step S42, generating discrete travel trajectory data.

[0081] Each travel trajectory grid is assigned a number 1, 2, ..., m, where m is the number of travel trajectory grids; and the timestamp is calculated based on the accumulated time; in addition, since there is a deviation between the location of the actual travel trajectory data and the actual driving trajectory of the traveler, spatial disturbance is added to each travel trajectory grid, converting (x_i, y_i) to (locx_i, locy_i), making the generated trajectory data more realistic. Thus, labeled simulated discrete travel trajectory data has been obtained, including: ID, timestamp (time), longitude (locx), latitude (locy) and corresponding transportation mode (mode).

[0082] This embodiment generates discrete single-traffic travel trajectory data for 30 highways, national and provincial roads, high-speed railways, and conventional railways, as well as discrete single-traffic travel trajectory data for each of the 30 highways, national and provincial roads, high-speed railways, and conventional railways. Generate 10 complex traffic discrete trips from high-speed rail to national and provincial roads, from ordinary rail to national and provincial roads, and from expressway to national and provincial roads Number of tracks The generated single traffic discrete travel trajectory data format of expressway (GG), national and provincial highway (GSD), high-speed rail (TG), and ordinary rail (TS) is shown in Figure 7 , for a real map display, see Figure 8 The generated complex traffic discrete travel trajectory data formats of high-speed rail to national and provincial roads (TG-GSD), conventional rail to national and provincial roads (TS-GSD), and expressway to national and provincial roads (GG-GSD) are shown in Fig. 9 , for a real map display, see Fig.10 .

[0083] The above description is only a preferred embodiment of the present invention and does not constitute any other form of limitation to the present invention. Any modification or equivalent change made based on the technical essence of the present invention still falls within the scope of protection required by the present invention.

Claims

1. A method for generating space-time discrete labeled inter-city comprehensive transportation travel trajectories, characterized by: The following steps are involved: Step S10, obtaining relevant geographic information data of the study area, dividing the study area into grids through ArcGis software, and realizing map mapping of various infrastructures; Step S20, storing various infrastructure data and generating a complete grid base map of the study area; Step S30, selecting a travel starting point and generating a continuous travel trajectory of a single traffic trip or a complex traffic trip; Step S40, simulating the generation rules of actual travel trajectory data to generate simulated discrete travel trajectory data with labels.

2. The method for generating time-space discrete labeled inter-city integrated transportation travel trajectories according to claim 1 is characterized by: The step S10 specifically includes: Step S11, the relevant geographic information data of the research area includes administrative division information and transportation infrastructure information that various transportation modes rely on, such as high-speed rail / conventional rail, expressway, national and provincial road information, and railway station and expressway toll station site information. The geographic information data are all .shp files; Step S12, selecting a suitable size to divide the study area into grids by ArcGis software, and numbering the grids as 1, 2, ..., n, where n is the number of grids; Step S13, all traffic infrastructures are mapped to grids respectively through ArcGis software. If there is any infrastructure in the grid, it is stored as 1, otherwise it is stored as 0.

3. The method for generating time-space discrete labeled inter-city integrated transportation travel trajectories according to claim 1 is characterized by: The step S20 specifically includes: Step S21, storing own grid information; Arrange different infrastructures in a certain order, and the corresponding infrastructure information in each grid is a string of numbers with a length equal to the number of infrastructure types and a value of 1 or 0. In order to reduce storage space, the infrastructure information of the grid is converted from a binary string of 1 / 0 to a decimal number. At this point, all the infrastructure information in each grid is stored as a decimal number; Step S22, storing surrounding grid information; When the infrastructure information of each grid is complete, in order to generate a continuous travel trajectory to achieve the continuity of travel trajectory data, the infrastructure information of the eight surrounding grids is retrieved and stored. The information finally stored in each grid includes its own infrastructure information and the eight surrounding grids, in the form of [num_i_1, num_i_2, num_i_3, num_i_4, num_i_5, num_i_6, num_i_7, num_i_8, num_i_9], where i is the number of the grid, and num_i_1-num_i_9 are the decimal numbers of the infrastructure information of the grids to the upper left, directly above, upper right, directly left, itself, directly right, lower left, directly below, and lower right of grid i respectively; Step S23, grid number conversion; In order to facilitate the retrieval and verification of the mapping of different infrastructures, the grid numbers are converted into the form of (x_i, y_i) to indicate the row and column status of the grid, and finally a base map data dictionary file of the form {(x_1, y_1): [num_1_1, …, num_1_9], …, {(x_n, y_n): [num_n_1, …, num_n_9]} is formed.

4. The method for generating time-space discrete labeled inter-city integrated transportation travel trajectories according to claim 1 is characterized by: The step S30 specifically includes: Step S31, selecting a trajectory generation starting point; Under the condition of obtaining complete infrastructure information in step S20, for different transportation modes, the starting point grid of the transportation mode is randomly selected according to its actual situation; Step S32, generating a single traffic continuous travel trajectory; For a certain mode of transportation, after obtaining the starting grid in step S31, the infrastructure of the eight grids around it is obtained through the depth-first search method, and the next grid of the trajectory is randomly selected from the grids that allow this mode of transportation to pass, and the cumulative estimated trajectory length is calculated. From the grid data, it is statistically obtained that the distribution of the road length of each mode of transportation in the grid satisfies the normal distribution with a mean of 1 and a variance of 0.

1. The grid corresponding to the trajectory length is executed according to this rule, and so on. When the expected distance of the mode of transportation is reached or there is no next trajectory point around, the search stops, the trajectory generation ends, and the grids that have been passed will not be repeatedly selected to exclude special cases such as circling. If the trajectory generation ends because there is no next trajectory point around, the trajectory is discarded and a new trajectory is generated; Step S33, generating a complex traffic continuous travel trajectory; An example of generating trajectories for completing travel by two modes of transportation is given. The method for generating trajectories for completing travel by more than two modes of transportation is similar. The first half and the second half of the trajectory for completing travel by two modes of transportation are different modes of transportation. After obtaining the starting point by the first half of the travel mode, the infrastructure conditions of the eight surrounding grids are obtained by the depth-first search method. The next grid of the trajectory is randomly selected from the grids that allow the first half of the travel mode to pass, and so on, until the expected distance of the first half of the travel mode is reached. After that, the mode is switched. When selecting the next grid, the grids around that allow the first half of the travel mode to be switched to the second half of the travel mode are searched. One of them is randomly selected as the next grid of the trajectory. After that, the depth-first search method is continued to search and obtain the remaining trajectory until the length of the second half of the trajectory meets the requirement and the search ends. If there is no mode switching grid when the mode is switched or the length of the second half of the trajectory does not meet the requirement because the next grid cannot be found, the trajectory is discarded and regenerated until the condition is met.

5. The method for generating time-space discrete labeled inter-city integrated transportation travel trajectories according to claim 1 is characterized by: The step S40 specifically includes: Step S41, obtaining a travel trajectory dot grid; First, the starting point of the continuous trajectory is set as the starting point of the discrete travel trajectory data. Secondly, the travel speeds of different transportation modes are different. The travel distance of the transportation mode in this time period is calculated according to the operating speed of the corresponding transportation mode and the time interval of the trajectory data. The operating speed of the transportation mode is obtained by simulating the actual situation according to the actual acceleration / deceleration. By counting the time interval of the actual travel trajectory data, its distribution satisfies the normal distribution with a mean of 11 and a variance of 2. Then, according to the road length distribution, the number of grid passes in this time period is obtained, so as to obtain the grid dot corresponding to the next discrete travel trajectory data, and record the corresponding cumulative time and transportation mode, and so on, until the end of the travel trajectory. If it is a multi-transportation mode travel trajectory with a mode conversion point, the mode conversion grid will be retained in the travel trajectory dot grid sequence; Step S42, generating discrete travel trajectory data; Each travel trajectory grid is numbered 1, 2, …, m, where m is the number of travel trajectory grids; and the timestamp is calculated based on the accumulated time. In addition, since there is a deviation between the position of the actual travel trajectory data and the actual driving trajectory of the traveler, spatial disturbance is added to each travel trajectory grid, and x_i, y_i are converted into locx_i, locy_i to make the generated trajectory data more realistic. Thus, labeled simulated discrete travel trajectory data has been obtained, including: ID, timestamp time, longitude locx, latitude locy and corresponding transportation mode mode.

6. The system for generating a time-space discrete labeled method for inter-city integrated transportation travel trajectory according to any one of claims 1 to 5, characterized in that: It includes a grid division and infrastructure mapping module, an infrastructure data storage module, a continuous travel trajectory generation module and a discrete travel trajectory data generation module, wherein: The grid division and infrastructure mapping module is used to grid the study area and map the corresponding infrastructure to the corresponding grid, which is the base map data source used for subsequent trajectory generation; The infrastructure data storage module is used to reasonably and efficiently store information of different transportation infrastructures to facilitate grid search; The continuous travel trajectory generation module is used to generate travel trajectories for single-mode travel or multiple-mode travel, and is the basis for generating discrete travel trajectory data; The discrete travel trajectory data generation module is used to generate and store simulated discrete travel trajectory data from the continuous travel trajectory according to the actual travel trajectory data dotting rule.

Images