A mobile target behavior pattern trajectory simulation method

Abstract

The embodiment of the present disclosure provides a mobile target behavior mode track simulation method, which belongs to the technical field of data processing and specifically comprises the following steps: acquiring geographic information data in a target area; generating a geographic scene environment map based on the geographic information data; setting the type of a mobile target motion behavior mode and setting basic parameters of simulation track data accordingly; setting a specified motion behavior mode feature of the mobile target and setting track generation rule constraints accordingly; generating initial simulation track data corresponding to different mobile target behavior modes by using an A* algorithm based on the basic parameters and the track generation rule constraints; adding position noise and time noise to the initial simulation track data; obtaining multiple target simulation track data corresponding to the motion behavior mode of the mobile target in the geographic scene environment map and visualizing and storing the target simulation track data. The adaptability and automation of track data simulation are improved by the scheme of the present disclosure.

Description

Technical Field

[0001] This disclosure relates to the field of data processing technology, and in particular to a method for simulating the trajectory of a moving target's behavior pattern. Background Technology

[0002] With the development of satellite positioning technology and various sensor technologies, the activities of moving targets (vehicles, pedestrians) in urban space can be collected as continuous location point data, i.e., movement trajectory data, at a specific sampling frequency. This data reflects the daily activities and movement patterns of moving targets and can be used to explore the movement patterns of moving targets and analyze their behavioral patterns and social phenomena.

[0003] However, the current state of mobile target trajectory data has two main problems. First, most of the available mobile target trajectory data is held by the government or related companies (such as Gaode and Baidu). Due to the need to protect user privacy, the government or companies only release a limited amount of anonymized trajectory data. Second, the released trajectory data is incomplete and limited, making it difficult to cover a comprehensive range of mobile target movement patterns. Therefore, it has a significant impact on the study of mobile target movement patterns and behavioral patterns.

[0004] Mobile target trajectory data simulation refers to the virtual simulation of the movement of a mobile target in a city according to actual needs. It does not involve the user's real location, thus protecting the user's privacy, and can meet the requirements for data completeness and quantity. Therefore, it has become an effective means to solve the above problems.

[0005] Existing methods for simulating the motion trajectory data of moving targets can be divided into two categories: one is to manually simulate a certain amount of trajectory data. This method generates high-quality trajectory data, but it is time-consuming, labor-intensive, and inefficient, making it difficult to simulate large amounts of trajectory data; the other is to automatically generate the data using relevant simulation software. This method can simulate a large amount of trajectory data, but it lacks the ability to simulate trajectory data of urban moving targets under specified motion patterns under geographical constraints.

[0006] Overall, simulation of moving target trajectories is an effective method to resolve the contradiction between data privacy protection and data incompleteness and finiteness. However, existing moving target trajectory data simulation technologies lack methods that can automatically generate a large number of specified motion behavior patterns (one or more) and take into account geographical scene constraints.

[0007] It is evident that there is an urgent need for a highly adaptable and automated method for simulating the trajectory of moving target behavior patterns. Summary of the Invention

[0008] In view of this, the present disclosure provides a method for simulating the trajectory of a moving target's behavior pattern, which at least partially solves the problem of poor adaptability and automation in the prior art.

[0009] In a first aspect, embodiments of this disclosure provide a method for simulating the trajectory of a moving target's behavior pattern, including:

[0010] Step 1: Obtain geographic information data within the target area, wherein the geographic information data includes urban road network data, point of interest data, building data, terrain data, water system data, and administrative region data;

[0011] Step 2: Generate a geographic scene environment map based on geographic information data;

[0012] Step 3: Set the type of motion behavior mode of the moving target and set the basic parameters of the simulated trajectory data accordingly. The basic parameters include the number of moving targets, the number of trajectory points, the start and end positions of the motion trajectory, the sampling frequency, and the sampling interval.

[0013] Step 4: Set the specified motion behavior pattern features of the moving target and set trajectory generation rule constraints accordingly. The trajectory generation rule constraints include speed constraints, distance constraints, geometric shape constraints, relationship constraints between moving targets, and area access constraints.

[0014] Step 5: Based on the basic parameters and trajectory generation rule constraints, the A* algorithm is used to simulate and generate initial simulated trajectory data corresponding to different moving target behavior patterns;

[0015] Step 6: Add position and time noise from the trajectory sampling to the initial simulated trajectory data;

[0016] Step 7: Repeat steps 3 to 6 a preset number of times to obtain multiple simulated trajectory data of the moving target within the geographic scene environment map, and then visualize and store them.

[0017] According to a specific implementation of an embodiment of this disclosure, step 2 specifically includes:

[0018] Step 2.1: Project the geographic information data onto the same coordinate system.

[0019] Step 2.2: Process the urban road network data, extract the road centerlines, reconstruct the road topology, and perform topology checks and processing to eliminate road topology errors;

[0020] Step 2.3: Perform topology checks and processing on the building data to remove overlapping building data;

[0021] Step 2.4: Extract the row and column numbers and elevation values ​​of grid points from the terrain data, and save the data processed in steps 2.1 to 2.4 as vector data;

[0022] Step 2.5: Read the vector data needed to generate a map of the geographic scene environment and assign different colors to different features;

[0023] Step 2.6: Use OpenCV to rasterize the processed vector data into an image, set the background color of the image to white, and overlay different elements to obtain a map image containing different geographical scene elements, which is then used as a geographical scene environment map.

[0024] According to a specific implementation of this disclosure, before step 3, the method further includes:

[0025] Within the passable area, a point is randomly selected as the starting point. Then, a buffer zone with a specified radius is generated with the starting point as the center. Within the passable area of ​​the buffer zone, a point is randomly selected as the ending point.

[0026] According to a specific implementation of an embodiment of this disclosure, the step of setting the specified motion behavior pattern characteristics of the moving target includes:

[0027] Based on the specified motion behavior pattern of the moving target, the trajectory attribute field of the moving target is set, wherein the trajectory attribute field includes the position coordinates of the trajectory point, time, speed, elevation and direction.

[0028] According to a specific implementation of this disclosure, the area access constraint is to set the areas that a moving target can pass through and the areas that cannot pass through when moving on the map, and to set accessibility weights for different types of accessible areas.

[0029] According to a specific implementation of an embodiment of this disclosure, step 5 specifically includes:

[0030] Step 5.1: Using the basic parameters for generating the trajectory of the moving target and the map image data under the constraints of specified rules as input, the A* algorithm is used to generate the shortest path for the movement of the moving target with specified start and end points;

[0031] Step 5.2: Obtain the baseline trajectory data of the moving target based on the shortest path and basic parameters.

[0032] Step 5.3: Using the baseline trajectory data as a reference, generate motion trajectory data of other moving targets based on trajectory generation rule constraints, and combine them with the baseline trajectory data to form initial simulated trajectory data.

[0033] According to a specific implementation of an embodiment of this disclosure, step 5.1 specifically includes:

[0034] Step 5.1.1: Construct the adjacency list for A* path planning, and encode the pixels of the passable areas in the graph as nodes in the table. Adjacent nodes satisfy the 8-connectivity relationship.

[0035] Step 5.1.2: Mark the starting and ending points of the moving target's movement within the passable area;

[0036] Step 5.1.3: Initialize the open list and the closed list, and add the starting point to the open list;

[0037] Step 5.1.4: Calculate the g(n) and h(n) values ​​of the current node according to the user-defined rules, and calculate the f(n) value according to the preset formula. Select the node n with the smallest f(n) value in the open list for expansion, delete node n from the open list, and add node n to the closed list. If expanding the node causes the path to pass through an impassable area, abandon the expansion direction and instead select the node with the second smallest f(n) value in the open list for expansion, thereby avoiding planning to impassable areas. The preset formula is...

[0038] f(n) = g(n) + h(n)

[0039] Where f(n) is the priority from the current node to the destination, g(n) is the cost from the starting point to the current node, and h(n) is the heuristic cost from the node to the destination.

[0040] Step 5.1.5: For each neighboring node m of node n, if m is in the closed list, then ignore node m; if it is not in the open list, calculate the value of f(m) of node m and add m to the open list; if m is in the open list, compare the existing value of f(m) with the newly calculated value of f(m) and keep the smaller one.

[0041] Step 5.1.6: Repeat steps 5.1.4 and 5.1.5 until the open list is empty or the endpoint is expanded. If the endpoint is in the closed list, trace the parent node of the node in the closed list and push back from the endpoint to the starting point to obtain the node sequence, which is the shortest path for the target movement.

[0042] According to a specific implementation of an embodiment of this disclosure, step 5.3 specifically includes:

[0043] Based on the specified motion behavior pattern characteristics and trajectory generation rules, and using the baseline trajectory data as a template, the trajectory data of the remaining moving targets are generated point by point, so that the trajectory data conforms to the specified motion behavior pattern characteristics and is combined with the baseline trajectory data to form the initial simulated trajectory data.

[0044] According to a specific implementation of an embodiment of this disclosure, step 6 specifically includes:

[0045] Step 6.1: Add position noise to the initial simulated trajectory data by simulating the sampling position deviation of the initial simulated trajectory data using a Gaussian distribution. The expression for the sampling position deviation of the initial simulated trajectory data generated by the Gaussian distribution simulation is as follows:

[0046] x_=x+random.gauss(μ,σ)

[0047] y_=y+random.gauss(μ,σ)

[0048] Where x and y are the positions of the generated trajectory points, μ and σ are Gaussian distribution parameters, and random.gauss() is the Gaussian distribution function provided by Python;

[0049] Step 6.2: Add time noise to the initial simulated trajectory data by randomly deleting a preset number of trajectory point data to simulate missing sampling signals, or add time noise to the initial simulated trajectory data by randomly selecting a preset number of trajectory points and setting a time lag of a predetermined time to simulate weak sampling signals.

[0050] According to a specific implementation of an embodiment of this disclosure, the step of visualizing and storing the data includes:

[0051] The generated target simulation trajectory data corresponding to the specified motion mode is uniformly named and stored in CSV file format, and the generated target simulation trajectory data is visualized using ArcGIS or QGIS tools.

[0052] The simulation scheme for simulating the trajectory of a moving target behavior pattern in this embodiment includes: Step 1, acquiring geographic information data within the target area, wherein the geographic information data includes urban road network data, point of interest data, building data, terrain data, water system data, and administrative region data; Step 2, generating a geographic scene environment map based on the geographic information data; Step 3, setting the type of the moving target's motion behavior pattern and setting the basic parameters of the simulated trajectory data accordingly, wherein the basic parameters include the number of moving targets, the number of trajectory points, the start and end positions of the motion trajectory, the sampling frequency, and the sampling interval; Step 4, setting the specified motion behavior pattern features of the moving target and setting trajectory generation rule constraints accordingly, wherein the trajectory generation rule constraints include speed constraints, distance constraints, geometric shape constraints, relationships between moving targets, and area access constraints; Step 5, using the A* algorithm to simulate and generate initial simulated trajectory data corresponding to different moving target behavior patterns based on the basic parameters and trajectory generation rule constraints; Step 6, adding position noise and time noise to the initial simulated trajectory data; Step 7, repeating steps 3 to 6 a preset number of times to obtain multiple target simulated trajectory data corresponding to the moving target's motion behavior pattern within the geographic scene environment map and visualizing and storing them.

[0053] The beneficial effects of the embodiments of this disclosure are as follows: By setting certain rule constraints based on the multi-dimensional characteristics of the moving target and the specified motion behavior pattern and the geographical scene environment, the movement trajectory data of a specified number of urban moving targets with specified motion behavior patterns can be simulated and generated, thereby improving the adaptability and automation of trajectory data simulation. Attached Figure Description

[0054] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 A flowchart illustrating a method for simulating the trajectory of a moving target's behavior pattern, provided in an embodiment of this disclosure;

[0056] Figure 2 A schematic diagram of a moving target accompaniment mode provided in an embodiment of this disclosure;

[0057] Figure 3 Visualization results of simulation data on the accompaniment pattern of a moving target provided in the embodiments of this disclosure;

[0058] Figure 4 Visualization results of simulation data for the moving target search pattern provided in the embodiments of this disclosure. Detailed Implementation

[0059] The embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0060] The following specific examples illustrate the implementation of this disclosure. Those skilled in the art can easily understand other advantages and effects of this disclosure from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. This disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this disclosure. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0061] It should be noted that various aspects of embodiments within the scope of the appended claims are described below. It will be apparent that the aspects described herein can be embodied in a wide variety of forms, and any particular structure and / or function described herein is merely illustrative. Based on this disclosure, those skilled in the art will understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects set forth herein can be used to implement the device and / or practice the method. Additionally, this device and / or method can be implemented using structures and / or functionalities other than one or more of the aspects set forth herein.

[0062] It should also be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this disclosure. The drawings only show the components related to this disclosure and are not drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0063] Furthermore, specific details are provided in the following description to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the described aspects can be practiced without these specific details.

[0064] Simulation of moving target trajectories is an effective method to resolve the contradiction between data privacy protection and data incompleteness and finiteness. However, current moving target trajectory data simulation technology lacks a method that can automatically generate a large number of specified motion behavior patterns (one or more) and take into account geographical scene constraints.

[0065] This disclosure provides a method for simulating the trajectory of a moving target's behavior pattern, which can be applied to urban traffic management.

[0066] See Figure 1 This is a flowchart illustrating a method for simulating the trajectory of a moving target's behavior pattern, provided in an embodiment of this disclosure. Figure 1 As shown, the method mainly includes the following steps:

[0067] Step 1: Obtain geographic information data within the target area, wherein the geographic information data includes urban road network data, point of interest data, building data, terrain data, water system data, and administrative region data;

[0068] In practice, geographic information data within the target area can be obtained through the National Geographic Information Public Service Platform or open street view map websites. This geographic information data includes urban road network data, points of interest data, building data, terrain data, water system data, and administrative region data, etc., for subsequent steps.

[0069] Step 2: Generate a geographic scene environment map based on geographic information data;

[0070] Furthermore, step 2 specifically includes:

[0071] Step 2.1: Project the geographic information data onto the same coordinate system.

[0072] Step 2.2: Process the urban road network data, extract the road centerlines, reconstruct the road topology, and perform topology checks and processing to eliminate road topology errors;

[0073] Step 2.3: Perform topology checks and processing on the building data to remove overlapping building data;

[0074] Step 2.4: Extract the row and column numbers and elevation values ​​of grid points from the terrain data, and save the data processed in steps 2.1 to 2.4 as vector data;

[0075] Step 2.5: Read the vector data needed to generate a map of the geographic scene environment and assign different colors to different features;

[0076] Step 2.6: Use OpenCV to rasterize the processed vector data into an image, set the background color of the image to white, and overlay different elements to obtain a map image containing different geographical scene elements, which is then used as a geographical scene environment map.

[0077] In practice, the following steps can be used to generate a geographic scene environment map:

[0078] 1) Project all acquired data onto the same coordinate system to ensure the consistency of positional accuracy and correspondence between different types of data. In this embodiment, Gauss-Krüger is selected as the projection coordinate system.

[0079] 2) Process the urban road network data, extract the road centerline, reconstruct the road topology and perform topology checks and processing to eliminate road topology errors, such as road hanging points, pseudo nodes, and road line overlaps.

[0080] There are several ways to extract the road centerline. The method used here is based on ArcGIS. First, the vector linear road network data is converted into polygon data. Then, the dissolve tool is used to merge adjacent polygon features. Finally, the polygon feature to line feature tool is executed to obtain the road centerline vector data.

[0081] 3) Perform topology checks and processing on building data to ensure that there is no overlap between buildings;

[0082] 4) Process the terrain data, extract the row and column numbers and elevation values ​​of the grid points, and save them as vector data;

[0083] 5) Read the data such as road network, buildings, points of interest, water system, elevation, and administrative regions required to generate the geographic scene environment map, and assign different colors to different elements, such as gray for roads, blue for water system, red for buildings, and yellow for administrative regions;

[0084] 6) Use OpenCV to rasterize the processed vector data into an image. Set the background color of the image to white except for the geographic features. Overlay the images of different features to obtain a map image containing different geographic scene features and use it as a geographic scene environment map.

[0085] Step 3: Set the type of the moving target's motion behavior mode and set the basic parameters of the simulated trajectory data accordingly. The basic parameters include the number of moving targets, the number of trajectory points, the start and end positions of the motion trajectory, the sampling frequency, and the sampling interval.

[0086] Optionally, before step 3, the method further includes:

[0087] Within the passable area, a point is randomly selected as the starting point. Then, a buffer zone with a specified radius is generated with the starting point as the center. Within the passable area of ​​the buffer zone, a point is randomly selected as the ending point.

[0088] For example, this invention uses the adjoint mode as an example, and examples of adjoint modes are as follows: Figure 2 As shown.

[0089] It should be noted that any motion behavior pattern that can be described by multidimensional features and constraints can be generated by the method of this invention, including but not limited to accompanying patterns, aggregation patterns, separation patterns, lingering patterns, periodic patterns, and frequent patterns.

[0090] Then, set the basic parameters for the simulated trajectory data of the moving targets: including the number of moving targets, the number of trajectory points, the start and end positions of the motion trajectory, the sampling frequency, and the sampling distance interval. The number of moving targets is the number of vehicles or pedestrians participating in this motion mode trajectory simulation; the number of trajectory points is the number of trajectory points for each moving target's simulated trajectory data, which is an optional parameter; the start and end positions of the motion trajectory are the coordinates of the starting and ending points of each moving target's simulated trajectory, which is an optional parameter; the sampling frequency is the sampling time interval for trajectory generation; and the sampling distance interval is the distance between the initially generated trajectory points.

[0091] The trajectory start and end positions can be generated in two ways: user-specified or randomly generated. The random generation method involves first randomly selecting a point within the passable area as the starting point, then generating a buffer zone with a specified radius centered on this starting point, and finally randomly selecting a point within the passable area of ​​the buffer zone as the end point. The buffer radius can be set according to the trajectory generation requirements. This method aims to prevent the start and end points from being too far apart, which would result in a longer computation time.

[0092] In this example, the moving target is set as a moving pedestrian, and the number is 3; two points in the passable area of ​​the urban scene are randomly selected as the start and end points of the simulated trajectory data in the manner described above; the sampling frequency is 30 seconds and the sampling interval is 50 meters.

[0093] To avoid the path between the randomly selected start and end points being too long, the location of the end point is restricted to a buffer zone with a radius of 5km centered on the start point.

[0094] Step 4: Set the specified motion behavior pattern features of the moving target and set trajectory generation rule constraints accordingly. The trajectory generation rule constraints include speed constraints, distance constraints, geometric shape constraints, relationship constraints between moving targets, and area access constraints.

[0095] Optionally, the step of setting the specified motion behavior pattern characteristics of the moving target includes:

[0096] Based on the specified motion behavior pattern of the moving target, the trajectory attribute field of the moving target is set, wherein the trajectory attribute field includes the position coordinates of the trajectory point, time, speed, elevation and direction.

[0097] Optionally, the area access constraint is to set the areas that a moving target can pass through and the areas that cannot pass through when it moves on the map, and to set accessibility weights for different types of accessible areas.

[0098] In practice, the rules and constraints for the movement of targets in the accompanying pattern include that the speeds of the movements are similar, the distances are kept within a certain threshold, the directions of movement are similar, and the geometric shapes of the trajectories are similar.

[0099] Specifically, the area accessibility constraints for trajectory generation refer to setting the areas that a moving target can and cannot pass through on the map, and assigning accessibility weights to different types of passable areas. For example, assuming that a moving target cannot pass through buildings or waterways and tends to move on roads, the pixel values ​​of buildings and waterways can be set to infinity to represent impassable areas, the pixel values ​​of roads can be set to 1, and other pixel values ​​can be set to 2. The larger the pixel value, the lower the passability.

[0100] Step 5: Based on the basic parameters and trajectory generation rule constraints, the A* algorithm is used to simulate and generate initial simulated trajectory data corresponding to different moving target behavior patterns;

[0101] Based on the above embodiments, step 5 specifically includes:

[0102] Step 5.1: Using the basic parameters for generating the trajectory of the moving target and the map image data under the constraints of specified rules as input, the A* algorithm is used to generate the shortest path for the movement of the moving target with specified start and end points;

[0103] Step 5.2: Obtain the baseline trajectory data of the moving target based on the shortest path and basic parameters.

[0104] Step 5.3: Using the baseline trajectory data as a reference, generate motion trajectory data of other moving targets based on trajectory generation rule constraints, and combine them with the baseline trajectory data to form initial simulated trajectory data.

[0105] Furthermore, step 5.1 specifically includes:

[0106] Step 5.1.1: Construct the adjacency list for A* path planning, and encode the pixels of the passable areas in the graph as nodes in the table. Adjacent nodes satisfy the 8-connectivity relationship.

[0107] Step 5.1.2: Mark the starting and ending points of the moving target's movement within the passable area;

[0108] Step 5.1.3: Initialize the open list and the closed list, and add the starting point to the open list;

[0109] Step 5.1.4: Calculate the g(n) and h(n) values ​​of the current node according to the user-defined rules, and calculate the f(n) value according to the preset formula. Select the node n with the smallest f(n) value in the open list for expansion, delete node n from the open list, and add node n to the closed list. If expanding the node causes the path to pass through an impassable area, abandon the expansion direction and instead select the node with the second smallest f(n) value in the open list for expansion, thereby avoiding planning to impassable areas. The preset formula is...

[0110] f(n) = g(n) + h(n)

[0111] Where f(n) is the priority from the current node to the destination, g(n) is the cost from the starting point to the current node, and h(n) is the heuristic cost from the node to the destination.

[0112] Step 5.1.5: For each neighboring node m of node n, if m is in the closed list, then ignore node m; if it is not in the open list, calculate the value of f(m) of node m and add m to the open list; if m is in the open list, compare the existing value of f(m) with the newly calculated value of f(m) and keep the smaller one.

[0113] Step 5.1.6: Repeat steps 5.1.4 and 5.1.5 until the open list is empty or the endpoint is expanded. If the endpoint is in the closed list, trace the parent node of the node in the closed list and push back from the endpoint to the starting point to obtain the node sequence, which is the shortest path for the target movement.

[0114] Furthermore, step 5.3 specifically includes:

[0115] Based on the specified motion behavior pattern characteristics and trajectory generation rules, and using the baseline trajectory data as a template, the trajectory data of the remaining moving targets are generated point by point, so that the trajectory data conforms to the specified motion behavior pattern characteristics and is combined with the baseline trajectory data to form the initial simulated trajectory data.

[0116] In practice, the steps for generating initial simulated trajectory data corresponding to different moving target behavior patterns can be as follows:

[0117] S51 takes the basic parameters for generating the trajectory of the moving target and the map image data under the constraints of specified rules as input, and uses the A* algorithm to generate the shortest path for the movement of the moving target with specified start and end points.

[0118] Specifically, it includes the following sub-steps:

[0119] 1) Construct an adjacency list for A* path planning, and encode the pixels of passable areas in the graph as nodes in the list (e.g., roads, blank areas), with adjacent nodes satisfying 8-connectivity;

[0120] 2) Mark the starting and ending points of the moving target's movement on the map (ensuring they are within the passable area);

[0121] 3) Initialize the open list and the closed list, and add the starting point to the open list. The open list stores discovered nodes that have not yet been expanded, while the closed list is initially empty and stores expanded nodes.

[0122] 4) Calculate the g(n) and h(n) values ​​of the current node according to the rules set by the user, and calculate the f(n) value according to formula (1). Select the node n with the smallest f(n) value in the open list for expansion, delete node n from the open list, and add node n to the closed list. If expanding the node causes the path to pass through an impassable area, abandon the expansion direction and select the node with the second smallest f(n) value in the open list for expansion, thereby avoiding planning to an impassable area;

[0123] 5) For each neighboring node m of node n, if m is in the closed list, ignore node m; if it is not in the open list, calculate the value of f(m) of node m and add m to the open list; if m is in the open list, compare the existing value of f(m) with the newly calculated value of f(m) and keep the smaller one.

[0124] f(n)=g(n)+h(n) (1)

[0125] In the formula, f(n) is the total cost from the current node to the destination, g(n) is the cost from the starting point to the current node, and h(n) is the heuristic cost from the node to the destination. In this invention, the Dijkstra algorithm is used to calculate the shortest distance between two nodes as the value of h(n). The distance is Euclidean distance, as shown in formula (2).

[0126]

[0127] In the formula, (x1, y1) and (x2, y2) are the position coordinates of the two trajectory points.

[0128] 6) Repeat steps 4) and 5) until the open list is empty or the endpoint is expanded. If the endpoint is in the closed list, trace the parent node of the node in the closed list, and the sequence of nodes obtained by tracing back from the endpoint to the starting point is the shortest path for the target movement.

[0129] S52, based on the obtained shortest path and the set sampling frequency, sampling interval, etc., obtain the reference trajectory data of the moving target.

[0130] S53, using this baseline trajectory as a reference, generates simulated trajectory data for other moving targets based on rule constraints (such as distance constraints and velocity constraints between different targets).

[0131] Step 6: Add position and time noise from the trajectory sampling to the initial simulated trajectory data;

[0132] Based on the above embodiments, step 6 specifically includes:

[0133] Step 6.1: Add position noise to the initial simulated trajectory data by simulating the sampling position deviation of the initial simulated trajectory data using a Gaussian distribution. The expression for the sampling position deviation of the initial simulated trajectory data generated by the Gaussian distribution simulation is as follows:

[0134] x_=x+random.gauss(μ,σ)

[0135] y_=y+random.gauss(μ,σ)

[0136] Where x and y are the positions of the generated trajectory points, μ and σ are Gaussian distribution parameters, and random.gauss() is the Gaussian distribution function provided by Python;

[0137] Step 6.2: Add time noise to the initial simulated trajectory data by randomly deleting a preset number of trajectory point data to simulate missing sampling signals, or add time noise to the initial simulated trajectory data by randomly selecting a preset number of trajectory points and setting a time lag of a predetermined time to simulate weak sampling signals.

[0138] In practice, the sampling position deviation of the trajectory is generated through Gaussian distribution simulation, as shown in the following formula:

[0139] x_=x+random.gauss(μ,σ)

[0140] y_=y+random.gauss(μ,σ) (3)

[0141] In the formula, x and y are the positions of the generated trajectory points, μ and σ are Gaussian distribution parameters, μ is set to 0, σ is set to 20 meters, and random.gauss() is the Gaussian distribution function provided by Python.

[0142] The sampling time deviation is that a certain number of trajectory point data are randomly deleted to simulate a missing sampling signal, and a certain number of trajectory point data are randomly delayed for a certain period of time to simulate a weak sampling signal.

[0143] Step 7: Repeat steps 3 to 6 a preset number of times to obtain multiple simulated trajectory data of the moving target within the geographic scene environment map, and then visualize and store them.

[0144] Optionally, the step of visualizing and storing it includes:

[0145] The generated target simulation trajectory data corresponding to the specified motion mode is uniformly named and stored in CSV file format, and the generated target simulation trajectory data is visualized using ArcGIS or QGIS tools.

[0146] For example, steps 3 through 6 can be repeated 100 times. After completing the repetitions, multiple moving target accompaniment pattern trajectory data within the urban scene can be obtained. The generated trajectory data for the specified motion patterns can then be uniformly named and stored in CSV file format. Furthermore, the generated trajectory data can be visualized using tools such as ArcGIS or QGIS to intuitively verify the quality of the generated data. The visualization results of the moving target accompaniment pattern simulation data are shown below. Figure 3 As shown, similarly, when the mode is search mode, the simulation data visualization results of the moving target search mode are as follows: Figure 4 As shown.

[0147] Furthermore, to further enhance visualization, a mobile target motion pattern trajectory simulation system can be designed and developed. This system utilizes web development technology and adopts a B / S architecture, integrating functions such as geographic scene generation and loading, trajectory data simulation of several motion behavior patterns of mobile targets, visualization, and saving. In this embodiment, the simulated mobile target motion pattern data includes: accompanying mode, aggregation mode, separation mode, loitering mode, periodic mode, frequent mode, and patrol mode.

[0148] The mobile target behavior pattern trajectory simulation method provided in this embodiment can simulate and generate a specified number of urban mobile target movement trajectory data with specified movement behavior patterns by setting certain rule constraints based on the multi-dimensional features of the mobile target and the specified movement behavior pattern and the geographical scene environment, thereby improving the adaptability and automation of trajectory data simulation.

[0149] The units described in the embodiments of this disclosure can be implemented in software or in hardware.

[0150] It should be understood that the various parts of this disclosure can be implemented in hardware, software, firmware, or a combination thereof.

[0151] The above description is merely a specific embodiment of this disclosure, but the scope of protection of this disclosure is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this disclosure should be included within the scope of protection of this disclosure. Therefore, the scope of protection of this disclosure should be determined by the scope of the claims.

Claims

1. A method for simulating the trajectory of a moving target's behavior pattern, characterized in that, include: Step 1: Obtain geographic information data within the target area, wherein the geographic information data includes urban road network data, point of interest data, building data, terrain data, water system data, and administrative region data; Step 2: Generate a geographic scene environment map based on geographic information data; Step 3: Set the type of motion behavior mode of the moving target and set the basic parameters of the simulated trajectory data accordingly. The basic parameters include the number of moving targets, the number of trajectory points, the start and end positions of the motion trajectory, the sampling frequency, and the sampling interval. Step 4: Set the specified motion behavior pattern features of the moving target and set trajectory generation rule constraints accordingly. The trajectory generation rule constraints include speed constraints, distance constraints, geometric shape constraints, relationship constraints between moving targets, and area access constraints. Step 5: Based on the basic parameters and trajectory generation rule constraints, the A* algorithm is used to simulate and generate initial simulated trajectory data corresponding to different moving target behavior patterns; Step 6: Add position and time noise from the trajectory sampling to the initial simulated trajectory data; Step 7: Repeat steps 3 to 6 a preset number of times to obtain multiple simulated trajectory data of the moving target within the geographic scene environment map, and then visualize and store them.

2. The method according to claim 1, characterized in that... Step 2 specifically includes: Step 2.1: Project the geographic information data onto the same coordinate system. Step 2.2: Process the urban road network data, extract the road centerlines, reconstruct the road topology, and perform topology checks and processing to eliminate road topology errors; Step 2.3: Perform topology checks and processing on the building data to remove overlapping building data; Step 2.4: Extract the row and column numbers and elevation values ​​of grid points from the terrain data, and save the data processed in steps 2.1 to 2.4 as vector data; Step 2.5: Read the vector data needed to generate a map of the geographic scene environment and assign different colors to different features; Step 2.6: Use OpenCV to rasterize the processed vector data into an image, set the background color of the image to white, and overlay different elements to obtain a map image containing different geographical scene elements, which is then used as a geographical scene environment map.

3. The method according to claim 2, characterized in that... Before step 3, the method further includes: Within the passable area, a point is randomly selected as the starting point. Then, a buffer zone with a specified radius is generated with the starting point as the center. Within the passable area of ​​the buffer zone, a point is randomly selected as the ending point.

4. The method according to claim 3, characterized in that... The step of setting the specified motion behavior pattern characteristics of the moving target includes: Based on the specified motion behavior pattern of the moving target, the trajectory attribute field of the moving target is set, wherein the trajectory attribute field includes the position coordinates of the trajectory point, time, speed, elevation and direction.

5. The method according to claim 4, characterized in that... The area access constraints are defined as the areas that a moving target can pass through and the areas that cannot pass through when it moves on the map, and access weights are set for different types of accessible areas.

6. The method according to claim 5, characterized in that... Step 5 specifically includes: Step 5.1: Using the basic parameters for generating the trajectory of the moving target and the map image data under the constraints of specified rules as input, the A* algorithm is used to generate the shortest path for the movement of the moving target with specified start and end points; Step 5.2: Obtain the baseline trajectory data of the moving target based on the shortest path and basic parameters. Step 5.3: Using the baseline trajectory data as a reference, generate motion trajectory data of other moving targets based on trajectory generation rule constraints, and combine them with the baseline trajectory data to form initial simulated trajectory data.

7. The method according to claim 6, characterized in that... Step 5.1 specifically includes: Step 5.1.1: Construct the adjacency list for A* path planning, and encode the pixels of the passable areas in the graph as nodes in the table. Adjacent nodes satisfy the 8-connectivity relationship. Step 5.1.2: Mark the starting and ending points of the moving target's movement within the passable area; Step 5.1.3: Initialize the open list and the closed list, and add the starting point to the open list; Step 5.1.4: Calculate the g(n) and h(n) values ​​of the current node according to the user-defined rules, and calculate the f(n) value according to the preset formula. Select the node n with the smallest f(n) value in the open list for expansion, delete node n from the open list, and add node n to the closed list. If expanding the node causes the path to pass through an impassable area, abandon the expansion direction and instead select the node with the second smallest f(n) value in the open list for expansion, thereby avoiding planning to impassable areas. The preset formula is... f(n) = g(n) + h(n) Where f(n) is the priority from the current node to the destination, g(n) is the cost from the starting point to the current node, and h(n) is the heuristic cost from the node to the destination. Step 5.1.5: For each neighboring node m of node n, if m is in the closed list, then ignore node m; if it is not in the open list, calculate the value of f(m) of node m and add m to the open list; if m is in the open list, compare the existing value of f(m) with the newly calculated value of f(m) and keep the smaller one. Step 5.1.6: Repeat steps 5.1.4 and 5.1.5 until the open list is empty or the endpoint is expanded. If the endpoint is in the closed list, trace the parent node of the node in the closed list and push back from the endpoint to the starting point to obtain the node sequence, which is the shortest path for the target movement.

8. The method according to claim 7, characterized in that Step 5.3 specifically includes: Based on the specified motion behavior pattern characteristics and trajectory generation rules, and using the baseline trajectory data as a template, the trajectory data of the remaining moving targets are generated point by point, so that the trajectory data conforms to the specified motion behavior pattern characteristics and is combined with the baseline trajectory data to form the initial simulated trajectory data.

9. The method according to claim 8, characterized in that... Step 6 specifically includes: Step 6.1: Add position noise to the initial simulated trajectory data by simulating the sampling position deviation of the initial simulated trajectory data using a Gaussian distribution. The expression for the sampling position deviation of the initial simulated trajectory data generated by the Gaussian distribution simulation is as follows: x_=x+random.gauss(μ,σ) y_=y+random.gauss(μ,σ) Where x and y are the positions of the generated trajectory points, μ and σ are Gaussian distribution parameters, and random.gauss() is the Gaussian distribution function provided by Python; Step 6.2: Add time noise to the initial simulated trajectory data by randomly deleting a preset number of trajectory point data to simulate missing sampling signals, or add time noise to the initial simulated trajectory data by randomly selecting a preset number of trajectory points and setting a time lag of a predetermined time to simulate weak sampling signals.

10. The method according to claim 9, characterized in that... The steps of visualizing and storing it include: The generated target simulation trajectory data corresponding to the specified motion mode is uniformly named and stored in CSV file format, and the generated target simulation trajectory data is visualized using ArcGIS or QGIS tools.

Images