Pedestrian simulation initialization method and device based on OD matrix

Abstract

The invention discloses a pedestrian simulation initialization method and device based on an OD matrix. The method comprises the steps of obtaining scene structure data; laying simulation pedestrians according to the OD matrix by using the simulation model, and enabling the simulation pedestrians to walk from the corresponding pedestrian starting point to the corresponding pedestrian ending point; according to a preset frequency, pedestrian density data of each simulation pedestrian in a preset radius are dynamically collected; and judging whether there is a difference between the pedestrian density data at the current time and the pedestrian density data at the last acquisition time so as to confirm whether pedestrian simulation initialization is completed. According to the technical scheme provided by the invention, each simulation pedestrian has a clear and non-random starting point and terminal point, and whether the initialization of the pedestrian simulation is completed or not is not judged by subjective judgment of workers in the process, so that the initialization of the pedestrian simulation is more reasonable and has a basis.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a pedestrian simulation initialization method and device. Background Art

[0002] Micro-pedestrian simulation is a computer simulation technology that treats each pedestrian as an independent individual and simulates their movement behavior. This model plays an important role in predicting the performance of pedestrian flows in pedestrian areas or buildings.

[0003] Pedestrian simulation has its own unique characteristics, and requires micro-pedestrian simulation initialization or the completion of the "micro-pedestrian simulation initialization phase". For example, flood simulation starts from the first second, when the first drop of rain falls, which is the starting point of the simulation model. However, if we simulate the pedestrian situation during the morning rush hour at a transportation hub, we need to make the number of pedestrians in the model reach the number of people during the peak hour, and we need to make these pedestrians have reasonable positions and directions. These cannot be completed at the beginning of the pedestrian simulation model, and it takes a certain amount of time to achieve this goal.

[0004] We can call the process from the beginning of the pedestrian simulation model to the time when each pedestrian in the model has a reasonable position and direction as "micro-pedestrian simulation initialization". This stage of pedestrian simulation can be called the "micro-pedestrian simulation initialization stage".

[0005] Pedestrian simulation initialization is an important part of pedestrian simulation research. It involves how to set a reasonable starting state for the simulation model so as to more accurately simulate the dynamic behavior of pedestrians in the real world. However, in many practical application scenarios, pedestrian simulation initialization is not reasonable enough. First, the existence and necessity of pedestrian simulation initialization are not fully realized; second, there is no accurate division between the completion of initialization and the formal start of the simulation task, and the staff conducting the simulation rely on subjective experience to make judgments. All of these will directly affect the subsequent simulation results.

[0006] Therefore, how to design a pedestrian simulation initialization method so that the simulation model can obtain a reasonable starting state has become a problem that needs to be solved in this field. Summary of the invention

[0007] In view of this, in a first aspect, the present application proposes a pedestrian simulation initialization method based on an OD matrix, the method comprising:

[0008] Acquire scene structure data; the scene structure data includes: at least one pedestrian starting point and at least one pedestrian end point;

[0009] Using the simulation model to arrange simulated pedestrians according to the pedestrian types and the number of pedestrians of each type generated by the OD matrix, so that the simulated pedestrians walk from the corresponding pedestrian starting point to the corresponding pedestrian end point;

[0010] Dynamically collect pedestrian density data of each simulated pedestrian within a preset radius according to a preset collection frequency;

[0011] Determine whether there is a difference between the pedestrian density data at the current time and the pedestrian density data at the previous collection time;

[0012] If the difference exists, the simulated pedestrian continues walking to continue the simulation initialization;

[0013] If the difference does not exist, the simulated pedestrian completes the simulation initialization.

[0014] Preferably, the using of the simulation model to arrange simulated pedestrians according to the OD matrix includes:

[0015] Determine the proportion of pedestrian categories based on the simulation scene information;

[0016] The simulation model is used to arrange simulated pedestrians according to the OD matrix and the ratio of the pedestrian categories.

[0017] Further preferred:

[0018] Each pedestrian category corresponds to a different preset walking speed, so that the walking speed of the simulated pedestrian of each pedestrian category from the corresponding pedestrian starting point to the corresponding pedestrian end point is different.

[0019] Further preferably, the determining the proportion of pedestrian categories according to the simulation scene information includes:

[0020] Determining the proportion of pedestrian categories on each route according to the simulation scenario information; or

[0021] The proportion of pedestrian categories in the entire scene is determined according to the simulation scene information.

[0022] Preferably, the deployment of simulated pedestrians includes:

[0023] According to the preset deployment frequency, simulated pedestrians are deployed one by one until all simulated pedestrians are deployed; or

[0024] Deploy all simulated pedestrians at one time.

[0025] Preferably, the determining whether there is a difference between the pedestrian density data at the current time and the pedestrian density data at the previous collection time includes:

[0026] One-way ANOVA was used to determine whether there was a difference between the pedestrian density data at the current time and the pedestrian density data at the previous collection time.

[0027] In a second aspect, the present application further provides a pedestrian simulation initialization device based on an OD matrix, the device comprising:

[0028] An acquisition module is configured to acquire scene structure data; the scene structure data includes: at least one pedestrian starting point and at least one pedestrian end point;

[0029] A simulation module is configured to arrange simulated pedestrians according to the types of pedestrians and the number of people of each type given by the matrix generation module using the simulation model, so that the simulated pedestrians walk from the corresponding pedestrian starting point to the corresponding pedestrian end point;

[0030] The collection module is configured to dynamically collect pedestrian density data of each simulated pedestrian within a preset radius according to a preset collection frequency;

[0031] A judgment module is configured to judge whether there is a difference between the pedestrian density data at the current time and the pedestrian density data at the previous collection time;

[0032] If the difference exists, the simulated pedestrian continues to walk to continue the simulation for initialization;

[0033] If the difference does not exist, the simulated pedestrian completes the simulation initialization.

[0034] Preferably, the simulation module is also used for:

[0035] Determine the proportion of pedestrian categories based on the simulation scene information;

[0036] The simulation model is used to arrange simulated pedestrians according to the OD matrix and the ratio of the pedestrian categories.

[0037] Preferably, the simulation module is also used for:

[0038] Determining the proportion of pedestrian categories on each route according to the simulation scenario information; or

[0039] The proportion of pedestrian categories in the entire scene is determined according to the simulation scene information.

[0040] Preferably, the judgment module is also used for:

[0041] One-way ANOVA was used to determine whether there was a difference between the pedestrian density data at the current time and the pedestrian density data at the previous collection time.

[0042] The pedestrian simulation initialization method based on the OD matrix provided in the present application utilizes a simulation model to arrange simulated pedestrians according to the OD matrix, so that the simulated pedestrians walk from the corresponding pedestrian starting point to the corresponding pedestrian end point, and judges whether there is a difference between the pedestrian density data at the current time and the pedestrian density data at the previous collection time, so as to confirm whether the pedestrian simulation initialization is completed. In this process, each pedestrian has a clear, non-random starting point and end point, and does not rely on the subjective judgment of the staff whether the initialization of the pedestrian simulation is completed, thereby making the pedestrian simulation initialization more reasonable and reliable.

[0043] Other features and advantages of the present application will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The accompanying drawings constituting a part of the present application are used to provide a further understanding of the present application, and the schematic implementation modes and descriptions of the present application are used to explain the present application. In the accompanying drawings:

[0045] Figure 1 A method flow chart of a pedestrian simulation initialization method based on an OD matrix according to a preferred embodiment of the present application;

[0046] Figure 2 A schematic diagram of a simulation scenario of a preferred embodiment of the present application;

[0047] Figure 3 This is a schematic diagram of the preferred implementation mode of the present application for centrally deploying simulated pedestrians at the starting point at one time;

[0048] Figure 4 This is a schematic diagram of arranging simulated pedestrians one by one at a starting point according to a frequency in a preferred embodiment of the present application;

[0049] Figure 5 It is a structural schematic diagram of a pedestrian simulation initialization device based on an OD matrix according to a preferred embodiment of the present application. DETAILED DESCRIPTION

[0050] The technical solution of the present application will be described in detail below with reference to the accompanying drawings and in combination with the implementation modes.

[0051] In a first aspect, the present application provides a pedestrian simulation initialization method based on an OD matrix, such as Figure 1 As shown, the method includes steps 110-160.

[0052] Step 110, obtaining scene structure data;

[0053] Specifically, the scene structure data can be obtained according to the design drawings, including: at least one pedestrian starting point and at least one pedestrian end point, and the simulated pedestrian moves along the route from the pedestrian starting point to the pedestrian end point. In addition, the scene structure data can also include structural data such as walls and stairs. The scene simulated according to the scene structure data is closer to the real pedestrian activity place.

[0054] In a specific embodiment, Figure 2 The simulation scene shown in the figure includes five points "A, B, C, D, E", which can be set as the starting point or end point as needed. For example, the four points "A, B, C, D" can be used as both the starting point and the end point of the pedestrian. Of course, the starting point and the end point of the pedestrian cannot be the same point. "E" can be set as an "escape exit" and only serves as the end point of the simulated "escape" scene, not the end point of the simulated pedestrian when the pedestrian simulation is initialized.

[0055] Step 120, using the simulation model to arrange simulated pedestrians according to the OD matrix, so that the simulated pedestrians walk from the corresponding pedestrian starting point to the corresponding pedestrian end point;

[0056] Specifically, the origin-destination (OD) matrix is ​​a matrix that is sorted by rows (origin areas) and columns (destination areas), with the number of pedestrians (OD quantity) between any two partitions as elements. According to the OD matrix, a route for pedestrians to walk can be connected between each pedestrian origin and pedestrian destination, and the present application presets the number of pedestrians of each category in each route as needed, that is, the number of pedestrians of each category between each pedestrian origin and pedestrian destination, and then uses the simulation model to arrange simulated pedestrians according to the OD matrix with the number of pedestrians of each category.

[0057] In a specific embodiment, the OD matrix with four points "A, B, C, D" as the pedestrian starting point and pedestrian end point is shown in Table 1:

[0058] Table 1: OD matrix (unit: person)

[0059] Start / End Point A B C D A - 19 16 19 B 16 - 19 16 C 19 16 - 19 D 16 19 16 -

[0060] In Table 1, in a route with "A" as the starting point of pedestrian movement and "B" as the end point of pedestrian movement, the required number of pedestrians is "19 people", and so on, the total number of pedestrians in the matrix is ​​210 people.

[0061] In a preferred embodiment, when arranging pedestrians, it is necessary to determine the proportion of pedestrian categories based on the simulation scene information, and then determine the number of pedestrians of each category based on the proportion of pedestrian categories.

[0062] Specifically, in actual scenes, different types of pedestrians have different speeds. For example, teenagers are shorter, have faster walking speeds but shorter strides, while adult men are taller, have relatively faster walking speeds and relatively longer strides. Therefore, it is necessary to introduce pedestrian category information to distinguish the movement patterns of different types of pedestrians, so that the simulation effect is closer to the real situation. That is, each type of pedestrian category is set to a corresponding different preset walking speed, so that the simulated pedestrians of each type of pedestrian category have different walking speeds from the corresponding pedestrian starting point to the corresponding pedestrian end point.

[0063] Determining the proportion of pedestrian categories based on the simulation scene information can be understood as setting the personnel composition according to the characteristics of the scene to be simulated. The specific method of determining the proportion of pedestrian categories based on the simulation scene information can be understood as confirming the proportion of pedestrian categories for each route based on the simulation scene information, or confirming the proportion of pedestrian categories for the entire scene based on the simulation scene information. For example, in the simulation scene of the children's park, there are mostly teenagers, so the proportion of pedestrian categories for the entire scene can be confirmed based on the simulation scene information. When the simulation model arranges pedestrians according to the OD matrix, the pedestrian categories from each pedestrian starting point to the pedestrian end point (each route) follow this principle. Alternatively, in the simulation scene of the city park, there are mostly elderly people in the sports equipment area, and the personnel categories in other areas are balanced. The proportion of pedestrian categories for each route can be determined based on the simulation scene information, and the proportion of the elderly can be appropriately increased in specific routes in the sports equipment area.

[0064] In a specific embodiment, after combining the OD matrix and the ratio of pedestrian categories, the obtained pedestrian composition quantity table is shown in Table 2:

[0065] Table 2: Pedestrian composition table (people)

[0066] Serial number 1 2 3 4 5 6 7 8 9 10 11 12 starting point A A A B B B C C C D D D end B C D A C D A B D A B C Adolescent males 7 4 7 4 7 4 7 4 7 4 7 4 Teenage women 4 4 4 4 4 4 4 4 4 4 4 4 Adult male 4 4 4 4 4 4 4 4 4 4 4 4 Adult women 4 4 4 4 4 4 4 4 4 4 4 4 total 19 16 19 16 19 16 19 16 19 16 19 16

[0067] In Table 2, in the route with "A" as the starting point of pedestrian movement and "B" as the end point of pedestrian movement, the required number of teenage male pedestrians is "7 people", the number of teenage female pedestrians is "4 people", the number of adult male pedestrians is "4 people", the number of adult female pedestrians is "4 people", and so on. The total number of people in the matrix is ​​210.

[0068] In another preferred embodiment, when arranging simulated pedestrians, the simulated pedestrians can be arranged one by one according to the preset arrangement frequency according to the simulation needs until all simulated pedestrians are arranged, or all simulated pedestrians can be arranged at one time. Among them, the arrangement method of arranging simulated pedestrians one by one according to the preset arrangement frequency will make the simulated pedestrians relatively dispersed and can achieve a reasonable distribution in space more quickly. The arrangement method of arranging all simulated pedestrians at one time requires fewer computing resources and operation steps.

[0069] In a specific embodiment, the simulated pedestrian layout diagram is as follows: Figure 3 , 4 As shown, in Figure 4 In the example, the simulated pedestrians are placed one by one at the starting point according to the frequency. Figure 3 In the middle, it is the effect of deploying simulated pedestrians at the starting point all at once, which takes a certain amount of time to disperse.

[0070] In addition, the simulation model can be implemented using the Unity engine combined with the C Sharp programming language. In order to keep the number of pedestrians in the scene unchanged at the number given by the OD matrix, after the simulated pedestrian starts walking from its corresponding pedestrian starting point to the corresponding pedestrian end point, the pedestrian disappears, and then the simulated pedestrian needs to be regenerated at the starting point and continue walking along the route.

[0071] Step 130, dynamically collecting pedestrian density data of each simulated pedestrian within a preset radius according to a preset collection frequency;

[0072] Specifically, the application collects pedestrian density data within a certain radius of all pedestrians at a certain time interval, and judges whether the pedestrian state is unstable by comparing whether there is a significant difference between two groups of pedestrian density data at a certain time interval, thereby judging whether the simulated pedestrian initialization is completed. In a specific embodiment, the preset collection frequency is set to 10 seconds, and the preset radius is 1 meter, that is, the pedestrian density within a radius of 1m around each simulated pedestrian is calculated at intervals of 10 seconds, where the pedestrian density data is in units of people / square meter.

[0073] Step 140, determining whether there is a difference between the pedestrian density data at the current time and the pedestrian density data at the previous collection time;

[0074] Specifically, the present application uses a one-way analysis of variance method to determine whether there is a difference between the pedestrian density data at the current time and the pedestrian density data at the previous collection time.

[0075] One-Way ANOVA is a statistical method used to analyze whether there are significant differences between three or more independent groups divided by an independent variable. Among them, P-value is an important concept in ANOVA, which is used to determine whether the null hypothesis is true. In ANOVA, we first assume that all samples come from the same source, so there is no significant difference between the means of these samples. Then, we collect data and perform statistical analysis to determine whether the null hypothesis is true, and the P-value is a statistical indicator used to determine whether the null hypothesis is rejected.

[0076] It should be noted that when calculating the P value, SPSS software, GraphPad software, etc. can be used to give the corresponding results.

[0077] After the P value is determined, the P value is compared with the significance level. In the present application, the preset significance level is 0.05. If the P value is not greater than 0.05, it means that there is a difference between the pedestrian density data at the current time and the pedestrian density data at the previous collection time, and step 160 is executed; if the P value is greater than 0.05, it means that there is no difference between the pedestrian density data at the current time and the pedestrian density data at the previous collection time, and step 170 is executed.

[0078] Step 150, the simulated pedestrian continues to walk to continue the simulation initialization;

[0079] Specifically, if the P value obtained in the above steps is not greater than 0.05, it means that the pedestrian state is still unstable and the pedestrian simulation initialization process continues, then return to step 130 to continue collecting pedestrian density data and determine whether there is a difference in the pedestrian density data between the two moments, until there is no difference in the pedestrian density data between the two moments.

[0080] Step 160, completing pedestrian simulation initialization.

[0081] Specifically, if the P value obtained in the above steps is greater than 0.05, it means that the pedestrian state has stabilized, and the pedestrian simulation initialization is completed, and the next step can be performed.

[0082] In a second aspect, the present application further provides a pedestrian simulation initialization device based on an OD matrix, the device being used to implement the pedestrian simulation initialization method described in the first aspect above, the device comprising:

[0083] The acquisition module 210 is configured to acquire scene structure data; the scene structure data includes: at least one pedestrian starting point and at least one pedestrian end point;

[0084] The simulation module 220 is configured to use the simulation model to arrange the simulated pedestrians according to the result of the OD matrix generation module, so that the simulated pedestrians walk from the corresponding pedestrian starting point to the corresponding pedestrian end point;

[0085] The collection module 230 is configured to dynamically collect pedestrian density data of each simulated pedestrian within a preset radius according to a preset collection frequency;

[0086] The determination module 240 is configured to determine whether there is a difference between the pedestrian density data at the current time and the pedestrian density data at the previous collection time;

[0087] If there is a difference, the simulated pedestrian continues to walk to continue the simulation initialization;

[0088] If there is no difference, the simulated pedestrian completes initialization.

[0089] Preferably, the simulation module 220 is also used for:

[0090] Determine the proportion of pedestrian categories based on the simulation scene information;

[0091] The number of pedestrians of each category is calculated according to the ratio of the pedestrian categories.

[0092] Preferably, the simulation module 220 is further used for:

[0093] Determine the proportion of pedestrian categories on each route based on the simulation scenario information; or

[0094] The proportion of pedestrian categories in the entire scene is confirmed based on the simulation scene information.

[0095] Preferably, the determination module 250 is further used for:

[0096] One-way ANOVA was used to determine whether there was a difference between the pedestrian density data at the current time and the pedestrian density data at the previous collection time.

[0097] Other preferred embodiments of the pedestrian simulation initialization device provided in this application, the technical problems that can be solved and the technical effects that can be achieved are the same as those of the above-mentioned pedestrian simulation initialization method and will not be described in detail here.

[0098] The preferred embodiments of the present application are described in detail above; however, the present application is not limited to the specific details in the above embodiments. Within the technical concept of the present application, a variety of simple modifications can be made to the technical solution of the present application, and these simple modifications all fall within the protection scope of the present application.

[0099] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, this application will not further describe various possible combinations.

[0100] In addition, the various implementation modes of the present application may be arbitrarily combined, and as long as they do not violate the concept of the present application, they should also be regarded as the contents disclosed by the present invention.

Claims

1. A pedestrian simulation initialization method based on OD matrix, characterized in that: The method comprises: Acquire scene structure data; the scene structure data includes: at least one pedestrian starting point and at least one pedestrian end point; Using the simulation model to arrange simulated pedestrians according to the OD matrix, so that the simulated pedestrians walk from a corresponding pedestrian starting point to a corresponding pedestrian end point; Dynamically collect pedestrian density data of each simulated pedestrian within a preset radius according to a preset collection frequency; Determine whether there is a difference between the pedestrian density data at the current time and the pedestrian density data at the previous collection time; If the difference exists, the simulated pedestrian continues walking to continue the simulation initialization; If the difference does not exist, the simulated pedestrian completes the simulation initialization.

2. The method according to claim 1, characterized in that The method of using the simulation model to arrange the simulated pedestrians according to the OD matrix includes: Determine the proportion of pedestrian categories based on the simulation scene information; The simulation model is used to arrange simulated pedestrians according to the OD matrix and the ratio of the pedestrian categories.

3. The method according to claim 2, characterized in that: Each pedestrian category corresponds to a different preset walking speed, so that the simulated pedestrians of each pedestrian category have different walking speeds from the corresponding pedestrian starting point to the corresponding pedestrian end point.

4. The method according to claim 2, characterized in that: The determining the proportion of pedestrian categories according to the simulation scene information includes: Determining the proportion of pedestrian categories on each route according to the simulation scenario information; or The proportion of pedestrian categories in the entire scene is determined according to the simulation scene information.

5. The method according to claim 1, characterized in that The step of arranging simulated pedestrians comprises: According to the preset deployment frequency, simulated pedestrians are deployed one by one until all simulated pedestrians are deployed; or Deploy all simulated pedestrians at one time.

6. The method according to claim 1, characterized in that The determining whether there is a difference between the pedestrian density data at the current time and the pedestrian density data at the last collection time includes: One-way ANOVA was used to determine whether there was a difference between the pedestrian density data at the current time and the pedestrian density data at the previous collection time.

7. A pedestrian simulation initialization device based on OD matrix, characterized in that: The device comprises: An acquisition module is configured to acquire scene structure data; the scene structure data includes: at least one pedestrian starting point and at least one pedestrian end point; A simulation module is configured to arrange simulated pedestrians according to the OD matrix using a simulation model, so that the simulated pedestrians walk from a corresponding pedestrian starting point to a corresponding pedestrian end point; The collection module is configured to dynamically collect pedestrian density data of each simulated pedestrian within a preset radius according to a preset collection frequency; A judgment module is configured to judge whether there is a difference between the pedestrian density data at the current time and the pedestrian density data at the previous collection time; If the difference exists, the simulated pedestrian continues walking to continue the simulation initialization; If the difference does not exist, the simulated pedestrian completes the simulation initialization.

8. The device according to claim 7, characterized in that The simulation module is also used for: Determine the proportion of pedestrian categories based on the simulation scene information; The simulation model is used to arrange simulated pedestrians according to the OD matrix and the ratio of the pedestrian categories.

9. The device according to claim 8, characterized in that The simulation module is also used for: Determining the proportion of pedestrian categories on each route according to the simulation scenario information; or The proportion of pedestrian categories in the entire scene is determined according to the simulation scene information.

10. The device according to claim 7, characterized in that The judging module is also used for: One-way ANOVA was used to determine whether there was a difference between the pedestrian density data at the current time and the pedestrian density data at the previous collection time.

Images