Crowd behavior simulation method, device, electronic device and storage medium

Abstract

The present invention provides a crowd behavior simulation method, device, electronic device and storage medium, the method comprising: determining a target perception domain and a target relative orientation of a target pedestrian; determining the target pedestrian's share of the avoidance speed based on the target perception domain and the target relative orientation; determining a back-off avoidance half-plane line based on the target pedestrian's expected speed, the back-off avoidance half-plane line being used to limit the target pedestrian's feasible speed to be located in the feasible solution region of the back-off avoidance half-plane line; and simulating crowd behavior through an ORCA model based on the share ratio and the back-off avoidance half-plane line. The present invention improves the basic ORCA model by introducing a perception domain and a back-off avoidance configuration, and can stably reproduce the "arch" crowding phenomenon, with sequential stability.

Description

Technical Field

[0001] The present invention relates to the field of electronic technology, and in particular to a crowd behavior simulation method, device, electronic equipment and storage medium. Background Art

[0002] Crowd behavior simulation is to study the behavioral characteristics and laws of crowds in various scenarios, establish a simulation model of crowd behavior, and realistically display the crowd behavior simulation process in a virtual environment.

[0003] At present, when electronic devices are used to simulate crowd behavior, the Optimal Reciprocal Collision Avoidance (ORCA) model is one of the classic micro crowd behavior simulation methods. The basic ORCA model aims to prevent collisions between pedestrians and between pedestrians and obstacles, fully considers the distribution of avoidance responsibilities between pedestrians, solves the relatively optimal travel speed, and then simulates the movement of the crowd. The basic ORCA model can achieve collision avoidance between multiple moving objects without central coordination when only limited motion information is perceived. It has been widely used in many fields such as micro crowd simulation, robots, and drones.

[0004] When the basic ORCA model simulates congested crowds, pedestrians will experience disorder and instability due to their unconditional avoidance of each other. It is difficult to reproduce classic crowd self-organization phenomena such as the "arch effect", and the simulation results are inconsistent with the actual situation.

[0005] In the prior art, a stopping rule is proposed, that is, when a pedestrian's speed is detected to be opposite to the speed at the previous moment, its speed is forcibly set to 0 to avoid disrupting the trajectory due to the pedestrian's retreat. This stopping rule is a forced injection outside the basic ORCA model, which destroys the integrity of the basic ORCA model calculation system. As time goes by, it is easy to cause pedestrians to enter a stagnant state one by one, which still causes the simulation results to be inconsistent with the actual situation. Summary of the invention

[0006] The present invention provides a crowd behavior simulation method, device, electronic device and storage medium, which are used to solve the problem that the simulation results of the basic ORCA model in the prior art are inconsistent with the actual situation, and realize that the improved ORCA model can stably reproduce the "arch" crowding phenomenon and has order stability.

[0007] In a first aspect, the present invention provides a crowd behavior simulation method, comprising:

[0008] Determine the target perception domain and target relative position of the target pedestrian;

[0009] Determining a proportion of the target pedestrian's contribution to the avoidance speed based on the target perception domain and the target relative position;

[0010] Based on the desired speed of the target pedestrian, determining a back-up avoidance half-plane line, wherein the back-up avoidance half-plane line is used to limit the feasible speed of the target pedestrian to be within a feasible solution region of the back-up avoidance half-plane line;

[0011] Based on the sharing ratio and the retreat avoidance half-plane line, crowd behavior simulation is performed through the ORCA model;

[0012] Among them, the target perception domain includes the front visible area of ​​the target pedestrian and the rear perception area of ​​the target pedestrian; the target relative position is used to indicate the relative position information of the target pedestrian and the adjacent pedestrians; the avoidance speed is the speed required for the target pedestrian to avoid collision with the adjacent pedestrians.

[0013] Optionally, according to a crowd behavior simulation method provided by the present invention, the determining the target perception domain and the target relative position of the target pedestrian includes:

[0014] Determining a target perception domain of the target pedestrian based on an equation of a forward visible area of ​​the target pedestrian and an equation of a rearward perception area of ​​the target pedestrian;

[0015] The equation of the front visible area is:

[0016]

[0017] The equation for the rear perception area is:

[0018]

[0019] Where A represents the target pedestrian, T A represents the target position of the target pedestrian, p A represents the current position of the target pedestrian, d A represents the front vector of the target pedestrian, r A represents the visible radius of the target pedestrian, θ A represents the sight range of the target pedestrian, p B represents the current position of the neighboring pedestrian, γ A Represents the rear perception radius of the target pedestrian.

[0020] Optionally, according to a crowd behavior simulation method provided by the present invention, the step of determining the retreat avoidance half-plane line based on the expected speed of the target pedestrian includes:

[0021] Based on the equation of the backsliding avoidance half-plane line, the backsliding avoidance half-plane line is determined; wherein the equation of the backsliding avoidance half-plane line is:

[0022]

[0023] Where A represents the target pedestrian, v1 represents the velocity vector that meets the conditions, The unit vector representing the expected velocity of the target pedestrian at time t, ORCA f Denotes that the retreat avoids the half-plane line.

[0024] Optionally, according to a crowd behavior simulation method provided by the present invention, the crowd behavior simulation is performed through an ORCA model based on the sharing ratio and the retreat avoidance half-plane line, including:

[0025] Determining a set of interpersonal collision avoidance half-plane lines based on the avoidance speed, the sharing ratio, the actual speed of the target pedestrian, and the actual speed of the adjacent pedestrians;

[0026] Determining a speed feasible solution region based on the set of interpersonal collision avoidance half-plane lines, the set of obstacle collision avoidance half-plane lines, and the retreat avoidance half-plane lines;

[0027] The speed feasible solution region is used to determine the target speed of the target pedestrian.

[0028] Optionally, according to a crowd behavior simulation method provided by the present invention, determining the proportion of the target pedestrian's share of the avoidance speed based on the target perception domain and the target relative position includes:

[0029] In a first perception situation where the target perception domain and the target relative position indicate a first perception, the sharing ratio is set to a preset ratio value;

[0030] In the second perception situation where the target perception domain and the target relative position indicate a second perception, the sharing ratio is set to 1;

[0031] Among them, the first perception situation is a situation where the target pedestrian and the adjacent pedestrian can perceive each other, and the second perception situation is a situation where the target pedestrian can perceive the adjacent pedestrian unidirectionally.

[0032] Optionally, according to a crowd behavior simulation method provided by the present invention, determining a set of interpersonal collision avoidance half-plane lines based on the avoidance speed, the sharing ratio, the actual speed of the target pedestrian, and the actual speed of the adjacent pedestrians includes:

[0033] Determining the interpersonal collision avoidance half-plane line based on the equation of the interpersonal collision avoidance half-plane line;

[0034] Wherein, in the first perception case of the target perception domain and the target relative position indication, the equation of the interpersonal collision avoidance half-plane line is:

[0035]

[0036] In the second perception condition of the target perception domain and the target relative position indication, the equation of the interpersonal collision avoidance half-plane line is:

[0037]

[0038] Wherein, A represents the target pedestrian, B represents the adjacent pedestrian, represents the interpersonal collision avoidance half-plane line imposed by B on A, v2 represents the velocity vector that satisfies the condition, is the optimal speed of A, n is the unit vector of A relative to B that satisfies collision avoidance, and u is the minimum velocity change of A relative to B that satisfies collision avoidance.

[0039] In a second aspect, the present invention further provides a crowd behavior simulation device, the device comprising:

[0040] The first determination module is used to determine the target perception domain and target relative position of the target pedestrian;

[0041] A second determination module is used to determine the target pedestrian's share of the avoidance speed based on the target perception domain and the target relative position;

[0042] A third determination module is used to determine a back-up avoidance half-plane line based on the expected speed of the target pedestrian, wherein the back-up avoidance half-plane line is used to limit the feasible speed of the target pedestrian to be within a feasible solution region of the back-up avoidance half-plane line;

[0043] A simulation module, used for simulating crowd behavior through an ORCA model based on the sharing ratio and the retreat avoidance half-plane line;

[0044] Among them, the target perception domain includes the front visible area of ​​the target pedestrian and the rear perception area of ​​the target pedestrian; the target relative position is used to indicate the relative position information of the target pedestrian and the adjacent pedestrians; the avoidance speed is the speed required for the target pedestrian to avoid collision with the adjacent pedestrians.

[0045] In a third aspect, the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any of the crowd behavior simulation methods described above are implemented.

[0046] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the crowd behavior simulation methods described above.

[0047] In a fifth aspect, the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any of the crowd behavior simulation methods described above.

[0048] The crowd behavior simulation method, device, electronic device and storage medium provided by the present invention improve the basic ORCA model by introducing a perception domain and a back-up avoidance configuration, wherein the perception domain is used to distinguish the influence of relative orientation on the avoidance speed sharing, and the back-up avoidance configuration is used to match the pedestrian's expectation of moving to the target point as quickly as possible, thereby achieving that in terms of space, the improved ORCA model can stably reproduce the "arch" congestion phenomenon, the pedestrian trajectory is more stable, and the deformation problem existing in the basic ORCA model is eliminated; in terms of time, the improved ORCA model has order stability, and the disorder problem existing in the basic ORCA model is eliminated. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 It is one of the schematic diagrams of simulation results of the basic ORCA model in the prior art;

[0051] Figure 2 This is the second schematic diagram of the simulation results of the basic ORCA model in the prior art;

[0052] Figure 3 This is the third schematic diagram of the simulation results of the basic ORCA model in the prior art;

[0053] Figure 4 This is the fourth schematic diagram of the simulation results of the basic ORCA model in the prior art;

[0054] Figure 5 This is the fifth schematic diagram of the simulation results of the basic ORCA model in the prior art;

[0055] Figure 6 This is the sixth schematic diagram of the simulation results of the basic ORCA model in the prior art;

[0056] Figure 7This is the seventh schematic diagram of the simulation results of the basic ORCA model in the prior art;

[0057] Figure 8 This is the eighth schematic diagram of the simulation results of the basic ORCA model in the prior art;

[0058] Fig. 9 It is a flow chart of the crowd behavior simulation method provided by the present invention;

[0059] Fig.10 is a schematic diagram of the perception domain provided by the present invention;

[0060] Fig.11 is a schematic diagram of a retreat-avoiding half-plane line provided by the present invention;

[0061] Fig.12 is a schematic diagram of a feasible speed region provided by the present invention;

[0062] Fig.13 is a simulation flow chart of the improved ORCA model provided by the present invention;

[0063] Fig.14 This is one of the schematic diagrams of simulation results of the improved ORCA model provided by the present invention;

[0064] Fig.15 This is the second schematic diagram of simulation results of the improved ORCA model provided by the present invention;

[0065] Fig.16 This is the third schematic diagram of simulation results of the improved ORCA model provided by the present invention;

[0066] Fig.17 This is the fourth schematic diagram of simulation results of the improved ORCA model provided by the present invention;

[0067] Fig.18 This is the fifth schematic diagram of simulation results of the improved ORCA model provided by the present invention;

[0068] Fig.19 This is the sixth schematic diagram of simulation results of the improved ORCA model provided by the present invention;

[0069] Fig. 20 This is the seventh schematic diagram of simulation results of the improved ORCA model provided by the present invention;

[0070] Fig.21 This is the eighth schematic diagram of simulation results of the improved ORCA model provided by the present invention;

[0071] Fig. 22 It is a structural schematic diagram of the crowd behavior simulation device provided by the present invention;

[0072] Fig.23It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0073] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0074] The "arch effect" is a typical self-organizing characteristic of crowded crowds, which is manifested as the arch formed by the crowd in a narrow passage.

[0075] Figure 1 This is one of the schematic diagrams of the simulation results of the basic ORCA model in the prior art. In the process of simulating the "arch effect" by the basic ORCA model, the first stage of the simulation is as follows: Figure 1 As shown in FIG, pedestrians can be divided into the first type of pedestrians, the second type of pedestrians, the third type of pedestrians and the fourth type of pedestrians according to the distance between their starting positions and the exit from near to far. Pedestrians move toward the exit, and after a period of time, the first type of pedestrians with a closer starting position arrive at the exit.

[0076] Figure 2 This is the second schematic diagram of the simulation results of the basic ORCA model in the prior art. The second stage of the simulation, such as Figure 2 As shown, due to the narrow exit, only some of the pedestrians who arrived first were able to pass through, while the rest of the pedestrians were stuck in congestion at the exit.

[0077] Figure 3 This is the third schematic diagram of the simulation results of the basic ORCA model in the prior art. The third stage of the simulation, such as Figure 3 As shown in the figure, the first type of pedestrians who arrived first and formed congestion were pushed away by the second type of pedestrians who arrived later and dispersed to both sides, while the second type of pedestrians occupied the exit. The same phenomenon occurred when the third and fourth types of pedestrians arrived at the exit. At this time, in terms of the overall shape and spatial distribution of the crowd, the arch reproduced by the basic ORCA model showed a flat feature.

[0078] Figure 4 This is the fourth schematic diagram of the simulation results of the basic ORCA model in the prior art, the fourth stage of the simulation, such as Figure 4 As shown, the pedestrians who are squeezed to the sides will try their best to gather towards the exit, and the shape of the group will present the characteristics of a spire.

[0079] Figure 5 This is the fifth schematic diagram of the simulation results of the basic ORCA model in the prior art. Figure 6This is the sixth schematic diagram of the simulation results of the basic ORCA model in the prior art, the fifth stage of the simulation, such as Figure 5 and Figure 6 As shown, it continuously and dynamically changes between flat and pointed at the moment of subsequent congestion.

[0080] It can be seen that the crowd arch characteristics simulated by the basic ORCA model are not obvious and unstable, and in the congestion process, there is also a "vortex" phenomenon in which some pedestrians walk in circles (such as Figure 4-Figure 6 As shown in the figure, the simulation results are inconsistent with the actual behavior.

[0081] In terms of population order, Figure 3 As shown in the basic ORCA model, the pedestrians who arrive at the exit first are squeezed to the sides by the pedestrians who arrive later, resulting in a disorder phenomenon where the pedestrians who arrive later are closer to the exit. Figure 4 and Figure 5 It can be seen that this disorder gradually intensifies over time, causing pedestrians who are closer at the start to be completely mixed with pedestrians who are farther away, and the crowd is no longer orderly.

[0082] Figure 7 This is the seventh schematic diagram of the simulation results of the basic ORCA model in the prior art, such as Figure 7 As shown, all pedestrian trajectory lines represent the trajectory lines of all pedestrians from the starting position to the exit when there are obstacles, reflecting the total activity range of pedestrians in congestion.

[0083] like Figure 7 As shown, 5 pedestrian trajectory lines are extracted, corresponding to pedestrians starting from different areas, reflecting the specific travel routes of the pedestrians, including the pedestrian trajectory of ID=83, the pedestrian trajectory of ID=87, the pedestrian trajectory of ID=249, the pedestrian trajectory of ID=182 and the pedestrian trajectory of ID=93.

[0084] The trajectory lines of the above five pedestrians can be roughly divided into the following three types: (1) the forward, backward, left and right swinging type (such as the pedestrian trajectories of ID=83, ID=87 and ID=249). The main feature of this type of trajectory line is that after entering the congested area, the pedestrians are squeezed by the pedestrians behind and dispersed to the left or right side of the crowd, and then move to the back of the crowd, and then are brought to the other side of the crowd, and finally pass through the exit; (2) the unilateral swinging type (such as the pedestrian trajectory of ID=182). The main feature of this type of trajectory line is that after entering the congested area, the pedestrians are squeezed and dispersed to one side of the crowd, and then turn back from this side along the back of the crowd and pass through the exit; (3) the direct passing type (such as the pedestrian trajectory of ID=93). The main feature of this type of trajectory line is that the pedestrians are avoided by other pedestrians or obtain a passing route by squeezing other pedestrians, and pass directly from the exit in an approximately straight line.

[0085] It can be seen that in the simulation results of the basic ORCA model, most pedestrians exhibit the characteristics of (1) and (2), and the crowd therefore shows the characteristics of frequent left-right swaying and front-back deformation; while the pedestrians who pass directly are mainly those in the rearmost area, who gain passage space by squeezing pedestrians in other areas.

[0086] Figure 8 This is the eighth schematic diagram of the simulation results of the basic ORCA model in the prior art, such as Figure 8 As shown, in the coordinate system, the horizontal axis represents the time when the pedestrian passes through the exit (in seconds), and the vertical axis represents the distance from the pedestrian's initial position to the exit (in meters).

[0087] from Figure 8 It can be seen that the relationship between the initial distance of each pedestrian in the basic ORCA model (i.e., the distance from the initial position to the exit) and the time of passing through the exit is that there is almost no correlation between the time when the pedestrian passes through the exit and the initial distance, and the crowd as a whole shows disordered or even disordered characteristics.

[0088] Fig. 9 is a flow chart of the crowd behavior simulation method provided by the present invention, such as Fig. 9 As shown, the method includes:

[0089] Step 901, determining the target perception domain and target relative position of the target pedestrian;

[0090] Optionally, the target perception domain includes a front visible area of ​​the target pedestrian and a rear perception area of ​​the target pedestrian.

[0091] Optionally, the perception domain may be the perception range of the target pedestrian in the crowd to the neighboring pedestrians around him.

[0092] It is understandable that the avoidance behavior between pedestrians will be affected by each other's positions. Generally speaking, the pedestrians behind are more affected by the pedestrians in front, while the pedestrians in front cannot always perceive the pedestrians behind due to orientation, line of sight and other reasons, and are relatively less affected by them.

[0093] Optionally, the adjacent pedestrian may be a pedestrian within a certain range around the target pedestrian.

[0094] Optionally, the target pedestrian will try to move toward the target point as much as possible during the process of moving, and the direction toward the target point can be used as the basis for determining the front and rear of the target pedestrian. The direction toward the target point can be the front of the target pedestrian, and the direction away from the target point can be the rear of the target pedestrian.

[0095] Optionally, the front visual domain of the perception domain may be a range visible to a pedestrian, and may be defined as an area in front of the pedestrian and within a certain radius.

[0096] Optionally, the rear perception domain of the perception domain may be a region where the target pedestrian senses the presence of some rear pedestrians through body contact or other reasons.

[0097] Optionally, the target relative orientation is used to indicate relative position information between the target pedestrian and adjacent pedestrians.

[0098] Therefore, in the present invention, by determining the target perception domain of the target pedestrian, the front visible area and the rear perception area of ​​the target pedestrian can be distinguished, so that the target pedestrian is affected by the adjacent pedestrians in front differently from the target pedestrian is affected by the adjacent pedestrians behind. This, combined with the target relative position, can be used in subsequent steps to determine the target pedestrian's share of the avoidance speed.

[0099] Step 902, determining a proportion of the target pedestrian's share of the avoidance speed based on the target perception domain and the target relative position;

[0100] Optionally, the avoidance speed is a speed required for the target pedestrian to avoid collision with an adjacent pedestrian.

[0101] Optionally, a set of pedestrians (adjacent pedestrians) within a certain range around the target pedestrian is obtained, and for the adjacent pedestrians therein, the avoidance speed required to achieve collision avoidance between the target pedestrian and the adjacent pedestrians is calculated based on parameters such as the respective radius, current actual speed and reaction time of the target pedestrian and the adjacent pedestrian.

[0102] Alternatively, the avoidance speed may be a minimum avoidance speed.

[0103] Optionally, the target pedestrian's share of the avoidance speed may be determined based on the circumstances.

[0104] Optionally, the first perception situation may be that the target pedestrian A and the adjacent pedestrian B behind the target pedestrian A can perceive each other, and the respective sharing ratios of the target pedestrian A and the adjacent pedestrian B are 0.5.

[0105] Optionally, the second perception situation may be that the target pedestrian A sees the adjacent pedestrian B in front of the target pedestrian A, while the adjacent pedestrian B cannot perceive the target pedestrian A, then the share ratio of the target pedestrian A may be 1.

[0106] Optionally, in the second perception situation, the sharing ratio of the adjacent pedestrian B may be 0.

[0107] Therefore, in the present invention, by determining the target pedestrian's share of the avoidance speed, it is possible to determine the share of the avoidance speed between the target pedestrian and the adjacent pedestrian based on the perception situation, so that the improved ORCA model is consistent with the actual behavior during the simulation process.

[0108] Step 903, based on the desired speed of the target pedestrian, determine a back-up avoidance half-plane line, where the back-up avoidance half-plane line is used to limit the feasible speed of the target pedestrian to be within a feasible solution region of the back-up avoidance half-plane line;

[0109] Optionally, according to the current position of the target pedestrian and the target position, a unit vector pointing from the current position to the target position can be calculated, and multiplied by the maximum walking speed to obtain the expected speed of the target pedestrian at the current moment.

[0110] It is understandable that when pedestrians are evacuating, they tend to move towards their destination and, unless there are special circumstances, they will try to avoid moving backwards away from the target point.

[0111] It can be understood that the "half-plane line" is the boundary line of a feasible speed area. In space, it is equivalent to dividing the entire speed plane into two, so it is called the "half-plane line".

[0112] It can be understood that the ORCA model can take the clockwise direction as the positive direction. After determining the positive direction of the half-plane line, looking along the positive direction of the half-plane line, the left half-plane on the left side of the half-plane line is the feasible solution area.

[0113] Therefore, in the present invention, by determining the retreat avoidance half-plane line, it can be achieved that the target pedestrian tends to always move toward the destination, so that the improved ORCA model is consistent with the actual behavior during the simulation process.

[0114] Step 904: Based on the sharing ratio and the retreat avoidance half-plane line, crowd behavior simulation is performed through the ORCA model.

[0115] Optionally, after the sharing ratio and the retreat avoidance half-plane line are determined, they can be applied to the basic ORCA model to obtain an improved ORCA model.

[0116] Optionally, crowd behavior simulation can be performed based on the improved ORCA model.

[0117] Therefore, in the present invention, by combining the sharing ratio and the retreat avoiding half-plane line with the basic ORCA model, an improved ORCA model can be obtained, and then the simulation of pedestrian group behavior can be achieved through the improved ORCA model.

[0118] Optionally, by determining the target perception domain of the target pedestrian, the front visible area and the rear perception area of ​​the target pedestrian can be distinguished. The sharing ratio can be determined through the target perception domain and the target relative position, so that the sharing of the avoidance speed between the target pedestrian and the adjacent pedestrian can be determined based on the perception situation; the retreat avoidance half-plane line can be determined through the expected speed of the target pedestrian, so that the target pedestrian tends to move towards the destination all the time; by combining the sharing ratio and the retreat avoidance half-plane line with the basic ORCA model, an improved ORCA model can be obtained, thereby realizing the simulation of pedestrian group behavior.

[0119] The present invention improves the basic ORCA model by introducing perception domain and back-off avoidance configuration, wherein the perception domain is used to distinguish the influence of relative orientation on avoidance speed sharing, and the back-off avoidance configuration is used to match the pedestrian's expectation of moving toward the target point as quickly as possible, thereby achieving that in terms of space, the improved ORCA model can stably reproduce the "arch" congestion phenomenon, the pedestrian trajectory is more stable, and the deformation problem existing in the basic ORCA model is eliminated. In terms of time, the improved ORCA model has order stability, eliminating the disorder problem existing in the basic ORCA model.

[0120] Optionally, determining a target perception domain and a target relative position of a target pedestrian includes:

[0121] Determining a target perception domain of the target pedestrian based on an equation of a forward visible area of ​​the target pedestrian and an equation of a rearward perception area of ​​the target pedestrian;

[0122] Among them, the equation of the front visible area is:

[0123]

[0124] The equation for the rear perception area is:

[0125]

[0126] Where A represents the target pedestrian, T A represents the target position of the target pedestrian, p A represents the current position of the target pedestrian, d A represents the front vector of the target pedestrian, r A represents the visible radius of the target pedestrian, θ A represents the sight range of the target pedestrian, p B represents the current position of the neighboring pedestrian, γ A Indicates the rear perception radius of the target pedestrian.

[0127] Optionally, the front of the target pedestrian is determined by an equation of the front vector of the target pedestrian.

[0128] Optionally, the equation of the front vector of the target pedestrian may be the following equation:

[0129] d A =T A -p A ;

[0130] Among them, A represents the target pedestrian, d A represents the front vector of the target pedestrian, T A represents the target position of the target pedestrian, p A Indicates the current position of the target pedestrian.

[0131] Optionally, the front visual domain of the perception domain may be a range visible to a pedestrian, and may be defined as an area in front of the pedestrian and within a certain radius.

[0132] Optionally, the forward visible range of the target pedestrian is determined by an equation of the forward visible range.

[0133] Optionally, the rear perception domain of the perception domain may be a region where the target pedestrian senses the presence of some rear pedestrians through body contact or other reasons.

[0134] Optionally, the rear perception field of the target pedestrian is determined by an equation of the rear perception field.

[0135] Fig.10 is a schematic diagram of the perception domain provided by the present invention, such as Fig.10 As shown, d A represents the direction of the target pedestrian A towards the target point, d A is the front of the target pedestrian A, r is the front visible radius of the target pedestrian A, γ is the rear perceptible radius of the target pedestrian A, and d A , r and the sight range angle of the target pedestrian A can determine the front visible area of ​​the target pedestrian A, combined with d A , γ and the rear perception range angle of the target pedestrian A can determine the rear perception area of ​​the target pedestrian A.

[0136] Therefore, in the present invention, the target pedestrian's perception of the adjacent pedestrian in front can be determined through the equation of the front visible area, and the target pedestrian's perception of the adjacent pedestrian in the rear can be determined through the equation of the rear perception area. Then, combined with the perception of the front and rear adjacent pedestrians, the target perception domain of the target pedestrian can be obtained. Then, through the target perception domain, the front visible area and the rear perception area of ​​the target pedestrian can be distinguished, so that the target pedestrian is affected by the adjacent pedestrian in front differently from the target pedestrian is affected by the adjacent pedestrian in the rear. Then, combined with the target relative position, it can be used to determine the target pedestrian's share of the avoidance speed.

[0137] Optionally, the step of determining the retreating half-plane line to avoid based on the expected speed of the target pedestrian includes:

[0138] Based on the equation of the backsliding avoidance half-plane line, the backsliding avoidance half-plane line is determined; wherein the equation of the backsliding avoidance half-plane line is:

[0139]

[0140] Where A represents the target pedestrian, v1 represents the velocity vector that meets the conditions, The unit vector representing the expected velocity of the target pedestrian at time t, ORCA f Denotes that the retreat avoids the half-plane line.

[0141] Optionally, according to the current position of the target pedestrian and the target position, a unit vector pointing from the current position to the target position can be calculated, and multiplied by the maximum walking speed to obtain the expected speed of the target pedestrian at the current moment.

[0142] It is understandable that when pedestrians are evacuating, they tend to move towards their destination and, unless there are special circumstances, they will try to avoid moving backwards away from the target point.

[0143] It can be understood that the "half-plane line" is the boundary line of a feasible speed area. In space, it is equivalent to dividing the entire speed plane into two, so it is called the "half-plane line".

[0144] Understandably, ORCA f The model can be taken as the clockwise direction. After determining the positive direction of the half-plane line, looking along the positive direction of the half-plane line, the left half-plane on the left side of the half-plane line is the feasible solution area.

[0145] Fig.11 Schematic diagram of the retreat-avoidance half-plane line provided by the present invention, such as Fig.11 As shown, in the velocity space coordinate system, the backward avoidance velocity half plane line ORCA f Passing through the velocity origin, the unit vector e of the target pedestrian's desired velocity is perpendicular to the backward avoidance velocity half-plane line ORCA f .

[0146] Therefore, in the present invention, the retreat avoidance half-plane line can be determined by the equation of the retreat avoidance half-plane line, and then by the retreat avoidance half-plane line, it can be achieved that the target pedestrian tends to always move towards the destination, so that the improved ORCA model is consistent with the actual behavior during the simulation process.

[0147] Optionally, the crowd behavior simulation is performed by using an ORCA model based on the sharing ratio and the retreat avoidance half-plane line, including:

[0148] Determining a set of interpersonal collision avoidance half-plane lines based on the avoidance speed, the sharing ratio, the actual speed of the target pedestrian, and the actual speed of the adjacent pedestrians;

[0149] Determining a speed feasible solution region based on the set of interpersonal collision avoidance half-plane lines, the set of obstacle collision avoidance half-plane lines, and the retreat avoidance half-plane lines;

[0150] The speed feasible solution region is used to determine the target speed of the target pedestrian.

[0151] Optionally, based on the avoidance speed, the sharing ratio, the actual speed of the target pedestrian, and the actual speed of the adjacent pedestrians, a set of interpersonal collision avoidance half-plane lines may be determined.

[0152] Optionally, based on a set of interpersonal collision avoidance half-plane lines, a set of obstacle collision avoidance half-plane lines, and a back-off avoidance half-plane line, a speed feasible solution region may be determined.

[0153] Optionally, based on the speed feasible solution region, the target speed of the target pedestrian can be determined by a linear programming algorithm.

[0154] Fig.12 is a schematic diagram of the feasible speed area provided by the present invention, such as Fig.12 As shown, A represents the target pedestrian, v A max represents the maximum walking speed of the target pedestrian, D(0, v A max ) is a circle with the origin of the velocity coordinate system as the center and v A max The circle with a radius of represents the feasible speed range of the target pedestrian A without being affected by adjacent pedestrians and obstacles.

[0155] Optionally, the neighboring pedestrians of the target pedestrian A may be pedestrians B, pedestrians D, and pedestrians E, and the obstacle may be object C. Accordingly, Fig.12 The interpersonal collision avoidance half-plane line imposed by the adjacent pedestrian on the target pedestrian A can be and The obstacle avoidance half-plane line imposed by object C on target pedestrian A can be

[0156] In the velocity space coordinate system, retreat to avoid the velocity half-plane line ORCA f Passing through the velocity origin, the unit vector e of the target pedestrian's desired velocity is perpendicular to the backward avoidance velocity half-plane line ORCA f .

[0157] Fig.12The oblique line area in is the area to be solved, which is the feasible speed area after considering adjacent pedestrians, obstacles, and retreat avoidance factors. Then, an optimal speed can be solved from this feasible speed area.

[0158] Optionally, based on a linear programming algorithm, when the solution region is not empty, an optimal speed can be calculated, and when the solution region is empty, these half-plane lines can be appropriately broken through to calculate an optimal feasible solution.

[0159] It is understandable that in real life, pedestrians generally move towards the target point, and in extremely crowded situations, pedestrians can retreat a little to gain space for the crowd. Retreating to avoid the half-plane line does not absolutely prohibit pedestrians from retreating, but is to avoid the unreasonable phenomenon of large-scale and frequent retreats of pedestrians, so that the simulation results of the improved ORCA model are consistent with real life behavior.

[0160] Therefore, in the present invention, the speed feasible solution area can be determined through the set of interpersonal collision avoidance half-plane lines, the obstacle collision avoidance half-plane line set and the retreat avoidance half-plane line, and then the target speed of the target pedestrian can be determined through the speed feasible solution area, and then the position information and orientation information of the target pedestrian in the next time unit / moment can be determined through the target speed, and then through multiple target speed solutions, the simulation of pedestrian group behavior can be realized.

[0161] Optionally, determining the proportion of the target pedestrian's share of the avoidance speed based on the target perception domain and the target relative position includes:

[0162] In a first perception situation where the target perception domain and the target relative position indicate a first perception, the sharing ratio is set to a preset ratio value;

[0163] In the second perception situation where the target perception domain and the target relative position indicate a second perception, the sharing ratio is set to 1;

[0164] Among them, the first perception situation is a situation where the target pedestrian and the adjacent pedestrian can perceive each other, and the second perception situation is a situation where the target pedestrian can perceive the adjacent pedestrian unidirectionally.

[0165] Optionally, the target pedestrian's share of the avoidance speed may be determined based on the circumstances.

[0166] Optionally, the first perception situation may be that the target pedestrian A and the adjacent pedestrian B behind the target pedestrian A can perceive each other, and the respective sharing ratios of the target pedestrian A and the adjacent pedestrian B are 0.5.

[0167] Optionally, in the first perception case, the sharing ratio of the target pedestrian A can be 0.4, and correspondingly, the sharing ratio of the adjacent pedestrian B can be 0.6; the sharing ratio of the target pedestrian A can be 0.6, and correspondingly, the sharing ratio of the adjacent pedestrian B can be 0.4. There is no restriction on the sharing ratio here.

[0168] Optionally, the second perception situation may be that the target pedestrian A sees the adjacent pedestrian B in front of the target pedestrian A, while the adjacent pedestrian B cannot perceive the target pedestrian A, then the share ratio of the target pedestrian A may be 1.

[0169] Optionally, in the second perception situation, the sharing ratio of the adjacent pedestrian B may be 0.

[0170] Therefore, in the present invention, by dividing the perception situation of the target pedestrian into a first perception situation and a second perception situation, and then setting the sharing ratio of the target pedestrian in different perception situations differently, it is possible to determine the sharing of the avoidance speed between the target pedestrian and the adjacent pedestrian based on the perception situation, so that the improved ORCA model is consistent with the actual behavior during the simulation process.

[0171] Optionally, determining a set of interpersonal collision avoidance half-plane lines based on the avoidance speed, the sharing ratio, the actual speed of the target pedestrian, and the actual speed of the adjacent pedestrian includes:

[0172] Determining the interpersonal collision avoidance half-plane line based on the equation of the interpersonal collision avoidance half-plane line;

[0173] Wherein, in the first perception case of the target perception domain and the target relative position indication, the equation of the interpersonal collision avoidance half-plane line is:

[0174]

[0175] In the second perception condition of the target perception domain and the target relative position indication, the equation of the interpersonal collision avoidance half-plane line is:

[0176]

[0177] Wherein, A represents the target pedestrian, B represents the adjacent pedestrian, represents the interpersonal collision avoidance half-plane line imposed by B on A, v2 represents the velocity vector that satisfies the condition, is the optimal speed of A, n is the unit vector of A relative to B that satisfies collision avoidance, and u is the minimum velocity change of A relative to B that satisfies collision avoidance.

[0178] Optionally, in the first perception case, the collision avoidance half-plane line applied by the adjacent pedestrian to the target pedestrian can be determined by the equation of the interpersonal collision avoidance half-plane line, where the proportion of A to the minimum rate change is 0.5.

[0179] Optionally, in the second perception situation, the collision avoidance half-plane line imposed by the adjacent pedestrian on the target pedestrian can be determined by the equation of the interpersonal collision avoidance half-plane line, where the proportion of A to the minimum rate change is 1.

[0180] Therefore, in the present invention, the interpersonal collision avoidance half-plane line equation can be used to determine the interpersonal collision avoidance half-plane line according to the perception situation, and then the interpersonal collision avoidance half-plane line imposed on the target pedestrian by adjacent pedestrians under different perception situations can be set differently, so that the improved ORCA model is consistent with the actual behavior during the simulation process.

[0181] The present invention improves the basic ORCA model by introducing perception domain and back-off avoidance configuration, wherein the perception domain is used to distinguish the influence of relative orientation on avoidance speed sharing, and the back-off avoidance configuration is used to match the pedestrian's expectation of moving toward the target point as quickly as possible, thereby achieving that in terms of space, the improved ORCA model can stably reproduce the "arch" congestion phenomenon, the pedestrian trajectory is more stable, and the deformation problem existing in the basic ORCA model is eliminated. In terms of time, the improved ORCA model has order stability, eliminating the disorder problem existing in the basic ORCA model.

[0182] Fig.13 is a simulation flow chart of the improved ORCA model provided by the present invention, such as Fig.13 As shown, at time t, for a virtual crowd queue that already exists in the virtual scene, traverse each target pedestrian i in the queue and perform the following calculations:

[0183] Step 1: According to the current position of target pedestrian i and the target position, calculate the unit vector from the current position to the target position, and multiply it by the maximum walking speed to get the expected speed of target pedestrian i at time t;

[0184] Step 2: Obtain a set of pedestrians (adjacent pedestrians) within a certain range around the target pedestrian i. For the adjacent pedestrian j, calculate the minimum avoidance speed required for the target pedestrian i and the adjacent pedestrian j to avoid collision based on the respective radius of the target pedestrian i and the adjacent pedestrian j (in the ORCA model, a circle is used to represent pedestrians, and the radius of the circle can be the radius of the pedestrian), the current actual speed, the reaction time (the reaction time of the target pedestrian to avoid the adjacent pedestrian, the shorter the reaction time, the faster the avoidance behavior is executed, and the longer the reaction time, the slower the avoidance behavior is executed) and other parameters. Then, according to the relative orientation of the target pedestrian i and the adjacent pedestrian j, based on the perception domain model, determine the proportion of the target pedestrian i and the adjacent pedestrian j to share the minimum avoidance speed, and determine the interpersonal collision avoidance half-plane line between the target pedestrian i and the adjacent pedestrian j based on the actual speed. This step can obtain the set of all interpersonal collision avoidance half-plane lines of the target pedestrian i at time t;

[0185] Step 3: Obtain the set of obstacles within a certain range around the target pedestrian i. For any obstacle line k, the obstacle avoidance half-plane line of obstacle k to the target pedestrian i is constructed based on the principle that all collision avoidance responsibilities are borne by the individual pedestrian i (the target pedestrian's share ratio is 1). This step can obtain the set of all obstacle avoidance half-plane lines of the target pedestrian i at time t;

[0186] Step 4: Based on the principle of avoiding retreat, a speed line passing through the speed origin and perpendicular to the desired speed direction of the target pedestrian i is constructed as the retreat avoidance half-plane line. The retreat avoidance half-plane line can ensure that all feasible speeds are located in the feasible solution area of ​​the half-plane line;

[0187] It can be understood that the ORCA model can be clockwise as the positive direction. After determining the positive direction of the half-plane line, looking along the positive direction of the half-plane line, the left half-plane on the left side of the half-plane line is the feasible solution area;

[0188] Step 5: Steps 2, 3, and 4 result in a feasible speed solution region Q for the target pedestrian i surrounded by multiple half-plane lines. In step 5, a linear programming algorithm can be used to solve the speed region. When Q is not empty, an optimal speed closest to the expected speed can be obtained; when Q is empty, a three-dimensional linear programming algorithm can be used to appropriately break through Q to obtain a speed that is optimal to the expected speed. This speed solution is the optimal speed of the target pedestrian i.

[0189] Step 6: Multiplying it with the simulation step length dt and adding it to the position at time t, we can get the position of the target pedestrian i at time t+1, and use it to update the position attribute of the target pedestrian i;

[0190] Subsequent steps: If the target pedestrian i has reached the target point and there is no next target point, the target pedestrian i stops moving and exits the subsequent cycle; if the target pedestrian i has not reached the target point, or there is still a next target point, repeat steps 1 to 6 based on the crowd attributes at time t+1.

[0191] Fig.14 is one of the schematic diagrams of simulation results of the improved ORCA model provided by the present invention, such as Fig.14 As shown in Figure 2, in the process of improving the ORCA model to simulate the “arch effect”, the first stage of the simulation is as follows: Fig.14 As shown in the figure, pedestrians can be divided into the first type of pedestrians, the second type of pedestrians, the third type of pedestrians and the fourth type of pedestrians according to the distance between their starting position and the exit from near to far. In the process of moving towards the target point, the crowd gradually converges.

[0192] Fig.15 FIG. 2 is a second schematic diagram of simulation results of the improved ORCA model provided by the present invention. In the second stage of the simulation, Fig.15 As shown, the first type of pedestrians (in blue) who arrived first caused congestion after reaching the exit, and most of the pedestrians who arrived later (the second type of pedestrians, the third type of pedestrians, and the fourth type of pedestrians) queued behind the first type of pedestrians.

[0193] Fig.16 FIG. 3 is a schematic diagram of simulation results of the improved ORCA model provided by the present invention. The third stage of the simulation is as follows: Fig.16 As shown, a small number of pedestrians on the periphery continue to move forward, showing the characteristics of being distributed around the circumference of the first type of pedestrians.

[0194] Fig.17 FIG. 4 is a fourth schematic diagram of simulation results of the improved ORCA model provided by the present invention, the fourth stage of the simulation, as shown in FIG. Fig.17 As shown, the crowd as a whole presents a typical arch feature

[0195] Fig.18 This is the fifth schematic diagram of simulation results of the improved ORCA model provided by the present invention. Fig.19 FIG6 is a sixth schematic diagram of simulation results of the improved ORCA model provided by the present invention, the fifth stage of the simulation, such as Fig.18 and Fig.19 As shown, during the subsequent congested traffic process, the arch-shaped feature was always maintained, and no abnormal behaviors such as "vortex" occurred.

[0196] In terms of order, in the improved ORCA model, even if there is congestion between pedestrians, there is always a clear order of precedence, and the later arrivals are distributed around the earlier arrivals (e.g. Figure 15-17 shown); although Fig.18There is a certain degree of mixing between the third and fourth types of pedestrians. This is because the fourth type of pedestrians surround the third type of pedestrians and try to pass through the side exit, resulting in a certain degree of fusion. This behavior is also consistent with real behavior.

[0197] Fig. 20 FIG. 7 is a seventh schematic diagram of simulation results of the improved ORCA model provided by the present invention, such as Fig. 20 As shown, all pedestrian trajectory lines represent the trajectory lines of all pedestrians from the starting position to the exit when there are obstacles, reflecting the total activity range of pedestrians in congestion.

[0198] like Fig. 20 As shown, 7 pedestrian trajectory lines are extracted, corresponding to pedestrians starting from different areas, reflecting the specific routes of the pedestrians, including the trajectories of pedestrians ID=6, ID=13, ID=43, ID=141, ID=163, ID=186 and ID=188.

[0199] from Fig. 20 It can be seen that no matter which area the pedestrians are in, although their trajectories may bend partially due to avoidance and crowding, they basically maintain the characteristics of moving steadily towards the exit, without obvious left-right swings or back-and-forth turns. At the same time, it can also be seen that some pedestrians at the back move around the edge of the crowd and then move along the wall to the exit (such as pedestrians ID=13 and ID=188).

[0200] Fig.21 FIG8 is an eighth schematic diagram of simulation results of the improved ORCA model provided by the present invention, such as Fig.21 As shown, in the coordinate system, the horizontal axis represents the time when the pedestrian passes through the exit (in seconds), and the vertical axis represents the distance from the pedestrian's initial position to the exit (in meters).

[0201] from Fig.21 It can be seen that the initial distance of each pedestrian in the improved ORCA model (i.e., the distance from the initial position to the exit) shows an obvious correlation with the time of passing through the exit. The time when the pedestrian passes through the exit is related to the initial distance. The initial position farther from the exit corresponds to a longer walking time, and the initial position closer to the exit corresponds to a shorter walking time. This further shows that the order of pedestrians is basically maintained when passing through the crowded exit. It can be seen that the simulation results of the improved ORCA model have a high order stability.

[0202] The present invention improves the basic ORCA model by introducing perception domain and back-off avoidance configuration, wherein the perception domain is used to distinguish the influence of relative orientation on avoidance speed sharing, and the back-off avoidance configuration is used to match the pedestrian's expectation of moving toward the target point as quickly as possible, thereby achieving that in terms of space, the improved ORCA model can stably reproduce the "arch" congestion phenomenon, the pedestrian trajectory is more stable, and the deformation problem existing in the basic ORCA model is eliminated. In terms of time, the improved ORCA model has order stability, eliminating the disorder problem existing in the basic ORCA model.

[0203] The crowd behavior simulation device provided by the present invention is described below. The crowd behavior simulation device described below and the crowd behavior simulation method described above can be referenced to each other.

[0204] Fig. 22 Schematic diagram of the structure of the crowd behavior simulation device provided by the present invention. Fig. 22 As shown, the device includes: a first determination module 2201, a second determination module 2202, a third determination module 2203 and a simulation module 2204; wherein:

[0205] The first determination module 2201 is used to determine the target perception domain and target relative position of the target pedestrian;

[0206] The second determination module 2202 is used to determine the target pedestrian's share of the avoidance speed based on the target perception domain and the target relative position;

[0207] The third determination module 2203 is used to determine a back-up avoidance half-plane line based on the expected speed of the target pedestrian, wherein the back-up avoidance half-plane line is used to limit the feasible speed of the target pedestrian to be within the feasible solution region of the back-up avoidance half-plane line;

[0208] The simulation module 2204 is used to simulate crowd behavior through the ORCA model based on the sharing ratio and the retreat avoidance half-plane line;

[0209] Among them, the target perception domain includes the front visible area of ​​the target pedestrian and the rear perception area of ​​the target pedestrian; the target relative position is used to indicate the relative position information of the target pedestrian and the adjacent pedestrians; the avoidance speed is the speed required for the target pedestrian to avoid collision with the adjacent pedestrians.

[0210] Optionally, the device determines the target perception domain and target relative position of the target pedestrian through the first determination module 2201, and then determines the target pedestrian's share of the avoidance speed through the second determination module 2202, and then determines the retreat avoidance half-plane line through the third determination module 2203. The retreat avoidance half-plane line is used to limit the feasible speed of the target pedestrian to be within the feasible solution area of ​​the retreat avoidance half-plane line, and then performs crowd behavior simulation through the simulation module 2204.

[0211] The present invention improves the basic ORCA model by introducing perception domain and back-off avoidance configuration, wherein the perception domain is used to distinguish the influence of relative orientation on avoidance speed sharing, and the back-off avoidance configuration is used to match the pedestrian's expectation of moving toward the target point as quickly as possible, thereby achieving that in terms of space, the improved ORCA model can stably reproduce the "arch" congestion phenomenon, the pedestrian trajectory is more stable, and the deformation problem existing in the basic ORCA model is eliminated. In terms of time, the improved ORCA model has order stability, eliminating the disorder problem existing in the basic ORCA model.

[0212] Optionally, the first determining module is further used for:

[0213] Determine the target perception domain of the target pedestrian based on the equation of the front visible area of ​​the target pedestrian and the equation of the rear perception area of ​​the target pedestrian;

[0214] The equation for the visible area ahead is:

[0215]

[0216] The equation for the rear perception area is:

[0217]

[0218] Among them, A represents the target pedestrian, T A represents the target position of the target pedestrian, p A represents the current position of the target pedestrian, d A represents the front vector of the target pedestrian, r A represents the visible radius of the target pedestrian, θ A represents the sight range of the target pedestrian, p B represents the current position of the neighboring pedestrian, γ A Represents the rear perception radius of the target pedestrian.

[0219] Optionally, the third determining module is further used for:

[0220] Based on the equation of the retreat avoidance half-plane line, the retreat avoidance half-plane line is determined; wherein the equation of the retreat avoidance half-plane line is:

[0221]

[0222] Among them, A represents the target pedestrian, v1 represents the velocity vector that meets the conditions, The unit vector representing the expected velocity of the target pedestrian at time t, ORCA f Indicates retreat to avoid half-plane line.

[0223] Optionally, the simulation module is also used to:

[0224] Based on the avoidance speed, the sharing ratio, the actual speed of the target pedestrian, and the actual speed of the adjacent pedestrians, a set of interpersonal collision avoidance half-plane lines is determined;

[0225] A fourth determining unit, configured to determine a speed feasible solution region based on a set of interpersonal collision avoidance half-plane lines, a set of obstacle collision avoidance half-plane lines, and a retreat avoidance half-plane line;

[0226] Among them, the speed feasible solution region is used to determine the target speed of the target pedestrian.

[0227] Optionally, the second determining module is further used for:

[0228] In the first perception case of the target perception domain and the target relative position indication, the sharing ratio is set to a preset ratio value;

[0229] A second setting unit, configured to set the sharing ratio to 1 in a second perception situation where the target perception domain and the target relative position indicate a second perception;

[0230] Among them, the first perception situation is a situation where the target pedestrian and the adjacent pedestrian can perceive each other, and the second perception situation is a situation where the target pedestrian can perceive the adjacent pedestrian unidirectionally.

[0231] Optionally, the simulation module is also used to:

[0232] Based on the equation of the interpersonal collision avoidance half-plane line, determine the interpersonal collision avoidance half-plane line;

[0233] Wherein, in the case of the first perception of the target perception domain and the target relative position indication, the equation of the interpersonal collision avoidance half-plane line is:

[0234]

[0235] In the case of the target perception domain and the second perception of the target relative position indication, the equation of the interpersonal collision avoidance half-plane line is:

[0236]

[0237] Among them, A represents the target pedestrian, B represents the adjacent pedestrian, represents the interpersonal collision avoidance half-plane line imposed by B on A, v2 represents the velocity vector that satisfies the condition, is the optimal speed of A, n is the unit vector of A relative to B that satisfies collision avoidance, and u is the minimum velocity change of A relative to B that satisfies collision avoidance.

[0238] The present invention improves the basic ORCA model by introducing perception domain and back-off avoidance configuration, wherein the perception domain is used to distinguish the influence of relative orientation on avoidance speed sharing, and the back-off avoidance configuration is used to match the pedestrian's expectation of moving toward the target point as quickly as possible, thereby achieving that in terms of space, the improved ORCA model can stably reproduce the "arch" congestion phenomenon, the pedestrian trajectory is more stable, and the deformation problem existing in the basic ORCA model is eliminated. In terms of time, the improved ORCA model has order stability, eliminating the disorder problem existing in the basic ORCA model.

[0239] Fig.23 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Fig.23 As shown, the electronic device may include: a processor (processor) 2310, a communication interface (Communications Interface) 2320, a memory (memory) 2330 and a communication bus 2340, wherein the processor 2310, the communication interface 2320, and the memory 2330 communicate with each other through the communication bus 2340. The processor 2310 can call the logic instructions in the memory 2330 to execute the crowd behavior simulation method, which includes: determining the target perception domain and the target relative orientation of the target pedestrian; based on the target perception domain and the target relative orientation, determining the target pedestrian's share of the avoidance speed; based on the expected speed of the target pedestrian, determining the retreat avoidance half-plane line, the retreat avoidance half-plane line is used to limit the feasible speed of the target pedestrian to be located in the feasible solution area of ​​the retreat avoidance half-plane line; based on the share ratio and the retreat avoidance half-plane line, crowd behavior simulation is performed through the ORCA model; wherein the target perception domain includes the front visible area of ​​the target pedestrian and the rear perception area of ​​the target pedestrian; the target relative orientation is used to indicate the relative position information of the target pedestrian and the adjacent pedestrians; the avoidance speed is the speed required for the target pedestrian to achieve collision avoidance with the adjacent pedestrians.

[0240] In addition, the logic instructions in the above-mentioned memory 2330 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0241] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the crowd behavior simulation method provided by the above-mentioned methods, which includes: determining the target perception domain and target relative orientation of the target pedestrian; based on the target perception domain and the target relative orientation, determining the target pedestrian's share of the avoidance speed; based on the expected speed of the target pedestrian, determining a retreat avoidance half-plane line, the retreat avoidance half-plane line is used to limit the feasible speed of the target pedestrian to be located in the feasible solution area of ​​the retreat avoidance half-plane line; based on the share ratio and the retreat avoidance half-plane line, crowd behavior simulation is performed through the ORCA model; wherein, the target perception domain includes the target pedestrian's front visible area and the target pedestrian's rear perception area; the target relative orientation is used to indicate the relative position information of the target pedestrian and the adjacent pedestrians; the avoidance speed is the speed required for the target pedestrian to achieve collision avoidance with the adjacent pedestrians.

[0242] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the crowd behavior simulation method provided by the above-mentioned methods, the method comprising: determining the target perception domain and target relative orientation of the target pedestrian; determining the target pedestrian's share of the avoidance speed based on the target perception domain and the target relative orientation; determining a retreat avoidance half-plane line based on the target pedestrian's expected speed, the retreat avoidance half-plane line being used to limit the target pedestrian's feasible speed to be within the feasible solution region of the retreat avoidance half-plane line; based on the share ratio and the retreat avoidance half-plane line, performing crowd behavior simulation through the ORCA model; wherein the target perception domain includes the target pedestrian's front visible area and the target pedestrian's rear perception area; the target relative orientation is used to indicate the relative position information of the target pedestrian and the adjacent pedestrians; the avoidance speed is the speed required for the target pedestrian to achieve collision avoidance with the adjacent pedestrians.

[0243] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0244] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0245] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A crowd behavior simulation method, characterized in that: The method comprises: Determine the target perception domain and target relative position of the target pedestrian; the target perception domain of the target pedestrian is determined based on the equation of the front visible area of ​​the target pedestrian and the equation of the rear perception area of ​​the target pedestrian; wherein the equation of the front visible area is: ； The equation for the rear perception area is: ； in, represents the target pedestrian, represents the target position of the target pedestrian, represents the current position of the target pedestrian, represents the front vector of the target pedestrian, Indicates the visible radius of the target pedestrian, Indicates the sight range of the target pedestrian. represents the current position of the neighboring pedestrians, Indicates the rear perception radius of the target pedestrian; In the first perception situation where the target perception domain and the target relative direction indicate a first perception situation, the sharing ratio is set to a preset ratio value; in the second perception situation where the target perception domain and the target relative direction indicate a second perception situation, the sharing ratio is set to 1; wherein the first perception situation is a situation where the target pedestrian and the adjacent pedestrian can perceive each other, and the second perception situation is a situation where the target pedestrian can perceive the adjacent pedestrian unidirectionally; Based on the equation of the backsliding avoidance half-plane line, the backsliding avoidance half-plane line is determined; wherein the equation of the backsliding avoidance half-plane line is: ； in, represents the target pedestrian, represents the velocity vector that satisfies the condition, Indicated in The unit vector of the target pedestrian’s expected velocity at time , represents the retreat avoidance half-plane line, and the retreat avoidance half-plane line is used to limit the feasible speed of the target pedestrian to be located in the feasible solution area of ​​the retreat avoidance half-plane line; Based on the equation of the interpersonal collision avoidance half-plane line, the interpersonal collision avoidance half-plane line is determined; wherein, in the first perception case of the target perception domain and the target relative orientation indication, the equation of the interpersonal collision avoidance half-plane line is: ； In the second perception condition of the target perception domain and the target relative position indication, the equation of the interpersonal collision avoidance half-plane line is: ； in, represents the target pedestrian, B represents the adjacent pedestrian, It represents the interpersonal collision avoidance half-plane line imposed by B on A. represents the velocity vector that satisfies the condition, is the optimal speed of A, n is the unit vector of A relative to B that satisfies collision avoidance, and u is the minimum velocity change of A relative to B that satisfies collision avoidance; Determining a speed feasible solution region based on the set of interpersonal collision avoidance half-plane lines, the set of obstacle collision avoidance half-plane lines, and the retreat avoidance half-plane lines; Among them, the speed feasible solution area is used to determine the target speed of the target pedestrian, the target perception domain includes the front visible area of ​​the target pedestrian and the rear perception area of ​​the target pedestrian; the target relative orientation is used to indicate the relative position information of the target pedestrian and the adjacent pedestrians; the avoidance speed is the speed required for the target pedestrian to avoid collision with the adjacent pedestrians.

2. A crowd behavior simulation device, characterized in that: The device comprises: The first determination module is used to determine the target perception domain and the target relative position of the target pedestrian; the target perception domain of the target pedestrian is determined based on the equation of the front visible area of ​​the target pedestrian and the equation of the rear perception area of ​​the target pedestrian; wherein the equation of the front visible area is: ； The equation for the rear perception area is: ； in, represents the target pedestrian, represents the target position of the target pedestrian, represents the current position of the target pedestrian, represents the front vector of the target pedestrian, Indicates the visible radius of the target pedestrian, Indicates the sight range of the target pedestrian. represents the current position of the neighboring pedestrians, Indicates the rear perception radius of the target pedestrian; A second determination module is configured to set the sharing ratio to a preset ratio value when the target perception domain and the target relative orientation indicate a first perception situation; and set the sharing ratio to 1 when the target perception domain and the target relative orientation indicate a second perception situation; wherein the first perception situation is a situation where the target pedestrian and the adjacent pedestrian can perceive each other, and the second perception situation is a situation where the target pedestrian can perceive the adjacent pedestrian unidirectionally; The third determination module is used to determine the retreat avoidance half-plane line based on the equation of the retreat avoidance half-plane line; wherein the equation of the retreat avoidance half-plane line is: ； in, represents the target pedestrian, represents the velocity vector that satisfies the condition, Indicated in The unit vector of the target pedestrian’s expected velocity at time , represents the retreat avoidance half-plane line, and the retreat avoidance half-plane line is used to limit the feasible speed of the target pedestrian to be located in the feasible solution area of ​​the retreat avoidance half-plane line; A simulation module is used to determine the interpersonal collision avoidance half-plane line based on the equation of the interpersonal collision avoidance half-plane line; wherein, in the first perception case of the target perception domain and the target relative orientation indication, the equation of the interpersonal collision avoidance half-plane line is: ； In the second perception condition of the target perception domain and the target relative position indication, the equation of the interpersonal collision avoidance half-plane line is: ； in, represents the target pedestrian, B represents the adjacent pedestrian, It represents the interpersonal collision avoidance half-plane line imposed by B on A. represents the velocity vector that satisfies the condition, is the optimal speed of A, n is the unit vector of A relative to B that satisfies collision avoidance, and u is the minimum velocity change of A relative to B that satisfies collision avoidance; Determining a speed feasible solution region based on the set of interpersonal collision avoidance half-plane lines, the set of obstacle collision avoidance half-plane lines, and the retreat avoidance half-plane lines; Among them, the target perception domain includes the front visible area of ​​the target pedestrian and the rear perception area of ​​the target pedestrian; the target relative position is used to indicate the relative position information of the target pedestrian and the adjacent pedestrians; the avoidance speed is the speed required for the target pedestrian to avoid collision with the adjacent pedestrians.

3. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the crowd behavior simulation method as claimed in claim 1 are implemented.

4. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the crowd behavior simulation method as claimed in claim 1 are implemented.

5. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the crowd behavior simulation method as claimed in claim 1 are implemented.

Images