CN106096072B - Dense crowd emulation mode based on intelligent body - Google Patents
Dense crowd emulation mode based on intelligent body Download PDFInfo
- Publication number
- CN106096072B CN106096072B CN201610327242.0A CN201610327242A CN106096072B CN 106096072 B CN106096072 B CN 106096072B CN 201610327242 A CN201610327242 A CN 201610327242A CN 106096072 B CN106096072 B CN 106096072B
- Authority
- CN
- China
- Prior art keywords
- pedestrian
- distance
- circle
- pedestrians
- crowd
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
Abstract
The Dense crowd emulation mode based on intelligent body that the present invention provides a kind of.This method specifically includes that the corresponding three circles manikin of construction pedestrian, the circle that the three circle manikin is intersected by three form, wherein intermediate great circle indicates that trunk, the roundlet of two sides indicate two shoulders；The clear distance between the corresponding three circles manikin of two pedestrians is calculated, using the clear distance as the distance between described two pedestrians.The embodiment of the present invention is by three circle manikins under the premise of not increasing algorithm complexity, it dexterously solves the untrue property of circular granular simulation pedestrian's body, lays a good foundation for the calculating and highdensity calculating of pedestrian's individual physical contact force in crowded crowd.Crowd density calculation method, the considerations of making crowd density from individual angle fining, describes each pedestrian to the subjective feeling of itself local environment with the method for quantization, this method provides possibility for the calculating and analysis of pedestrian's individual physical contact force.
Description
Technical field
The present invention relates to pedestrian simulation modelling technique field more particularly to a kind of Dense crowd emulation based on intelligent body
Method.
Background technique
As major urban track traffic Operation Network gradually forms, Network effects are more and more obvious, and the volume of the flow of passengers is substantially
Increase, subway station is crowded to capacity, and service level is low, and security risk is very big.Therefore, peak period congestion state, city are studied
The behavioural characteristic and stress of rail traffic station pedestrian develops congestion state debarkation stop pedestrian simulation using computer technology
Model and implementation method provide for the crowded risk key node in evaluation station, improvement facility layout and station one skilled in the art's organization scheme
Foundation ensures station safe operation and the security of the lives and property of passenger etc., is of great significance.
Currently, pedestrian simulation model mainly includes both macro and micro model.Micromodel simulation result is fine, accurate, adds
Computer technology rapid development, become pedestrian simulation field using most simulation models.It is micro- according to spatial description method
It sees model and is divided into spatial spreading model and space continuous model；Wherein, spatial spreading model is using cellular Automation Model as representative,
Space continuous model is using social force model as representative.
It is especially painstaking as the performance of the discrete model of representative using cellular automata in terms of Dense crowd moves simulation.It is first
First, discrete model is to divide the space into grid, and a pedestrian occupies one or more grid, the shaped form and net of pedestrian's body
The shape of lattice is difficult to fit like a glove, and is extremely difficult to be in close contact between pedestrian, limits the density threshold of crowd.In addition, discrete
Model is rule-based model, does not have physical contact force between pedestrian；Henein&White is to indicate contact force to high density
The influence of crowd introduces " field of force " conceptual construct efforts of everyone models, but still cannot intuitively reflect contact force numerical value.In this way,
The judge of simulation model effect can only be verified by macroscopical social phenomenon, lack finer quantitative verification approach.
The above-mentioned deficiency of discrete model can be overcome well using social force model as the continuous model of representative.In social force
In model, pedestrian's body indicates that each pedestrian by oneself requirement with comfortable speed in addition to being arrived at the destination using circular granular
Hope driving force, the two kinds of power also applied by other pedestrians or barrier: social force and physical force (contact force).Wherein
There is no actual physical resources for social force, but reflect pedestrian and wish that oneself is walked with comfortable speed along specific direction
To destination, while the psychology trend with other pedestrian's hypotelorisms or collision is avoided again, main includes expectation with comfortable speed
It moves towards the driving force of destination and avoids the repulsive force for bumping against or getting too close to pedestrian or barrier.And physical force is then only expert at
People's density is especially big, and generation when pedestrian is physically contacted between each other or between pedestrian and barrier mainly includes normal direction
Extruding force and tangential frictional force.
Social force model is widely recognized as because parameter has actual physical meaning, has business software Innovation Input to make at present
With；However crowd simulation density maximum has differences no more than the maximal density under the conditions of 8 people/m2, with actual operation.Study carefully its original
Cause, pedestrian's body uses round simulation, although reducing calculation amount, does not conform to the actual conditions.
Langston et al. is emulated using modified social force model with reference to parameter value in Helbing correlative study
Out when stand in 10m length 14 people when maximum extrusion pressure be 500N, stablize after pressure be 100-200N；Lu Chunxia is used social force
Model calculates the extruding force that can produce maximum 650N when Dense crowd evacuation in conjunction with power propagation model.Social force model, but
It is that reasonable explanation do not provided for the value of parameter in the research such as Helbing, and in Dense crowd simulation especially
Important human body elasticity modulus and social force sphere of action value always is to be considered with constant, is not conformed to the actual conditions, is thus led
Pedestrian in Dense crowd " oscillation ", extruding force calculating has been caused not to conform to the actual conditions.Therefore, Pelechano et al. is crowded in consideration
When, give up contact squeeze power, crowd's high density is realized using small personal space threshold value and repulsive force.
Pedestrian's implementation method used by the simulating scheme of the first Dense crowd is that round particle is real in the prior art
It is existing.The implementation method of round particle has a two benefits for Computer Simulation: first, pedestrian in normal traveling process by
It is different in the degree of concern to itself different directions, lead to barrier or pedestrian psychological repellence of the pedestrian in front of itself visual field more
By force, and its side or barrier behind or pedestrian then attention rate is smaller, this gives in social force model also by direction weight
To embody；Therefore, though pedestrian's body approximate ellipsoidal；But because the presence of this direction weight, causes pedestrian traveling across
Individual space in journey is closer round.Second, in clear distance between calculating particle, circle can simplify calculating, keep algorithm simpler
Clean, execution efficiency is higher；And under low density condition, since pedestrian's spacing is much larger than pedestrian's particle radii, come by circle in terms of
The loss of significance for calculating clear distance will not be too big.
In the prior art the shortcomings that the simulating scheme of the first Dense crowd are as follows: under conditions of high density, pedestrian group's
Flowing velocity is lower, and the personal air that pedestrian is formed by psychological application has been broken；So the travel region of pedestrian is from circle
Contracted the oval space become using pedestrian's individual space as boundary, or even the static item in rail transit cars in shape space
Under part, human body can be according to human body elasticity modulus moderate compression.Therefore, to realize the simulation model under high density crowed condition,
It needs to improve pedestrian dummy.
The simulating scheme of second of Dense crowd uses social force model in the prior art, and social force model is a kind of power
The psychological application of pedestrian is quantified as social force and physical force by driving model, the model, the social force and physical force collective effect
The motor behavior of pedestrian is driven, a kind of social force model schematic diagram in the prior art is as shown in Figure 1.Wherein,For pedestrian's fortune
Target point D is directed toward in dynamic driving force, direction；FαβFor the psychological application power to pedestrian β that pedestrian α is subject to, direction is directed toward by pedestrian β
Pedestrian α, shows as repulsion；For the psychological application power to barrier (wall) that pedestrian α is subject to, direction is perpendicular to obstacle
Pedestrian is directed toward on object (wall) surface, shows as repulsion；It is other pedestrians or place to the attraction of pedestrian α；ξ is random force,
To simulate the random variation of behavior in pedestrian's walking process, such as direction of travel is deposited in an alternative case, and random force can be more true
The random selection of real simulation pedestrian, direction and size are indefinite.
Therefore the dynamical motion equation that pedestrian α can be write out according to Newton's second law, as shown in formula (1)
In the prior art the shortcomings that the simulating scheme of second of Dense crowd are as follows: people has as a kind of intelligent beings
Visual capacity removes perception ambient enviroment, and there is decision-making capability to go to judge the track route of local optimum, has autonomous adjustment mechanism
Adjust the behavior state of itself.Pedestrian not only considers itself stress in decision, it is also contemplated that the factor of ambient enviroment.Pedestrian
It is exactly pedestrian density to perception information most intuitive around itself, and the decision-making foundation of pedestrian's direction of travel, speed etc. is also led
Consider the crowd density of surrounding pedestrian.Therefore, only behavior state can be made to lack authenticity with the walking behavior of power drive pedestrian.
In addition to this, the considerations of passing round in the social force model of the program to pedestrian is improperly worked as, it will be apparent that has ignored the autonomous of pedestrian
Property and dynamic role.As a large amount of pedestrian movements, social force and pedestrian's direction of travel is caused to have one since present position is interlaced
Clamp angle, this can make pedestrian's generation is visual to pass round behavior；But ought in special circumstances, for example two opposite pedestrians exist
It walks on straight line, point-blank, the program will lead to pedestrian can only be repeatedly for social repulsion direction and the direction of motion
Collision is met and cannot normally be passed round.
Summary of the invention
The Dense crowd emulation mode based on intelligent body that the embodiment provides a kind of, with realize it is true, have
Effect ground emulates Dense crowd.
The present invention provides following schemes:
A kind of Dense crowd emulation mode based on intelligent body, comprising:
The corresponding three circles manikin of pedestrian is constructed, the circle that the three circle manikin is intersected by three forms, wherein in
Between great circle indicate that trunk, the roundlets of two sides indicate two shoulders；
The clear distance between the corresponding three circles manikin of two pedestrians is calculated, using the clear distance as described two pedestrians
The distance between.
Further, the size value of the three circle manikins is shown in Table 1, wherein RsFor small radius of circle, RtFor
Big radius of circle, Rd=Rs+Rt, RdUsing normal distribution, speed is used and is uniformly distributed；
Table 1
Further, the clear distance calculated between the corresponding three circles manikin of two pedestrians, by the clear distance
As the distance between described two pedestrians, comprising:
If the center of circle of the three circle manikins of the first pedestrian is respectively as follows: E1、E2And E3, the three circle manikins of the second pedestrian
The center of circle be respectively as follows: F1、F2And F3；
l1For E1And F1Between Euclidean distance, l2For E1And F2Between Euclidean distance, l2For E1And F3Between
Euclidean distance, l4For E2And F1Between Euclidean distance, l5For E2And F2Between Euclidean distance, l6For E2
And F3Between Euclidean distance, l7For E3And F1Between Euclidean distance, l8For E3And F2Between Euclid away from
From l9For E3And F3Between Euclidean distance；
The distance between two pedestrians lminCalculation formula it is as follows:
lmin=min { l1,l2,l3,l4,l5,l6,l7,l8,l9(formula 2)
Further, the method further include:
The angular field of view of pedestrian is scanned by several visual field blocks, the pedestrian for traversing surrounding is angularly divided into, is remembered
Record falls in pedestrian in each visual field block within sweep of the eye, and calculates the averag density of each visual field block one skilled in the art, then and i-th
The calculation formula of the averag density of block visual field block one skilled in the art is as follows:
Wherein, n indicates the quantity of i-th piece of visual field block one skilled in the art, ρikIndicate area locating for i-th piece of visual field block one skilled in the art k
The crowd density in domain；
Pedestrian relatively finds out the smallest visual field block of pedestrian density in each visual field block as the direction of travel of oneself,
And target point is temporarily converted to point nearest apart from oneself in the smallest visual field block of the pedestrian density.
Further, when pedestrian nearest at a distance from pedestrian A with locating region is not overlapped mutually, the pedestrian A institute
Locate the crowd density ρ in regionikCalculation method it is as follows:
ρA=1/ [AA(1+d1/rA)2]
When nearest pedestrian overlaps each other at a distance from pedestrian A with locating region, the crowd in region locating for the pedestrian A
Density pikCalculation method it is as follows:
ρA=1/ [(AA-AAB)]
Wherein, ρAIndicate the density of crowd locating for pedestrian A, AAIndicate the body projected area of pedestrian A, d1It is pedestrian A and institute
Locate nearest the distance between the pedestrian of distance in region, rAIndicate distance d1The radius of corresponding pedestrian A body circle, AABIt indicates
Pedestrian A in locating region at a distance from nearest pedestrian's lap area.
Further, the method further include:
The calculation method for passing round power between pedestrian is as follows:
1) judge whether two pedestrians are opposite pedestrians, namely judge whether the velocity angle of two pedestrians is greater than 90 °；Such as
Fruit meets condition and then carries out in next step；
2) whether judge to predict two pedestrians it is necessary to evade, carried out as γ < θ in next step, wherein γ is pedestrian's fortune
It moves velocity vector and from the angle between the vector that self-position is directed toward nearest pedestrian position, θ is that two pedestrians wipe shoulder just
And speed and relative position angle when not having to hide are crossed, calculation formula is as follows:
Wherein, b1、b2It is two elliptical long axial lengths,It is two pedestrian centers away from the projection in x-axis；
3) pedestrian is judged whether in the sphere of action for passing round power, is calculated and between all pedestrians in sphere of action
Distance, and find out apart from nearest people；
4) it calculates and passes round power between nearest pedestrian, according to the calculated power of passing round to pedestrian and apart from most
Close pedestrian carries out passing round processing, passes round powerCalculation formula it is as follows；
Wherein, AC、BCIt is the action intensity and sphere of action for passing round power, size is determined by calibration simulated experiment；
ω is azimuthal influence coefficient,It is the normal direction of pedestrian's speed；
Above-mentioned calibration simulated experiment refers to and establishes a simulated scenario, wherein two opposite pedestrians connect along two people present positions
Line is walked in opposite directions, constantly adjusts analog parameter until pedestrian can be just around other side position be avoided, azimuthal influence coefficient is according to the following formula
It calculates:
Wherein, λ ∈ [0,1] is form factor, and the value of λ is used to adjust the value of azimuthal influence coefficient, and value is smaller, advances
The subsequent influence coefficient in direction is smaller；Value is bigger, and the subsequent influence coefficient of direction of advance is bigger.
As can be seen from the technical scheme provided by the above-mentioned embodiment of the present invention, the embodiment of the present invention passes through three circle human moulds
Type solves the untrue property of circular granular simulation pedestrian's body, dexterously under the premise of not increasing algorithm complexity to gather around
The calculating of pedestrian's individual physical contact force and highdensity calculating are laid a good foundation in the crowd of squeezing.
Crowd density calculation method, the considerations of making crowd density from individual angle fining, is described with the method for quantization
Each pedestrian proposes the calculating and analysis of pedestrian's individual physical contact force the subjective feeling of itself local environment, this method
Having supplied may.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, required use in being described below to embodiment
Attached drawing be briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this
For the those of ordinary skill of field, without any creative labor, it can also be obtained according to these attached drawings other
Attached drawing.
Fig. 1 is a kind of social force model schematic diagram provided in an embodiment of the present invention；
Fig. 2 is a kind of three circles manikin schematic diagram provided in an embodiment of the present invention；
Fig. 3 is the free distance computation schematic diagram in a kind of three circles manikin provided in an embodiment of the present invention；
Fig. 4 is a kind of people's angular field of view schematic diagram provided in an embodiment of the present invention；
Fig. 5 is that a kind of visual angle scanning theory diagram provided in an embodiment of the present invention is intended to；
Fig. 6 is a kind of addition density scan algorithm descendant's flow point Butut provided in an embodiment of the present invention；
Fig. 7 is a kind of action principle schematic diagram for the power of passing round provided in an embodiment of the present invention；
Fig. 8 is a kind of schematic diagram of collision avoidance judgment rule provided in an embodiment of the present invention；
Fig. 9 is opposite pedestrian stream emulation signal after a kind of addition prediction of collision Hedging mechanism provided in an embodiment of the present invention
Figure；
Figure 10 is a kind of human body stiffness measuring schematic diagram provided in an embodiment of the present invention；
Figure 11 is a kind of calculating schematic diagram of pedestrian's occupied area provided in an embodiment of the present invention.
Specific embodiment
Embodiments of the present invention are described below in detail, the example of the embodiment is shown in the accompanying drawings, wherein from beginning
Same or similar element or element with the same or similar functions are indicated to same or similar label eventually.Below by ginseng
The embodiment for examining attached drawing description is exemplary, and for explaining only the invention, and is not construed as limiting the claims.
Those skilled in the art of the present technique are appreciated that unless expressly stated, singular " one " used herein, " one
It is a ", " described " and "the" may also comprise plural form.It is to be further understood that being arranged used in specification of the invention
Diction " comprising " refer to that there are the feature, integer, step, operation, element and/or component, but it is not excluded that in the presence of or addition
Other one or more features, integer, step, operation, element, component and/or their group.It should be understood that when we claim member
Part is " connected " or when " coupled " to another element, it can be directly connected or coupled to other elements, or there may also be
Intermediary element.In addition, " connection " used herein or " coupling " may include being wirelessly connected or coupling.Wording used herein
"and/or" includes one or more associated any cells for listing item and all combinations.
Those skilled in the art of the present technique are appreciated that unless otherwise defined, all terms used herein (including technology art
Language and scientific term) there is meaning identical with the general understanding of those of ordinary skill in fields of the present invention.Should also
Understand, those terms such as defined in the general dictionary, which should be understood that, to be had and the meaning in the context of the prior art
The consistent meaning of justice, and unless defined as here, it will not be explained in an idealized or overly formal meaning.
In order to facilitate understanding of embodiments of the present invention, it is done by taking several specific embodiments as an example below in conjunction with attached drawing further
Explanation, and each embodiment does not constitute the restriction to the embodiment of the present invention.
The embodiment of the present invention enables pedestrian to perceive to construct and realize with the agent model of real human's characteristic
Thus surrounding population density makes behaviour decision making and adjusts behavior state.Propose a kind of Dense crowd based on intelligent body
Emulation mode, this method can demarcate the experimental method of human body rigidity from host computer and psychological forces sphere of action determines method；
And propose a kind of crowd density calculation method based on pedestrian's individual, it is more suitable for the calculating of Dense crowd density.
1: three circle manikin
The embodiment of the present invention is optimal in order to merge round calculated performance, and ellipse simulation two-dimension human body project it is optimal
Two big advantages construct three circle manikins.A kind of three circle manikin schematic diagram such as Fig. 2 institute provided in an embodiment of the present invention
Show, the circle which is intersected by three forms, wherein intermediate great circle indicates that trunk, the roundlet of two sides indicate two shoulders；By
This pedestrian's body two-dimensional projection obtained and ellipse are more approximate, while three circles also have specific physical significance, ginseng
Number calibration is more convenient.
Therefore, the embodiment of the present invention carries out the modeling of human body two-dimensional projection by three circle manikins of realization perfect；Meanwhile
In order to enable three circle manikins to embody Chinese visible human feature, to " Chinese adult human dimension (GB 10000-1988) "
In human dimension analyzed, and combine document in dimension model, propose three circle manikins size values, be shown in Table
1, wherein RsFor small radius of circle, RtFor big radius of circle, Rd=Rs+Rt, RdUsing normal distribution.
1 moulded dimension of table and velocity information
After three circle models, calculating for most short clear distance can be by calculating the clear distance between 9 circles and circle between pedestrian
It solves, the free distance computation schematic diagram in a kind of three circles manikin provided in an embodiment of the present invention is as shown in figure 3, calculation formula
See formula (2)
If the center of circle of the three circle manikins of the first pedestrian is respectively as follows: E1、E2And E3, the three circle manikins of the second pedestrian
The center of circle be respectively as follows: F1、F2And F3。
l1For E1And F1Between Euclidean distance, l2For E1And F2Between Euclidean distance, l2For E1And F3Between
Euclidean distance, l4For E2And F1Between Euclidean distance, l5For E2And F2Between Euclidean distance, l6For E2
And F3Between Euclidean distance, l7For E3And F1Between Euclidean distance, l8For E3And F2Between Euclid away from
From l9For E3And F3Between Euclidean distance.
lmin=min { l1,l2,l3,l4,l5,l6,l7,l8,l9}
(formula 2)
lminFor the distance between two pedestrians.
2: agent model
Pedestrian is as a kind of individual with wisdom and independent decision-making ability, available ambient condition information and other rows
People interacts；Therefore, the embodiment of the present invention can more effectively solve the row for relying on power drive merely by building intelligent body
What people's simulation model was shown does not meet actual behavior.Agent model is passed through based on the calculating of the power of social force model
Environmental information in visual perception scope of sight, such as barrier, walkable region, other pedestrians and destination orientation etc.；
Force Calculation is carried out by information in scope of sight, and behaviour decision making is made by environmental information, is calculated according to the result of decision and power
As a result behavior state is adjusted.For the communication exchange for expressing intelligent body and external environment, perception mechanism is added to based on pedestrian's individual,
That is the visible sensation method of pedestrian；It is expression intelligent body to the processing capacity of external information, decision-making mechanism is added to based on pedestrian's individual,
That is decision-making technique of the pedestrian according to current context information；It is expression intelligent body to the respond of external information, based on pedestrian
Body is added to state Regulation mechanism, i.e. pedestrian's method for adjusting current state according to the result of decision.Based on agent model, realize
Pedestrian's method for optimizing route and opposite pedestrian based on crowd density pass round method etc..
3: the local path optimization method based on density
This method is mainly exactly that pedestrian carries out statistics calculating to the density of itself field range one skilled in the art, and select accordingly compared with
For comfortably efficiently walking path, the optimization degree of Path selection is improved by changing visual angle piecemeal.According to the visual angle model of pedestrian
It encloses to calculate pedestrian's ambient density, and determines its final direction of travel according to the density and its target bearing, here it is close
Spend the core concept of scanning algorithm.In order to more realistically describe pedestrian to surrounding row person's development, density is introduced to horizon range
Influencing mechanism.
The basis that the algorithm is realized is the angular field of view for first finding out pedestrian, by inspection information it is found that the Binocular vison of pedestrian
Angle can reach 200 ° or so, however actually pedestrian is only more sensitive to the things of two 120 ° of ranges in side at the moment, within the scope of other
Susceptibility it is relatively low, a kind of people's angular field of view schematic diagram provided in an embodiment of the present invention is as shown in figure 4, using pedestrian visual angle
It is 120 °, horizon range is 5m to be realized.
A kind of visual angle scanning theory diagram provided in an embodiment of the present invention is intended to as shown in Figure 5, it is contemplated that needs to optimize row
The direction of motion of people, by the angular field of view of pedestrian by being angularly divided into several regions, the precision of the size of angle as required
It takes, for example takes 30.
Then the pedestrian around traversal is scanned, and record falls in the pedestrian of each block within sweep of the eye.And it calculates each
The averag density of a block one skilled in the art, then the density calculation formula of i-th piece of block is shown in formula (3).
Wherein, n indicates the quantity of the i-th visual field block one skilled in the art, ρikIndicate people from region locating for the i-th visual field block one skilled in the art k
Group's density, ρikCalculation method be described below.
Pedestrian relatively finds out the smallest visual field block of pedestrian density in each visual field block as the direction of travel of oneself,
And target point is automatically temporarily converted to the point nearest apart from oneself in the region, driving force is directed toward the direction, i.e. generation at this time
The direction of motion of table row people.If the pedestrian density on pedestrian both sides is symmetry equivalent or is all zero, pedestrian can walk along the former direction of motion.
When finding density minimum area of visual field, pedestrian will consider the drift angle size with oneself former target point.
In view of the field range that pedestrian pays close attention to can be also varied with surrounding population density: when ambient density is smaller
When, for the field range of pedestrian than broader, sighting distance is larger；But when density is larger more crowded, pedestrian can only pay close attention to feelings at the moment
Condition, sighting distance are smaller.It joined the algorithm with density adjustment sighting distance in model thus.It is close in the pedestrian that every piece of field range has been calculated
After degree, the averag density ρ for being averaged to obtain pedestrian's whole visual field range is summed it up, carrys out the functional relation tune according to setting accordingly
Itself whole horizon range.The functional relation present invention is indicated using the piecewise function as shown in formula (4), passes through survey
It is more satisfactory to try effect.
A kind of addition density scan algorithm descendant's flow point Butut provided in an embodiment of the present invention is as shown in fig. 6, be added density
By multiple pedestrian simulation it has been observed that pedestrian can swing the run trace of oneself constantly to seek impedance most after scanning algorithm
It is small, most efficiently walking path, walking process Density Distribution are more uniform.
4, opposite pedestrian passes round method
A kind of action principle schematic diagram of power of passing round provided by the invention is as shown in fig. 7, pedestrian A is gone with speed VA movement
For people B with speed VB movement, then it is α with VA angle that pedestrian A, which is RV relative to the speed of pedestrian B,；Pedestrian A needs to pass round pedestrian B
So that relative velocity is rotated angle beta with acceleration a again, keep it tangent with the personal space of pedestrian B.Then the calculating of acceleration a is public
Shown in formula such as formula (5)；
A=VA × cos α × tan β
(formula 5)
The present invention introduces relevant Rule of judgment and calculation method on the basis of using for reference its thinking.Pass through point before
Analysis discovery, the proposition of the proposition and social force of passing round power have very big similitude, and the most important is exactly both psychological
Effect, therefore similar expression-form can be used.It is formula (6) that power formula is then passed round used in the present invention:
Wherein, AC、BCIt is the action intensity and sphere of action for passing round power, size is determined by calibration simulated experiment；
ω is azimuthal influence coefficient,It is the normal direction of pedestrian's speed.
Above-mentioned calibration simulated experiment refers to and establishes a simulated scenario, wherein two opposite pedestrians connect along two people present positions
Line is walked in opposite directions, constantly adjusts analog parameter until pedestrian can be just around avoiding other side position.Azimuthal influence coefficient is according to the following formula
It calculates:
Wherein, λ ∈ [0,1] is form factor, the value of the adjustable azimuthal influence coefficient of its value；Value is smaller, preceding
It is smaller into the subsequent influence coefficient in direction；Value is bigger, and the subsequent influence coefficient of direction of advance is bigger.
It should be noted that passing round power whole effect unlike social force, it needs certain Rule of judgment, right
Need to consider direction, the position etc. of pedestrian in the formulation of the action condition.Its judgment step is as follows:
1) judge whether two pedestrians are opposite pedestrians, namely judge whether the two velocity angle is greater than 90 °, velocity angle
Refer to the angle of two pedestrian's velocity vectors；It is carried out if meeting condition in next step.
2) whether judge to predict two pedestrians it is necessary to evade, rule principle is as shown in Figure 8.It is carried out as γ < θ next
Step, wherein γ is pedestrian movement's velocity vector and from the angle between the vector that self-position is directed toward nearest pedestrian position,
θ is that two pedestrians are brushed past just without speed when hiding and relative position angle, and calculation formula is such as shown in (7):
Wherein, b1、b2It is two elliptical long axial lengths,It is two pedestrian centers away from the projection in x-axis.
3) judge that pedestrian whether in the sphere of action for passing round power, calculates and the institute in sphere of action according to formula 2
There is the distance between pedestrian, and finds out apart from nearest people.
4) it is calculated according to formula 6 and passes round power between nearest pedestrian, according to the calculated power of passing round to row
It people and carries out passing round processing apart from nearest pedestrian.
Opposite pedestrian stream emulates schematic diagram such as Fig. 9 after a kind of addition prediction of collision Hedging mechanism provided in an embodiment of the present invention
It is shown.Being added after the algorithm has very big improvement for the walking decision of pedestrian, by different numbers, different situations downlink
It is obvious that people carries out analogue simulation discovery effect.
5 parameter calibrations and calculation method
5.1 human body Rigidity Experiment standardizations
Organization of human body is complicated, and overall stiffness has anisotropy；In addition to integumentary musculature, under crowed condition, bone frame
Also there is certain compressibility, directly measuring human body rigidity, there are larger difficulties.Therefore, the embodiment of the present invention assumes human body rigidity
It is uniformly distributed, does not change with physical feeling, and by the limb flexibility and skeleton compressibility two parts of attachment skin and muscle
Composition.Wherein, limb flexibility is obtained by experimental calculation, and skeleton compressibility with the fitting of actual extruding force data by obtaining.
Human body limb is assumed to be a spring-damping system, subject lies in physical feeling to be measured naturally
On experimental bench, by loading device corresponding site apply a constant load after slowly under have an high regard for object then order for discharge its freely shake
Dynamic, a kind of human body stiffness measuring schematic diagram provided in an embodiment of the present invention is as shown in Figure 10.The pressure sensor being attached on limbs
Record vibration time-history curves, can calculate system angular frequency and load mass m using oscillating curve.For have damping from
Seen formula (8) by vibration motion equation, system angle frequency equation is shown in formula (9)
Wherein, c is system damping, and m is the load mass for being applied to test position, k0For system resilience modulus.
System resilience formula of modulus can be obtained by formula (9) and see formula (10), wherein damping combines vibration time-history curves by transporting
Dynamic equation solution, is shown in formula (11).Bone frame elastic is considered using skeleton compressed coefficient C, finally obtains human body rigidity expression formula
See formula (12).
k0=m (ω2+c2/4m2) (formula 10)
K=C*k0(formula 12)
Wherein, A0For the amplitude of vibrational system, t is the time, since system vibrate when calculate,For vibration
Initial phase, C are the skeleton compressed coefficient, k0For limbs rigidity, k is human body rigidity.
5.2 crowd density calculation methods
Use for reference Thiessen polygon method in from pedestrian's individual angle consider crowd density thinking, with pedestrian A to be studied
The ratio and pedestrian A body perspective plane to be studied of the shortest distance of nearest pedestrian B and corresponding body radius of circle within sweep of the eye
Product calculates area occupied by pedestrian A for foundation, shown in the dash area as shown in character A in Figure 11, so that it is determined that pedestrian A institute
The density of place crowd, when nearest pedestrian is not overlapped mutually at a distance from pedestrian A with locating region, calculation formula is shown in formula
(13)。
ρA=1/ [AA(1+d1/rA)2] (formula 13)
When nearest pedestrian overlaps each other at a distance from pedestrian A with locating region, i.e., when two pedestrians squeeze overlapping, with
The body projected area of pedestrian A to be studied, which is subtracted, calculates area occupied by pedestrian A with the squish area of pedestrian B for foundation, such as
Shown in Figure 11 dash area, so that it is determined that the crowd density of pedestrian A at this time, is shown in formula (14).
ρA=1/ [(AA-AAB)] (formula 14)
Wherein, ρAIndicate the density of crowd locating for pedestrian A, AAIndicate the body projected area of pedestrian A, d1It is pedestrian A and institute
Locate nearest the distance between the pedestrian of distance in region, is calculated using formula 2, rAIndicate distance d1Corresponding pedestrian A body circle
Radius, that is, d is calculated1The radius of used pedestrian A body circle.
AABIndicate the area of two pedestrian's laps.
5.3 psychological forces sphere of actions determine method
Psychological forces simulate pedestrian as a kind of fictitious force and keep certain personal empty between stranger under normal circumstances
Between demand.Personal space is the concept in environmental psychology, size with the psychological condition of individual and the variation of environment and
Dynamic change, especially in crowded public place, when space length needed for environment is unable to satisfy itself, personal space can root
It is shunk according to ambient conditions.Therefore, Most models by the method for psychological forces sphere of action value of fixed constant are not conform at present
Reason；The basic reason for needing to analyse in depth comfortable space generation, thus redefines the obtaining value method of personal space.
Personal space is divided into comfortable space and the rank of safe space two；It is to guarantee meeting that wherein comfortable space, which is pedestrian,
The necessary distance to be kept can be made a response and adjusted calmly when to special circumstances, and safe space is that pedestrian is to guarantee meeting
It can quickly make a response when to special circumstances and necessary distance that Emergency avoidance to be kept.As a result, in conjunction with autonomous Design
Experimental result and investigation result, the distance range for extrapolating comfortable space under normal circumstances and safe space are respectively
Dc∈[1.15m,2.0m],Ds∈[0.4m,0.9m]
However, becoming smaller as crowd density increases pedestrian's speed of travel, while being important to note that ring around under crowded environment
Border, it is rapider to the reaction of environmental change；Required personal space can become smaller therewith, and pedestrian can give up under certain density conditions
It abandons comfortable space and retains necessary safe space.Therefore, it is considered herein that working as region crowd density locating for pedestrian cannot be guaranteed just
When safe space demand in normal situation, pedestrian gives up comfortable space and only retains safe space, i.e. pedestrian can be in the reaction time
Inside stop immediately；When crowd density can meet comfortable space demand under normal circumstances very well, the personal space of pedestrian is
Comfortable space, i.e. pedestrian are on the basis of guaranteeing the reaction time plus previous step buffers and the distance of a successive step；Work as crowd density
In between the two when, pedestrian need on the basis of guaranteeing the reaction time add a successive step distance, adjustment distance AαMeter
It is following (15) to calculate formula.
Wherein, v is the speed of travel, and Δ t is the reaction time, S be it is long step by step, ρ is the crowd of pedestrian region crowd
Density, ρDIt is d1The crowd density calculated when taking v Δ t+2S, ρdIt is d1The crowd density being calculated when taking v Δ t.d1It is the row
People in locating region at a distance from nearest the distance between pedestrian, calculated using formula 2.
In conclusion the embodiment of the present invention passes through three circle manikins under the premise of not increasing algorithm complexity, it is ingenious
Ground solve circular granular simulation pedestrian's body untrue property, be crowded crowd in pedestrian's individual physical contact force calculating with
And highdensity calculating is laid a good foundation.
Crowd density calculation method, the considerations of making crowd density from individual angle fining, is described with the method for quantization
Each pedestrian experiences the supervisor of itself local environment, and this method proposes the calculating and analysis of pedestrian's individual physical contact force
Having supplied may.
Agent model is constructed according to the decision reflex action in pedestrian's walking process, mainly to crowd density and opposite row
People constructs perception, decision and reflex action, and effect is simulated when carrying out channel bidirectional flow scene simulation and more meets reality.
The measuring method of human body elasticity modulus can measure the true elasticity modulus of human body by laboratory facilities, make model
Value is truer；The elasticity modulus measuring method can measure the human body elasticity modulus under the conditions of different clothing simultaneously, to grind
Dense crowd emulation improves method basis under the conditions of studying carefully Various Seasonal difference clothing amount.
The determination method of psychological forces sphere of action elaborates the physics meaning of psychological forces sphere of action from psychologic angle
The foundation of justice, parameter value is stronger.
Those of ordinary skill in the art will appreciate that: attached drawing is the schematic diagram of one embodiment, module in attached drawing or
Process is not necessarily implemented necessary to the present invention.
As seen through the above description of the embodiments, those skilled in the art can be understood that the present invention can
It is realized by the mode of software plus required general purpose hardware unit.Based on this understanding, technical solution of the present invention essence
On in other words the part that contributes to existing technology can be embodied in the form of software products, the computer software product
It can store in storage medium, such as ROM/RAM, magnetic disk, CD, including some instructions are used so that a computer equipment
(can be personal computer, server or the network equipment etc.) executes the certain of each embodiment or embodiment of the invention
Method described in part.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for device or
For system embodiment, since it is substantially similar to the method embodiment, so describing fairly simple, related place is referring to method
The part of embodiment illustrates.Apparatus and system embodiment described above is only schematical, wherein the conduct
The unit of separate part description may or may not be physically separated, component shown as a unit can be or
Person may not be physical unit, it can and it is in one place, or may be distributed over multiple network units.It can root
According to actual need that some or all of the modules therein is selected to achieve the purpose of the solution of this embodiment.Ordinary skill
Personnel can understand and implement without creative efforts.
The foregoing is only a preferred embodiment of the present invention, but scope of protection of the present invention is not limited thereto,
In the technical scope disclosed by the present invention, any changes or substitutions that can be easily thought of by anyone skilled in the art,
It should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with scope of protection of the claims
Subject to.
Claims (3)
1. a kind of Dense crowd emulation mode based on intelligent body characterized by comprising
The corresponding three circles manikin of pedestrian is constructed, the circle that the three circle manikin is intersected by three forms, wherein intermediate
Great circle indicates that trunk, the roundlet of two sides indicate two shoulders；
The clear distance between the corresponding three circles manikin of two pedestrians is calculated, using the clear distance as between described two pedestrians
Distance, specific calculating process includes:
If the center of circle of the three circle manikins of the first pedestrian is respectively as follows: E1、E2And E3, the circle of the three circle manikins of the second pedestrian
The heart is respectively as follows: F1、F2And F3；
l1For E1And F1Between Euclidean distance, l2For E1And F2Between Euclidean distance, l3For E1And F3Between Europe
Distance, l are obtained in several4For E2And F1Between Euclidean distance, l5For E2And F2Between Euclidean distance, l6For E2And F3
Between Euclidean distance, l7For E3And F1Between Euclidean distance, l8For E3And F2Between Euclidean distance, l9
For E3And F3Between Euclidean distance；
The distance between two pedestrians lminCalculation formula it is as follows:
lmin=min { l1,l2,l3,l4,l5,l6,l7,l8,l9}
The angular field of view of pedestrian is scanned by several visual field blocks, the pedestrian for traversing surrounding is angularly divided into, record is fallen
Pedestrian in each visual field block within sweep of the eye, and calculate the averag density of each visual field block one skilled in the art, then it regards for i-th piece
The calculation formula of the averag density of wild block one skilled in the art is as follows:
Wherein, n indicates the quantity of i-th piece of visual field block one skilled in the art, ρikIndicate region locating for i-th piece of visual field block one skilled in the art k
Crowd density；
Pedestrian relatively finds out the smallest visual field block of pedestrian density in each visual field block as the direction of travel of oneself, and will
Target point is temporarily converted to point nearest apart from oneself in the smallest visual field block of the pedestrian density；
When nearest pedestrian is not overlapped mutually at a distance from pedestrian A with locating region, the crowd in region locating for the pedestrian A is close
Spend ρikCalculation method it is as follows:
ρA=1/ [AA(1+d1/rA)2]
When nearest pedestrian overlaps each other at a distance from pedestrian A with locating region, the crowd density in region locating for the pedestrian A
ρikCalculation method it is as follows:
ρA=1/ [(AA-AAB)]
Wherein, ρAIndicate the density of crowd locating for pedestrian A, AAIndicate the body projected area of pedestrian A, d1It is pedestrian A and locating area
Nearest the distance between the pedestrian of distance in domain, rAIndicate distance d1The radius of corresponding pedestrian A body circle, AABIndicate pedestrian
A in locating region at a distance from nearest pedestrian's lap area.
2. the Dense crowd emulation mode according to claim 1 based on intelligent body, which is characterized in that three circles
The size value of manikin is as shown in the table, wherein RsFor small radius of circle, RtFor big radius of circle, Rd=Rs+Rt, RdUsing just
State distribution, speed are as shown in the table using being uniformly distributed:
3. the Dense crowd emulation mode according to claim 1 based on intelligent body, which is characterized in that the method
Further include:
The calculation method for passing round power between pedestrian is as follows:
1) judge whether two pedestrians are opposite pedestrians, namely judge whether the velocity angle of two pedestrians is greater than 90 °；If full
Sufficient condition then carries out in next step；
2) whether judge to predict two pedestrians it is necessary to evade, carried out as γ < θ in next step, wherein γ is pedestrian movement's speed
It spends vector and from the angle between the vector that self-position is directed toward nearest pedestrian position, θ is that two pedestrians brush past just
Speed and relative position angle, calculation formula are as follows when without hiding:
Wherein, b1、b2It is two elliptical long axial lengths,It is two pedestrian centers away from the projection in x-axis,
3) judge pedestrian whether in the sphere of action for passing round power, calculate and between all pedestrians in sphere of action away from
From, and find out apart from nearest people；
4) it calculates and passes round power between nearest pedestrian, according to the calculated power of passing round to pedestrian and apart from nearest
Pedestrian carries out passing round processing, passes round powerCalculation formula it is as follows；
Wherein, AC、BCIt is the action intensity and sphere of action for passing round power, size is determined by calibration simulated experiment；ω is
Azimuthal influence coefficient,It is the normal direction of pedestrian's speed, rαβIndicate the sum of pedestrian α and the radius of pedestrian β, dαβIndicate pedestrian α and
Distance between the mass center of pedestrian β；
Above-mentioned calibration simulated experiment refers to and establishes a simulated scenario, wherein two opposite pedestrians are along two people present position line phases
To walking, analog parameter is constantly adjusted until pedestrian can be just around other side position be avoided, azimuthal influence coefficient calculates according to the following formula:
Wherein, λ ∈ [0,1] is form factor, and the value of λ is used to adjust the value of azimuthal influence coefficient, and value is smaller, direction of advance
Subsequent influence coefficient is smaller；Value is bigger, and the subsequent influence coefficient of direction of advance is bigger.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610327242.0A CN106096072B (en) | 2016-05-17 | 2016-05-17 | Dense crowd emulation mode based on intelligent body |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610327242.0A CN106096072B (en) | 2016-05-17 | 2016-05-17 | Dense crowd emulation mode based on intelligent body |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106096072A CN106096072A (en) | 2016-11-09 |
| CN106096072B true CN106096072B (en) | 2019-06-25 |
Family
ID=57230040
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610327242.0A Active CN106096072B (en) | 2016-05-17 | 2016-05-17 | Dense crowd emulation mode based on intelligent body |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106096072B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107665269B (en) * | 2017-08-11 | 2021-01-08 | 山东师范大学 | Rapid crowd evacuation simulation method and device based on geographic information |
| JP6623203B2 (en) * | 2017-10-24 | 2019-12-18 | 日本電信電話株式会社 | Personal space estimation device, personal space estimation method, and personal space estimation system |
| CN107909190B (en) * | 2017-10-27 | 2021-06-29 | 天津理工大学 | Dynamic prediction simulation device for crowd evacuation behaviors in case of toxic gas leakage accident and working method thereof |
| CN108256155B (en) * | 2017-12-20 | 2021-03-26 | 同济大学 | Passenger getting-off point selection method for T-junction passenger car |
| CN108170995B (en) * | 2018-01-29 | 2020-09-18 | 北京航空航天大学 | Pedestrian motion simulation method and system |
| CN108717596A (en) * | 2018-04-19 | 2018-10-30 | 北京交通大学 | The passenger flow traffic efficiency evaluation method in T fonts channel in track traffic station |
| CN108961313B (en) * | 2018-06-29 | 2021-06-29 | 大连民族大学 | Overlooking pedestrian risk quantification method of two-dimensional world coordinate system |
| CN109190325B (en) * | 2018-11-20 | 2019-05-28 | 山东师范大学 | Crowd evacuation Path Planning Simulation method based on the analysis of pedestrian's crowding |
| CN111339668B (en) * | 2020-02-28 | 2022-05-10 | 西南交通大学 | Crowd evacuation visualization method based on emotion cognition |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104239636A (en) * | 2014-09-16 | 2014-12-24 | 北京航空航天大学 | Fire emergency evacuation simulation method |
| CN104574352A (en) * | 2014-09-02 | 2015-04-29 | 重庆大学 | Crowd density grade classification method based on foreground image |
-
2016
- 2016-05-17 CN CN201610327242.0A patent/CN106096072B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104574352A (en) * | 2014-09-02 | 2015-04-29 | 重庆大学 | Crowd density grade classification method based on foreground image |
| CN104239636A (en) * | 2014-09-16 | 2014-12-24 | 北京航空航天大学 | Fire emergency evacuation simulation method |
Non-Patent Citations (4)
| Title |
|---|
| Agent-based continuous-space;陈峰等;《Proceedings of the Institution of Civil Engineers》;20151231;第336-345页 |
| FDS+Evac人员疏散模拟软件的理论探析与实战技巧;颜春萍等;《消防技术与产品信息》;20080715;第8-12页 |
| 基于行人仿真的轨道交通车站设施规模及布局研究;王子甲;《中国博士学位论文全文数据库 工程科技II辑》;20140515;第C033-17页 |
| 高层建筑火灾时的人员安全疏散研究;王爱莎;《中国优秀硕士学位论文全文数据库 工程科技II辑》;20140515;第C038-411页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106096072A (en) | 2016-11-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106096072B (en) | Dense crowd emulation mode based on intelligent body | |
| Palechano et al. | Virtual crowds: Methods, simulation, and control | |
| Martinez-Gil et al. | Modeling, evaluation, and scale on artificial pedestrians: a literature review | |
| Bruneau et al. | Going through, going around: A study on individual avoidance of groups | |
| Pelechano Gómez et al. | Crowd simulation incorporating agent psychological models, roles and communication | |
| Massaguer et al. | Multi-agent simulation of disaster response | |
| Dai et al. | Simulation of pedestrian counter flow through bottlenecks by using an agent-based model | |
| Radianti et al. | Crowd models for emergency evacuation: A review targeting human-centered sensing | |
| Ronchi et al. | New approaches to evacuation modelling for fire safety engineering applications | |
| Song et al. | Simulation of pedestrian rotation dynamics near crowded exits | |
| Manenti et al. | An agent-based proxemic model for pedestrian and group dynamics: motivations and first experiments | |
| Cheng et al. | A review of pedestrian group dynamics and methodologies in modelling pedestrian group behaviours | |
| Fridman et al. | The impact of culture on crowd dynamics: An empirical approach | |
| Che et al. | A novel simulation framework based on information asymmetry to evaluate evacuation plan | |
| Ma et al. | Pedestrian dynamics in real and simulated world | |
| Asano et al. | A pedestrian model considering anticipatory behaviour for capacity evaluation | |
| Kremer et al. | Modelling distracted agents in crowd simulations | |
| Wang et al. | An immersive multi-agent system for interactive applications | |
| CN105205216B (en) | The cross-cutting emulation mode of extensive pedestrian movement and device in hinge | |
| Skondras et al. | A route selection scheme for supporting virtual tours in sites with cultural interest using drones | |
| Torrens | Behavioral intelligence for geospatial agents in urban environments | |
| Torrens | Exploring behavioral regions in agents’ mental maps | |
| Szymanezyk et al. | Agent-based crowd simulation in airports using games technology | |
| CN108629121B (en) | Virtual reality crowd simulation method and system based on terrorist valley effect avoidance | |
| Curtis et al. | Virtual Tawaf: A velocity-space-based solution for simulating heterogeneous behavior in dense crowds |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |