51 results about "Crowd simulation" patented technology

The present invention discloses a computer simulation system for large-size crowd evacuation and a method therefor. The system comprises: an evacuation scene model establishing unit, used for extracting a feature of an evacuation scene, so as to obtain a three-dimensional model of the evacuation scene; an evacuation scene global path planning unit, used for performing a global path plan on the evacuation scene according to the three-dimensional model of the evacuation scene, and calculating global paths of all entrances and exits in the evacuation scene; a crowd activity generation unit, used for performing crowd grouping according to the feature of the evacuation scene, a relationship among individuals and individual position information, so as to obtain a plurality of groups; and a crowd simulation unit, used for screening a leader of each group and calculating a real-time speed of an individual according to a modified social force model. In a simulation process, a moving direction of each group is decided by the leader, and the motion speed of the individual is adjusted in real time, so as to maintain travel consistency of each group and achieve behavior simulation of evacuated crowd.

The invention provides an underground emergency evacuation virtual crowd simulation method based on multiple intelligent agents. A virtual mine emergency evacuation crowd simulation frame is built by using the underground emergency evacuation virtual crowd simulation method based on the multiple intelligent agents. The virtual mine emergency evacuation crowd simulation frame is mainly composed of five portions of a crowd simulation engine, a geometric scene simulation engine, a scene database, an event recorder and a human-computer interaction interface. In consideration of different understandings on mine knowledge and different behavior decisions of miners playing different roles when mine disasters occur, a multiple intelligent agent miner crowd emergency evacuation module and a crowd emergency evacuation algorithm for the different roles of the miners such as job foremen, expert miners, common miners and rescuers are built by using the underground emergency evacuation virtual crowd simulation method based on the multiple intelligent agents according to a role theory. The underground emergency evacuation virtual crowd simulation method based on the multiple intelligent agents can simulate true situations of a mine well, has good universality and verisimilitude, and can provide effective technical support for underground security accident analysis and prevention, mine safety training and rescue exercises.

The invention discloses a simulation system and method of an evacuation crowd behavior based on grid-density-relation. The system comprises an evacuation scene model building unit, an evacuation scene global path planning unit, a crowd activity generation unit and a crowd simulation unit, wherein the evacuation scene model building unit is used for extracting characteristics of an evacuation scene according to structural information of the evacuation scene to obtain a three-dimensional model of the evacuation scene; the evacuation scene global path planning unit used for performing global path planning on the evacuation scene according to the three-dimensional model of the evacuation scene, and calculating global paths of all entrances in the evacuation scene; the crowd activity generation unit used for performing crowd grouping according to the quantity of all entrances, a relationship between individuals in the crowd and individual location information in the evacuation scene; and the crowd simulation unit used for calculating the real-time speed of individuals in a simulation process according to a social force model, and in the simulation process, correcting the moving speed of each individual in the same group in real time, so as to keep the advancing consistency of each group, and realize behavior simulation of the evacuation crowd.

The invention discloses a crowd-evacuation simulating method based on an extended BDI model. The crowd-evacuation simulating method, on the basis of the extended BDI model and with the cellular automaton and multi-agent technologies, can simulate the crowd evacuation in a public place in emergency (e.g. a fire, etc.) according to the plane layout of a building. The extended BDI structure provides a certain support for the factors in the spiritual aspect, so the method fully considers the influence of the psychological state of the people on the evacuation in the process of crowd simulation; the method is used for simulation of a crowd with incomplete environmental information; and the method can conform better to the practice. The method has an extensive application prospect in simulation studies of safe and fast evacuation of a large crown in a building and can test design proposals or responding proposals, and therefore the invention is an economical and feasible crowd-evacuation simulating method.

The invention provides a crowd evacuation simulation system based on a composite potential energy field. An improved potential energy field model is adopted for the system, a traditional potential energy field under Dirichlet boundary conditions is linearly combined with a potential energy field under Neumann boundary conditions, a local potential energy field for solving collision prevention problems between people is added into the combined potential energy field, and thus the composite potential energy field is obtained; by combining an update strategy of pedestrians and a pedestrian speed control method, the simulation system of personnel evacuation in emergency on different scales of scenes can be established according to the layout of the actual scene. In the crowd simulation process, influences of path plans of moving individuals on the evacuation are fully considered, and the potential energy field method can play a certain role in removing the influence factor. The system has extensive application prospects in research on simulation of safe and fast evacuation of a large number of personnel on difference scales of scenes, the design defects on the scenes can be found, and the crowd evacuation simulation system can assist in making execution schemes in emergency and is economically feasible.

The invention discloses a dynamic data driven modeling and simulation method for crowd evacuation and transportation behaviors in a sudden-onset disaster. According to the method, real-time disaster information is acquired by analyzing high resolution remote sensing images, a simulation model of a landslide and other natural disasters is used for outputting virtual data about disaster evolution, a crowd simulation platform describes individual behaviors on a disaster circumstance, and then real-time online modeling of the landslide disaster is achieved. A decision is made for the individual behaviors according to environment information by describing individuals through several different attributes and defining several different environment description attributes. Experiments prove that through the method, better crowd evacuation efficiency and availability are achieved in an actual rescue activity in the landslide and other natural disasters.

The invention provides a visibility judgment method of a virtual dynamic crowd, which is used for judging visibility of each individual in the virtual dynamic crowd based on crowd simulation data; wherein, the crowd simulation data comprises individuals, locations of the individuals at a certain time and IDs of scene nodes corresponding to the individuals at a certain time; the visibility judgment method comprises the following steps: visibility eliminating is carried out for a static scene, a visibility result of each frame is updated to the visibility information array of the nodes in time; the individuals in the simulation data are combined with the scene nodes corresponding to the IDs of the scene nodes, and then the visibility eliminating of each individual is realized according to the visibility of the scene nodes in the visibility information array of the nodes. The method of the invention takes an individual in the crowd as one part of the corresponding scene node of the individual, and directly attaches the scene visibility to the individual, which avoids solely judging the visibility of a great quantity of individuals, and greatly reduces consumption caused by spatial data structure update and individual shading query.

The invention relates to a character model reconstruction method, and discloses a character segmented modeling method. The method includes: building a character head model by employing a three-dimensional scanner to acquire cloud data of character head points; pre-processing the built character head model, and obtaining the pre-processed character head model; simplifying the pre-processed character head model by employing an edge collapse simplification method based on edge curvature and area error, and maintaining detail features of the character head model by employing an interactive method; and performing matching and fusion of the obtained character head model and a human torso model in a human torso database, and finally obtaining a complete human body model. According to the method, the modeling efficiency is improved, the diversity of models is enhanced, and the application value in crowd simulation is high.

One embodiment of the invention sets forth a technique for avoiding collisions between agents traversing different lanes of a multi-lane path in a crowd simulation. A perception data structure is maintained per agent that keeps track of the speeds and locations of neighboring agents in the crowd simulation. At each frame, the perception data structure corresponding to a particular agent is analyzed to identify any projected collisions. If the perception data structure indicates that the agent is projected to collide with another agent, then a collision avoidance operation is executed to avoid the collision.

The invention discloses a crowd behavior simulation method based on a grid agent. The crowd behavior simulation method comprises the following steps: step 1, grid data processing, wherein input data of the grid data processing are three-dimensional model data and a ground object topological relation, output data are grid object information, and the grid data processing comprises three sub-steps: 1, definition of a grid object; 2, searching and setting of a boundary grid; and 3, defining a topological connection relation of the grid object; step 2, crowd behavior calculation: the main function of the crowd behavior calculation is to provide the crowd behavior calculation with an individual as a unit, wherein input data are crowd behavior parameters and the grid object information obtained in step 1, output data are crowd behavior information, and the crowd behavior calculation comprises two sub-steps: 1, setting the crowd behavior parameters; and 2, calculating a crowd behavior trace; and step 3, crowd simulation visualization, wherein the main function of the step is to realize the visualization of the geographical environment and human behaviors.

The invention discloses an emergency crowd evacuation simulation method and system based on a group effect. The emergency crowd evacuation simulation method comprises the steps that (1) building a scene model and an individual model in a Unity 3D scene; Step (2), creating an interaction relationship between scenes and individuals, creating an interaction relationship between individuals, and generating an emergency scene; And step (3), realizing emergency crowd evacuation simulation based on a group effect in an emergency scene. According to the crowd evacuation simulation method, the advantages of unity3D software are utilized, a scene is more conveniently set up, modeling is conducted through multi-body dynamics crowd simulation and cohesion models based on the position, a model and a social model, and a built role animation controller is utilized, so that the crowd evacuation simulation process is smoother, more stable and more efficient.

The invention discloses a virtual reality crowd simulation method and system based on eye movement tracking. A crowd simulation database is established, which comprises a building and obstacle model database, a human voice and environment sound collection database, a real human database, a human model database and a human attribute database. The virtual reality scene is constructed by selecting building and obstacle models from the building and obstacle model database and combining with the human voice and environment sound collection database. Based on the real character database, the human model database and the human attribute database, the virtual human with high simulation degree is constructed and placed in the virtual reality scene. According to the route planning of crowd escape orevacuation in emergencies, the virtual human is simulated and the eye movement tracking data is collected in simulation. The eye movement tracking data is labeled on the surface of the building modelto determine whether the locations of safety signs in the building are reasonable or not. The invention is helpful to improve the efficiency of crowd escape and evacuation in emergencies such as disasters.

The invention discloses a user-oriented planned virtual crowd simulation framework. The user-oriented planned virtual crowd simulation framework comprises a group, triggers, a motion range limiter and a cloud control end. The group is defined and described by a staff composition module, an activity range module, an objective module and an individual behavior and group behavior module and used for simulation of group behaviors so that simulation of various complex group behaviors is achieved. The triggers comprise the path point trigger, the flow trigger, the time trigger, the speed trigger and other triggers and are used for planning crowd paths and determining path topological information and perceptual information of crowd behaviors. The activity range limiter is used for providing position information of static barriers in the environment. The cloud control end is used for planning optimal paths without congestion for crowds according to the path topological information specified by users and the objective of the group. The user-oriented planned virtual crowd simulation framework has good verisimilitude and real-time performance, high flexibility and convenient and fast operability and can simulate verisimilar virtual crowd behaviors on various scenes according to intentions of the users.

The invention provides a vision field-based crowd grouping and crowd evacuation simulation system, and the system comprises an evacuation scene model establishing unit used for extracting features ofan evacuation scene to obtain a three-dimensional model of the evacuation scene; an evacuation scene global path planning unit which is used for carrying out global path planning on the evacuation scene according to the three-dimensional model of the evacuation scene and calculating global paths of all entrances and exits in the evacuation scene; a crowd activity generation unit which is used forgrouping the crowds according to the number of all entrances, the global paths of all entrances, the vision fields of individuals in the crowds and the individual position information to obtain grouping information; a crowd simulation unit which is used for processing the grouping information by adopting a social force model added with a visual factor and dividing pedestrians into a leader type and a follower type to realize behavior simulation of evacuated crowds. According to the invention, the evacuated crowds are divided into groups to move, the evacuation efficiency is improved, the building evacuation performance can be detected, the actual evacuation process is optimized, and the evacuation efficiency is improved.

The invention provides a data-driven crowd movement simulation method based on a generative adversarial network. The method relates to the fields of crowd simulation and deep learning. The method is used on the basis of a pedestrian trajectory data set extracted from pedestrian movement video data. Virtual pedestrians which do not exist in a data set are generated in a simulation scene, and complete path planning closer to real pedestrian response is carried out on the generated virtual pedestrians according to given conditions such as initial positions and destinations and other factors in the whole scene. According to the method, a generative adversarial network (GAN) based on a long short-term memory (LSTM) is applied to train a simulation model. Compared with a traditional crowd simulation method based on rules, the simulation effect of the movement trail of the virtual pedestrian simulated through the method is more realistic, and the movement trail of the virtual pedestrian is closer to the movement situation of a real pedestrian. According to the invention, the trajectory planning task for the virtual pedestrian is completed, and the authenticity of the crowd movement simulation effect is effectively improved.

The invention discloses a hierarchical crowd simulation method and system used for a complex scene. The method comprises the steps that regional division is carried out on the complex scene, and a topological structure diagram of the complex scene is obtained; global path planning is carried out on the complex scene, and a global path which comprises a scene entrance and a scene exit and a specific interest region is calculated, wherein the scene entrance, the scene exit and the specific interest region are in the complex scene; crowd grouping information and position information of individuals, of the crowd, in the complex scene are confirmed, a potential field of the individuals is established in the complex scene, and according to the potential field of the individuals, acting force exerted on the individuals in the complex scene is calculated; real-time velocities of the individuals during the crowd simulation process are calculated, and by using a traffic flow model, the real-time velocities of the individuals during the crowd simulation process are corrected in real time, so that the crowd simulation is achieved. According to the technical scheme, a velocity control mechanism in a potential field method crowd simulation is improved, and the problem that the velocity variation is unnatural is solved, so that the velocity variation is more conformed to the crowd moving rule, and the actual requirement of the crowd simulation is met.

The invention discloses a crowd simulation evacuation method and a device based on artificial fish swarm algorithm and target detection. The method comprises the following steps: a topological map isestablished according to an evacuation scene plan, wherein a node represents an evacuation outlet, a connecting line represents two outlets are connected, and a final outlet is marked; the evacuated crowd is detected and tracked according to the video data, and the evacuated crowd individual speed is obtained; The individual speed is used to represent the feeding speed in the feeding behavior of the fish swarm, the fish swarm is initialized, the food is represented by the final exit position of the scene, and the path planning of the evacuated crowd is carried out based on the artificial fishswarm algorithm; When the number of evacuees at the final exit is equal to the total number of evacuees, the evacuation process ends and the evacuation path is obtained. The invention identifies the individuals with slower evacuation speed, guides the crowd to evacuate, thereby taking into account the vulnerable groups in the evacuation, improves the evacuation efficiency, and avoids excessive congestion.

The invention discloses a large-scale virtual crowd simulation system. The system comprises a crowd behavior model simulation module, an environment processing module and a behavior realization module; the crowd behavior model simulation module is used for creating a crowd perception model composed of regional perception and virtual sense organs, and providing technical support for people obtaining external information and conducting behavior decision-making; the environment processing module is used for conducting data processing on a static environment model and rendering navigation grids; the behavior realization module is used for processing information input of a crowd and judging output behaviors through logic. According to the large-scale virtual crowd simulation system, in application, after a virtual crowd simulation technology is applied to an emergency drilling simulation training system of subway and high-speed railway platforms, the application of the technology can be extended to the aspects such as emergency exit design and emergency plan optimization of public buildings such as high-speed railways and subways through the accumulation of application experience and in-depth research, and the system has huge social and economic benefits.

The invention discloses a large-scale crowd behavior evolution analysis method based on map-reduce and multi-agent models. The method is achieved through a map-reduce programming model, and simulation operation is achieved through data processing operation; by converting agent information in multi-agent crowd simulation into entries of a key value and attribute value list and by achieving logical judgment decision and status update of agents through map and reduce operations, large-scale crowd simulation is converted into data processing operation in a key value and attribute value list mode. The method is achieved on clusters based on an ApacheHadoop frame. It is proved through tests that the simulation crowds produced through the simulation method are in accordance with existing serialization simulation results in subsection and morphology; according to the method, the scale of the crowds capable of being simulated is large, the operation speed is high, and occupation of a memory working set is greatly reduced.

The invention discloses a virtual reality crowd simulation method and a system based on the avoidance of uncanny valley effect. The virtual reality crowd simulation method and the system based on theavoidance of uncanny valley effect comprises avoiding the 'uncanny valley' effect of vision and the 'uncanny valley' effect of tactile sensation. In the aspect of avoid the 'uncanny valley' effect ofvision, the virtual character library module is added, and in the aspect of avoid 'uncanny valley' effect of tactile sensation, the human voice and the environment sound acquisition library module andthe collision visual fuzzy module are added, the virtual reality crowd simulation method and the system based on the avoidance of uncanny valley effect can effectively reduce and avoid the discomfortof the visual and tactile 'uncanny valley' effect to the human and the loss of the immersion sense, so that the crowd can simulate more natural performance, and the person is more in-person and has higher authenticity.

One embodiment of the invention sets forth a technique for identifying and avoiding impending collisions between moving objects in an animation. Paths traversed by the moving objects intersect at pre-determined intersection points. As a moving object approaches an intersection point, a collision avoidance module determines whether the object is on course to collide with another moving object also approaching the intersection point. If a collision is detected, then the collision avoidance module modifies the speed of the moving object to avoid the collision.

The invention discloses a digital simulator for AMS risk assessment. The digital simulator comprises a crowd simulation module, a measurement interference simulation module and an AMS risk assessment module. The crowd simulation module is used for creating simulation crowds with different hypoxia tolerance capacities by applying a mathematical model of a human hypoxia stress system and a plurality of instance experiment data. the measurement interference simulation module is used for simulating measurement interference and deep respiration physiological interference in the SpO2 information, and superposing simulated interference signals onto the SpO2 generation curve of the simulated crowd; and the AMS risk assessment module obtains indexes used for assessing AMS risks, and the indexes are designed based on parameters of the identified model and used for representing the real hypoxia adaptability of different virtual crowds. According to the AMS risk assessment method, richer research data can be obtained for AMS risk assessment algorithm research, and effective and reliable AMS risk assessment is realized.

The invention provides a crowd evacuation simulation method in a dynamic environment based on deep reinforcement learning, and the method comprises the steps: using a crowd as a multi-agent system, employing a deep neural network to approach a mapping function from a state to an action for a single pedestrian agent to serve as a behavior controller of the agent; and by the pedestrian agent, makinga behavior decision by observing the dynamic environment state and utilizing the mapping function, and taking corresponding actions from the action space. In the evacuation simulation process of a discrete time sequence, one strategy refers to probability distribution of agent actions on the time sequence in a given state. The information feedback of the environment to the intelligent agent is reflected as a return value. The target of the intelligent agent is to maximize the expectation of the cumulative return value, that is, an optimal action value function is searched, and an optimal strategy is obtained. According to the method, crowd evacuation simulation which is difficult to realize in classical crowd simulation in a high-dynamic environment can be realized, the simulation effectis closer to the real situation, and the randomness of crowd behaviors is enhanced.

Pedestrian movement and the mutual relation between pedestrians and the environment are studied through crowd simulation, and a method is used for pre-event prediction or post-event inspection for urban infrastructure planning, large public activity planning, emergency evacuation of emergencies and the like. According to the method, common behaviors such as pedestrian avoidance, congestion, congestion, attraction and repulsion in real life can be simulated by constructing mechanical models between pedestrians and between pedestrians and the environment, and the movement mode of pedestrians indifferent scenes can be simulated by combining the particularity of pedestrians in different scenes. According to the method, the inter-group avoidance behavior and the coordination behavior of the sub-group in the group are added into the model; the avoidance behaviors in different environments can be simulated, and if the sub-groups are forced to be separated in the avoidance process, the different rows of sub-groups can be kept in central symmetry by adding the force that the rear row of sub-groups get close to the front row of sub-groups, and a certain distance beneficial to walking is kept. On one hand, the current crowd simulation is widely applied to the fields of travel, fire protection, buildings and the like, and the quality of a crowd simulation result meets the requirements ofreality and rapid visualization; and on the other hand, the model better conforms to real pedestrian behaviors, and special phenomena under different attribute pedestrians and scenes can be simulatedby adjusting parameters, so that the model has very high social application value.

The invention relates to the computer graphics and crowd simulation technology, in particular to a crowd simulation method for a social force model based on relative speed. The method comprises the following steps: carrying out simulation calculation on a simulation experiment by using an improved relative speed social force model; obtaining simulation data including the speed, the speed direction and the position of the pedestrian; performing simulation modeling according to the simulation data, and displaying a simulation result. Compared with an existing relative speed-based social force model formula, the improved relative speed-based social force model has the advantage that a calculation formula of the improved relative speed-based social force model adds a calculation item of the social psychological force of the pedestrian at the current position. By improving the relative speed of the social force model, the pedestrian oscillation problem can be effectively reduced, and the pedestrian overlapping problem can be avoided, so that the social force model is more authentic in crowd evacuation simulation application.

The invention relates to the field of crowd simulation, reinforcement learning and statistics, and especially relates to a crowd battle simulation method based on emotional infection and deep reinforcement learning. The method comprises the following steps: extracting people number information of all agents; randomly endowing each agent with ID information, an initial position, an initial emotionvalue, an initial reward value and an initial action; obtaining an initial average physiological influence through initial action calculation; inputting the people number information, the ID information, the initial positions, the initial emotion values, the initial reward values, the initial actions and the initial average physiological influence into a neural network to obtain the action of theagent at the moment t; and sequentially carrying out iterative training for a plurality of times. Therefore, when a violent event occurs suddenly, strategy guidance can be quickly provided for the right side by the method, the reverse side is subdued as soon as possible, the situation that the scene is out of control is avoided, and high earnings are obtained, so social safety is maintained. Meanwhile, a simulation model also has great reference and learning values for real battle exercises.

The invention provides a crowd simulation evacuation method and system based on a wolf pack algorithm and congestion detection, and the method comprises the steps: obtaining real-time real scene data,and obtaining a real-time scene graph and crowd parameter information through real-time video data in the real scene data; setting evacuation scene parameter information according to the real-time scene graph and the crowd parameter information, creating an evacuation scene model and a figure model, importing the figure model into the evacuation scene model to establish an evacuation scene modelwith figures, and taking the crowd parameter information as preset evacuation crowd parameter information to perform wolf pack initialization; simulating crowd evacuation by using a wolf pack algorithm, calculating a wolf pack target function value according to the exit congestion degree of the evacuation scene model with the person, and selecting an optimal exit according to the target function value function; and when the evacuation number of the final exit is equal to the total number, ending the evacuation process. According to the invention, the evacuation exit with too high congestion degree can be effectively identified, and the crowd is guided to evacuate, so that more efficient evacuation is realized, and excessive congestion is avoided.

The invention relates to a target line-based advancing target arrival judgment method in crowd simulation, which expands a crowd advancing target from a target point to a target line / target line segment, and takes arrival near the target line or crossing the target line as a judgment basis for target arrival. The algorithm can effectively alleviate congestion and an 'enclosure' effect near a target point caused by taking the 'point' as a target and a large number of people simulated by a computer, so that the computer simulation of crowd behaviors is more realistic.