44 results about "Crowd evacuation" patented technology

The invention relates to an unsteady-state crowd evacuation path planning method and device based on deep reinforcement learning. The method comprises the following steps of: simplifying an unsteady evacuation environment into a two-dimensional region by using a discrete grid method; modeling is performed on dangerous source dynamic diffusion, the fire source position and range are updated in real time, and dynamic fire development is simulated. A person is regarded as an intelligent agent and moves according to a decision rule guided by a potential energy field. And with maximization of evacuated people number and minimization of evacuation time as objective functions, constraint conditions are determined in combination with actual shopping malls and agent movement rules, and an unsteady crowd evacuation path planning model is constructed. An APF-MADDPG algorithm is adopted, path planning is converted into a Markov decision process, state features are designed and combined with a potential energy field to construct a deep reinforcement learning framework, and after a model is trained, unsteady crowd evacuation path planning is achieved. By adopting the method, the practicability and the real-time performance of crowd evacuation path planning can be improved.

The invention discloses a large-scale people flow evacuation simulation method considering congestion state real-time decision, and belongs to the technical field of crowd emergency evacuation, and the method comprises the steps: S1, obtaining to-be-evacuated crowds, protection places and road network data in a region; s2, constructing a road passing time calculation model; s3, constructing a protection place full-load early warning model; s4, constructing a shortest time flow model considering a congestion state real-time decision; s5, obtaining a crowd evacuation movement process by utilizing crowd simulation, and performing visual presentation; the method is high in simulation efficiency and good in precision, and has good accuracy and practicability for crowd evacuation simulation.

The invention is suitable for the technical field of crowd simulation and emergency evacuation, and provides an emotional infection and social force model coupling-based crowd evacuation method, which comprises the following steps of: constructing an emotional infection model based on an SEIRS framework, dividing individual emotional states into a susceptible state, an exposed state, an infected state and a recovery state, the infection rate and the recovery rate are dynamically adjusted through a feedback function; establishing a social force model, and embedding an emotion value into calculation of a target driving force and a repulsive force between individuals to realize real-time adjustment of the motion behavior by the emotion; global path planning and an ant colony pheromone mechanism are fused, a density perception cost function and pheromone guiding force are introduced, and dynamic blockage avoidance is achieved; a bidirectional feedback mechanism is designed, and local path reconstruction is triggered when congestion is detected. According to the method, bidirectional coupling of emotion propagation and behavior decision is realized, the authenticity and adaptability of evacuation simulation are improved, and a scientific basis can be provided for emergency management and evacuation strategy optimization in public places.

The invention discloses a tunnel fire intelligent evacuation guiding robot system based on multi-modal sensing, which belongs to the technical field of tunnel fire fighting equipment and comprises a multi-source sensing fusion unit, a hierarchical communication decision unit, a dynamic path planning unit, a multi-modal guiding unit and an intelligent inspection unit. The multi-source sensing fusion unit constructs a tunnel three-dimensional point cloud map with centimeter-level precision according to acquired data and performs dynamic environment updating in real time, and the hierarchical communication decision unit adopts dual-mode communication to ensure the stability and reliability of signal transmission under different working conditions; the dynamic path planning unit generates and optimizes a safe evacuation path in real time, and finally the multi-mode guiding unit performs guiding in three modes of visual sense, auditory sense and human body posture tracking so as to guarantee crowd evacuation. According to the robot system, intelligent and efficient evacuation guidance of a tunnel fire scene can be achieved, the efficiency and safety of rescue actions are improved, and unnecessary confusion and danger are avoided.

The invention discloses a crowd evacuation simulation method and system based on panic emotion propagation and hazard source influence, and the method comprises the steps: setting a simulation scene, and modifying the infection rate of infectious disease dynamics into a state transition function affected by an emotion threshold value; after simulation is started, calculating a state transfer function for each individual in the crowd, and establishing a panic emotion propagation model; setting danger repulsive force exclusive to a danger source, establishing a social force model based on the panic emotion propagation model and the danger repulsive force, and simulating crowd evacuation based on the social force model. According to the method, the infection rate of infectious disease dynamics is changed into a dynamic parameter influenced by an emotion threshold value from a static parameter, the dynamic parameter is used as a state transition function of individuals, the function is used as a bridge, a social force model and panic emotion propagation are fused together, individual differences in crowd evacuation are deeply reflected, and the social force model and the panic emotion propagation are integrated. Therefore, simulation of crowd evacuation is realized.

According to the intelligent real-time generation method for the building crowd evacuation emergency route diagram, firstly, building space personnel self-adaptive evacuation space-time trajectory modeling is carried out through a building emergency evacuation model space-time trajectory modeling method, then, the evacuation trajectory is calculated and analyzed through a clustering method based on space-time similarity and structural similarity, and the emergency route diagram is generated in real time. And clustering evacuation tracks according to a method of fusing the evacuation tracks and the evacuation tracks, designing and organizing an emergency route diagram after fine tuning based on the clustered tracks, constructing an evacuation simulation streamline diagram, and calculating real tracks of personnel in the system. The designed evacuation simulation streamline diagrams reflect the real tracks of the personnel in the simulation system. Therefore, the management and control of the evacuated people stream are guided, the evacuated people stream is effectively controlled and induced, the crowded people stream during emergencies is avoided and relieved, the maximization of the evacuation efficiency is realized, and the economic loss and casualties are effectively reduced.

A macroscopic simulation method and device for crowd evacuation considering the reduction of the pedestrian passing speed in crowded areas, which relate to the field of crowd evacuation simulation in buildings, solve the problems of how to judge the location where crowd congestion occurs and quantitatively calculate the reduction of the pedestrian passing speed at the congested location during the evacuation simulation process. The present invention provides the following solutions: the macroscopic simulation method includes: obtaining the building floor plan of the case to be analyzed, and converting the spatial information of the evacuation scene into an evacuation path geometric network model in the building; inputting the data of the evacuation path geometric network model in the building into the calculation program module of the improved macroscopic simulation method, that is, performing an evacuation process simulation analysis on the case to be analyzed, and calculating the reduction of the pedestrian passing speed at the congested location to obtain the evacuation process simulation result. It is also applicable to the global evacuation situation of large-scale crowd evacuation scenarios.

The invention discloses a global security monitoring method and system based on the Internet of Things, and belongs to the technical field of the Internet of Things, and the method comprises the steps: obtaining an electronic map of a target place, obtaining the number and position data of Internet of Things nodes of each partition, and generating the configuration data of an edge calculation terminal; the edge computing terminal identifies an abnormal condition and sends a patrol instruction to the security terminals in the partitions; the cloud fusion platform constructs a stampede risk prediction model to predict a stampede risk level, and if the stampede risk level exceeds an early warning level, an evacuation request is sent to the management terminal; and generating a crowd evacuation path plan based on the multi-channel spatial-temporal feature matrix and the trampling risk level, and sending a guide instruction to a security terminal to generate a visual interface. According to the global safety monitoring method and system based on the Internet of Things provided by the invention, full-coverage, real-time and refined risk prediction, early warning and management of crowded places are realized, and the crowd safety management efficiency and accuracy are effectively improved.

The invention discloses a large-scale subway group escape system based on panic emotion infection, and relates to the technical field of escape systems.The large-scale subway group escape system based on panic emotion infection comprises the steps that 1, a crowd evacuation scene is constructed according to subway related information; step 2, initializing a crowd (Crowd); each individual in the crowd Crowd comprises personality characteristics phi, an initial emotion value e0 and an initial movement speed v0; 3, performing an emotion infection simulation algorithm; step 4: first; the range between the pedestrian carrying the luggage and the luggage is divided to be less than or equal to 0.5 m; carrying out related weighting and improvement on the evacuation speed and behavior of pedestrians according to the luggage volume and the luggage weight when the luggage volume and the luggage weight exceed the individual bearing capacity; a discarding behavior will occur; 5, calculating the crowd density rho of the walking area at the t + 1 moment according to the crowd movement information; the crowd evacuation behavior is updated; and step 6, realizing crowd evacuation visualization in a graphical mode.

The application discloses a large-scale crowd evacuation simulation method considering congestion state real-time decision, and belongs to the technical field of crowd emergency evacuation, and comprises the following steps: S1, acquiring crowd to be evacuated, a shelter and road network data in a region; S2, constructing a road passing time calculation model; S3, constructing a full shelter early warning model; S4, constructing a shortest time flow model considering congestion state real-time decision; and S5, obtaining a crowd evacuation movement process by using crowd simulation and performing visual presentation; the application has high simulation efficiency, good precision, good accuracy and practicability for crowd evacuation simulation.

The invention discloses a heterogeneous crowd evacuation model fusing multi-dimensional panic emotion characteristics in the technical field of crowd evacuation dynamics, which comprises a multi-dimensional panic emotion analysis unit and an improved social force model unit, the multi-dimensional panic emotion analysis unit is used for quantifying panic emotion influences of three dimensions of environment pressure, individual characteristics and group effect and generating dynamic multi-dimensional emotion pressure factors, and the improved social force model unit is used for adjusting expected speeds of heterogeneous individuals based on the multi-dimensional emotion pressure factors and realizing dynamic simulation of crowd evacuation behaviors. According to the dynamic adjustment mechanism, an evacuation path and a group cooperation strategy are dynamically optimized based on a real-time panic pressure factor, a multi-dimensional panic emotion analysis method and an improved social force model fusing multi-dimensional panic emotion features are innovatively designed, and the influence mechanism of different factors on the panic emotion and evacuation efficiency of pedestrians can be effectively analyzed. Therefore, real simulation of pedestrian evacuation behaviors is realized.

The invention is suitable for the field of computer vision, provides a crowd evacuation path dynamic recommendation method and system based on visual data, and discloses a crowd evacuation path dynamic recommendation method and system based on visual data in the field of computers. Crowd motion video streams are collected in real time, pedestrian head orientation vectors are extracted after preprocessing, a crowd motion velocity field is formed, and a crowd motion state database is established; constructing a visual inertia dynamic model to predict future crowd evacuation path distribution change; the real-time bearing rate of each channel is calculated, and grading early warning is generated by combining capacity threshold pre-judging saturation time; and when the channel reaches the optimal capacity threshold value, determining an upstream branch node, calculating a diversion starting time window and strength, generating a control instruction, and transmitting the control instruction to an indication execution end. Accurate dynamic path recommendation is realized, and evacuation safety and efficiency are improved.

This invention belongs to the field of computer simulation of crowd evacuation, and proposes a crowd congestion prediction method and system based on a spatiotemporal fusion neural network. The method includes: acquiring scene information and crowd state information of crowd movement; constructing a knowledge graph based on the scene information and crowd state information, and preprocessing the knowledge graph; inputting the preprocessed knowledge graph into a graph neural network to obtain a spatial feature matrix; inputting the preprocessed knowledge graph, spatial feature matrix, and crowd state information into the spatiotemporal fusion neural network to predict the movement speed of crowds within a region, and predicting the crowd congestion situation within the region based on the crowd movement speed. The knowledge graph-based construction method of this invention can more comprehensively store heterogeneous information related to crowd movement. Furthermore, using a knowledge graph to represent scene information and store crowd movement state attributes provides relatively complete input information for the next step of the neural network to mine the spatiotemporal correlation of crowd movement.

The present invention discloses a real-time crowd evacuation simulation method based on computer vision, wherein the method specifically comprises the following steps: collecting human images and creating a human data set; using the human data set to train a YOLOv3 human detection model; obtaining a camera RTSP video stream, inputting the RTSP video stream into the YOLOv3 human detection model to identify humans and obtain pixel coordinates; performing three-dimensional spatial positioning using monocular machine vision to convert the human pixel coordinates into world coordinates; extracting features of the evacuation scene, establishing a two-dimensional cellular automaton evacuation model of the evacuation scene; placing humans into a grid of the cellular automaton evacuation model using world coordinates, and performing real-time evacuation simulation, thereby achieving simulation of the crowd evacuation process. The present invention can perform real-time analysis based on the actual distribution of personnel and the disaster situation of the emergency scene, and generate an evacuation strategy in real time to help managers perform evacuation management, thereby achieving scientific guidance on emergency safety evacuation plans.

The application discloses an emergency shelter intelligent management and control system and method based on multi-source information fusion, and relates to the technical field of geographic data processing.The application carries out data fusion on real-time scene data, establishes a dynamic scene monitoring model according to the data fusion result, establishes a historical disaster evolution model according to historical case data, compares the historical disaster evolution model with the dynamic scene monitoring model, selects a comparison case according to the comparison result, simulates the disaster trend in the dynamic scene monitoring model according to the comparison case, covers the simulation disaster trend result on the dynamic scene monitoring model, simulates crowd evacuation with the emergency shelter as a simulation end point on the dynamic scene monitoring model, generates a personnel evacuation route according to the crowd evacuation simulation result, and dynamically adjusts according to real-time scene data during the execution of the personnel evacuation route.

The application is suitable for the field of computer vision, and provides a crowd evacuation path dynamic recommendation method and system based on visual data. The application discloses a crowd evacuation path dynamic recommendation method and system based on visual data in the field of computers. Real-time crowd motion video streams are collected, and pedestrian head orientation vectors are extracted after preprocessing to form a crowd motion velocity field and establish a crowd motion state database. A visual inertia dynamics model is constructed to predict future crowd evacuation path distribution changes. Real-time carrying rates of each channel are calculated, and a saturation time is generated by combining a capacity threshold value to generate a graded early warning. When the channel reaches the optimal capacity threshold value, an upstream bifurcation node is determined, a diversion start time window and intensity are calculated, and a control instruction is generated and transmitted to an indication execution end. The application realizes accurate dynamic path recommendation and improves evacuation safety and efficiency.

The invention discloses a pedestrian evacuation simulation method and system based on crowding and treading effects, and the method comprises the steps: adding a pedestrian crowding force generation and transmission superposition process, and adding an individual pushing and falling judgment mechanism at the same time, thereby carrying out the judgment of individual pushing and falling based on the resultant force generated by transmission superposition, and achieving the simulation of pedestrian evacuation. Further, pedestrian evacuation is simulated more truly, and more real evacuation time and actual tumble space of tumble pedestrians are obtained; thus, the problem that a traditional pedestrian dynamics microscopic simulation model cannot reproduce real crowd crowding force generation and propagation is solved from the technical level, the linkage process of high-density crowd crowding disaster can be completely reproduced, and then the reasonability and accuracy of crowd evacuation simulation data are improved; therefore, the method is very suitable for large-scale application and popularization.

This disclosure belongs to the field of crowd evacuation simulation technology, and provides a data-driven crowd group movement simulation method and system, including the following steps: acquiring crowd video data; dividing the acquired crowd video data into crowd groups; extracting crowd movement attributes based on the divided crowd groups; quantifying the extracted crowd movement attributes into group attributes to obtain group attribute quantification results; and obtaining a simulation animation of crowd group movement based on the obtained group attribute quantification results and a crowd simulation platform to realize the simulation of crowd group movement.

The application is suitable for the field of crowd evacuation simulation technology, and provides a subway fire scene robot-assisted pedestrian flow evacuation simulation method, which is applied to a subway fire scene, and the method comprises the following steps: obtaining a first position of a fire in a subway station and second positions of all pedestrians in a dangerous area of the fire; determining, according to the first position and the second positions, third positions of guide robots for evacuating and shunting the pedestrians in the dangerous area, and controlling the guide robots to move to the third positions; inputting the first position, the second positions and the third positions into an improved social force model for processing to obtain accelerations of each pedestrian in the dangerous area; determining, according to the accelerations of the pedestrians, positions of the pedestrians at a next moment, and guiding the pedestrians to move to the positions; and detecting whether there is an unevacuated person in the dangerous area, if there is, continuing to evacuate the unevacuated person, otherwise ending the simulation. The application has high intelligence and automation levels, and can quickly and effectively evacuate pedestrians.

The application relates to the field of building evacuation path optimization, and discloses a building evacuation path optimization method and system based on shortest global evacuation time, which can solve the problem that the path optimization method in the prior art cannot truly and accurately reflect the influence of crowd congestion on the evacuation process and cannot reasonably consider the actual situation of crowd evacuation under the congestion state in the path optimization process. The method comprises the following steps: generating a corresponding evacuation road network model from a building plan of a case to be analyzed; determining an evacuation personnel distribution scheme; determining a shortest path evacuation scheme, calculating an evacuation simulation result of the scheme, and obtaining a global evacuation time and a group number of the last completed evacuation; determining an evacuable group set according to a two-way mobilization strategy; setting an initial optimization scheme; iteratively optimizing the evacuation paths of groups in a first optimization adjustment area and a second optimization adjustment area. Finally, an optimized evacuation path scheme and corresponding evacuation process simulation calculation results are output.

The application relates to a non-steady crowd evacuation path planning method and device based on deep reinforcement learning. The method comprises the following steps: simplifying a non-steady evacuation environment into a two-dimensional region by using a discrete grid method. A dangerous source dynamic diffusion model is established, and the fire source position and range are updated in real time according to the model, so that the dynamic development of the fire is simulated. People are regarded as agents, and move according to the decision rule of the potential field guidance. A non-steady crowd evacuation path planning model is established by taking the maximum number of evacuees and the minimum evacuation time as the objective function, combining the actual situation of a shopping mall and the movement rule of the agents, and determining the constraint conditions. An APF-MADDPG algorithm is used to convert the path planning into a Markov decision process, a state feature is designed, a deep reinforcement learning framework is established by combining the potential field, and the model is used to realize non-steady crowd evacuation path planning after being trained. The method can improve the practicability and real-time performance of crowd evacuation path planning.

The invention discloses an artificial intelligence space people flow evacuation path planning method. The method comprises the steps of determining a space region, collecting space people flow data, establishing a path planning model, outputting an optimal evacuation path scheme and performing safety supervision. The invention belongs to the technical field of path planning, and particularly relates to an artificial intelligence space people flow evacuation path planning method. According to the scheme, a path traffic capacity model based on crowd density is adopted, and correction of the crowd density on traffic capacity is considered; the multi-target path fusion control model comprehensively balances safety, time and capacity factors to obtain a path comprehensive score, then the optimal evacuation direction trend is determined through dynamic path gradient field calculation, and a decision threshold is obtained through a fuzzy decision control equation to adapt to complex conditions; and a data processing technology and a visualization technology are adopted to organize optimal path scheme data of the evacuation planning model, and evaluation indexes are calculated and visually displayed to provide people evacuation navigation guidance.

The invention relates to the technical field of emergency evacuation prediction, and discloses a crowd emergency evacuation prediction method based on a multi-agent system, and the method comprises the steps: carrying out the fusion of multi-source positioning data after time alignment and space registration through employing a weighted time sequence fusion algorithm, and obtaining an environment state vector; taking to-be-evacuated persons as individual evacuation agents, generating individual behavior parameters based on historical individual information, and performing emergency evacuation response modeling on the individual evacuation agents to obtain perception delay of the individual evacuation agents; and predicting a moving track of the individual evacuation intelligent body, predicting a congestion risk and a fire risk of each building area, and carrying out dynamic re-planning evacuation path real-time correction processing on the individual evacuation intelligent body. According to the invention, on the basis of multi-source positioning perception, a multi-agent cooperative crowd evacuation mode is constructed, individual evacuation tracks and regional risks are dynamically predicted, and accurate prediction and evaluation of the crowd emergency evacuation process are realized through real-time path re-planning.

The invention discloses a crowd evacuation modeling method combining a stepping model and a neural network model. The crowd evacuation modeling method comprises the following steps of scene initialization and neural network training; the pedestrian motion simulation comprises pedestrian activation, pedestrian motion direction and speed calculation and pedestrian motion simulation, and pedestrian step length is calculated in combination with a pedestrian panic emotion calculation module; activating pedestrians; calculating the moving direction and speed of the pedestrian; pedestrian motion simulation: updating the pedestrian position according to the pedestrian motion direction and speed obtained in the above steps; and after single pedestrian motion simulation is completed, motion simulation of a next pedestrian is repeated until pedestrian motion simulation of all pedestrians in the model scene is completed. According to the crowd evacuation modeling method combining the stepping model and the neural network model, different models are adopted for simulation in different simulation scenes, the precision and efficiency of pedestrian evacuation simulation are improved, and pedestrian movement, flow distribution of a confluence platform and interaction among pedestrians can be accurately processed in a high-density environment.

The invention discloses an emergency shelter intelligent management and control system and method based on multi-source information fusion, and relates to the technical field of geographic data processing. The method comprises the following steps: performing data fusion on real-time scene data, establishing a dynamic scene monitoring model according to a data fusion result, establishing a historical disaster evolution model according to historical case data, comparing the historical disaster evolution model with the dynamic scene monitoring model, and selecting a comparison case according to a comparison result. Performing disaster trend simulation in the dynamic scene monitoring model according to the comparison case, covering the dynamic scene monitoring model with a disaster trend simulation result, performing crowd evacuation simulation on the dynamic scene monitoring model by taking the emergency shelter as a simulation terminal point, and generating a personnel evacuation route according to a crowd evacuation simulation result, and in the execution process of the personnel evacuation route, dynamic adjustment is performed according to real-time scene data.

The invention relates to the technical field of crowd evacuation guiding, in particular to a crowd dynamic evacuation guiding method based on an emotional infection model, which comprises the following steps: acquiring multi-source sensing data such as a crowd density thermodynamic diagram, panic scream frequency distribution and real-time passage counting of gates at entrances and exits of a target place; after carrying out space-time alignment on the crowd density thermodynamic diagram and panic scream frequency distribution, inputting an emotional infection propagation model to generate a dynamic emotional field distribution diagram; calculating an instantaneous evacuation potential value of a candidate evacuation channel in combination with real-time passage counting, and screening a priority evacuation channel set; and extracting channel surrounding environment constraint parameters to generate a path weight matrix with physical capacity limitation, superposing emotional excitement indexes to remove conflict paths, and determining a final guide path set. According to the method, more accurate and more efficient crowd dynamic evacuation guidance can be realized.

Systems, methods, and media for smart city crowd evacuation based on an Internet of Things (IoT) large model are provided. The method is executed by an emergency supervision management platform. The method includes: acquiring regional monitoring data based on an emergency supervision sensing control platform through a monitoring device deployed in a monitoring region; determining a regional congestion risk based on the regional monitoring data; in response to the regional congestion risk satisfying an evacuation condition: determining an emergency evacuation parameter based on the regional congestion risk; transmitting an emergency control signal to an emergency evacuation device deployed in the monitoring region via an emergency supervision sensing network platform based on the emergency evacuation parameter, to control an opening interval of a channel gate based on a traffic control parameter, and to control a display content of an electronic display board based on a traffic guidance parameter.

The application discloses a kind of crowd evacuation simulation method and system based on emotional infection dynamic evolution model, which comprises the following steps: constructing emotional infection network based on strong relationship and weak relationship in physical space and information space respectively;Determine the emotional infection rules based on strong relationship and weak relationship in information-physical space;Based on emotional infection network and emotional infection rules, combined with mean field theory, construct emotional infection dynamic evolution model based on strong relationship and weak relationship in information-physical space;Solve the emotional infection dynamic evolution model by using finite difference method, and obtain the proportion of different state individuals in crowd evacuation process at any time.The present application is based on the emotional infection rules of strong relationship and weak relationship between individuals in information-physical space, and constructs emotional infection dynamic evolution model, which can truly reflect the dynamic changes of individual social relationship in the evacuation process, provide guidance for crowd evacuation, and assist in crowd evacuation.

The application is suitable for the field of crowd simulation and emergency evacuation, and provides a crowd evacuation method based on emotion contagion and social force model coupling, which comprises the following steps: constructing an emotion contagion model based on SEIRS framework, dividing individual emotion state into susceptible state, exposed state, infected state and recovered state, and dynamically adjusting infection rate and recovery rate through feedback function; establishing a social force model, embedding emotion value into target driving force and individual repulsive force calculation, realizing real-time adjustment of emotion on movement behavior; fusing global path planning and ant colony pheromone mechanism, introducing density perception cost function and pheromone guiding force, realizing dynamic congestion avoidance; designing a bidirectional feedback mechanism, triggering local path reconstruction when congestion is detected. The application realizes the bidirectional coupling of emotion transmission and behavior decision, improves the reality and adaptability of evacuation simulation, and can provide a scientific basis for public place emergency management and evacuation strategy optimization.

The application discloses a kind of crowd evacuation flow prediction methods based on multi-level causal counterfactual memory enhancement, comprising: obtaining flow monitoring data and corresponding geographic spatial position data, establish the memory bank consisting of semantic, plot, procedure memory;Space-time data set is constructed based on flow monitoring data and geographic spatial position data;Procedure memory: for storing several kinds of intervention scripts, the intervention script is node flow limiting strategy;If it is judged that the current flow monitoring data is abnormal congestion state and is historical regularity congestion snapshot data, then extract the causal weight, time window segment and intervention script matched with the current from the memory bank, and inject it into the prompt word of large language model;The prompt word is input into large language model, and crowd flow is predicted.The application significantly improves the crowd flow prediction accuracy in large-scale activity complex environment.