An improved emergency simulation method for pedestrian blocks

Abstract

The present invention provides an improved pedestrian block emergency simulation method, comprising: preprocessing the road network vectors of the pedestrian block to generate an undirected graph; classifying the individual characteristics of people corresponding to the pedestrian block based on a preset theory; determining the neural response strength of different individuals based on the distance between people in the pedestrian block and the classified individual characteristics; constructing an intelligent individual speed function based on the neural response strength and the undirected graph; defining panic ratios for the individual characteristics of different classifications, and constructing corresponding path evacuation models based on the intelligent individual speed function; constructing the path evacuation models corresponding to the different individual characteristics, performing a shortest path search and obtaining a shortest time state sequence, and conducting an emergency evacuation simulation in a virtual scene constructed in the pedestrian block based on the search results and the obtained results. This improves the efficiency of emergency evacuation.

Description

Technical Field

[0001] The present invention relates to the field of intelligent simulation technology, and in particular to an improved pedestrian block emergency simulation method. Background Art

[0002] my country's rapid urbanization has led to a surge in pedestrian street construction across cities. These areas, integrating commercial, leisure, and tourism functions, have become one of the most popular public venues in cities, often becoming the most densely populated areas during holidays. While dense populations bring vitality to pedestrian streets, they also pose greater safety risks. When responding to emergencies, safe, rapid, and scientific evacuation strategies, along with appropriate spatial arrangements to support evacuation, are crucial factors in mitigating disaster losses.

[0003] During the evacuation of pedestrian streets, people usually evacuate manually, plan evacuation routes, and go to shelters for evacuation. However, this type of evacuation does not take into account the emotional state of the crowd in the face of an emergency, resulting in low evacuation efficiency.

[0004] Therefore, the present invention proposes an improved pedestrian block emergency simulation method. Summary of the Invention

[0005] The present invention provides an improved pedestrian block emergency simulation method to solve the above-mentioned technical problems.

[0006] The present invention proposes an improved pedestrian block emergency simulation method, comprising:

[0007] Step 1: Preprocess the pedestrian street network vector data to generate an undirected graph;

[0008] Step 2: Based on the preset theory, classify the personality characteristics of the people corresponding to the pedestrian blocks;

[0009] Step 3: Determine the neural response strength of different individuals based on the distance between people in the walking block and the classified personality characteristics;

[0010] Step 4: Construct an intelligent individual speed function based on the neural response intensity and the undirected graph;

[0011] Step 5: Define panic ratios for different categories of personality characteristics, and construct corresponding path evacuation models based on the intelligent individual speed function;

[0012] Step 6: When the personality characteristics are related to rational individual characteristics, a corresponding path evacuation model is obtained and a shortest path search is performed. Meanwhile, when the personality characteristics are related to irrational individual characteristics, a corresponding path evacuation model and an improved Viterbi algorithm are obtained to obtain the shortest time state sequence.

[0013] Step 7: Based on the search results and the obtained results, an emergency evacuation simulation is performed in the virtual scene constructed in the pedestrian street.

[0014] Preferably, step 1: performing data preprocessing on the road network vector of the pedestrian block to generate an undirected graph includes:

[0015] Obtaining the road layout of the pedestrian street;

[0016] Preliminary analysis of the road layout is performed based on road connectivity to determine a set of intersections and a set of roads in the pedestrian block;

[0017] Determining an intersection capacity attribute and an initial evacuation number attribute for each intersection in the intersection set;

[0018] Determining a road access attribute and a travel time attribute for each road in the road set;

[0019] Based on the adjacency table, first values related to the intersection capacity attribute and the initial evacuation number attribute are obtained respectively, and second values related to the road channel attribute and the travel time attribute are also obtained respectively;

[0020] Vector data is expressed based on the first value and the second value to generate an undirected graph of the pedestrian block.

[0021] Preferably, step 2: classifying the personality characteristics of the people corresponding to the pedestrian street based on a preset theory includes:

[0022] Based on a preset theory, determining the evacuation capability of corresponding people in the pedestrian street in response to evacuation signals during an emergency evacuation transfer process;

[0023] The individual characteristics of the corresponding human body are classified based on the evacuation capability.

[0024] Preferably, step 3: determining the neural response strength of different individuals based on the distance between people in the pedestrian block and the classified personality characteristics, includes:

[0025] Determine a first distance between a human body that does not receive an evacuation signal and a human body that sends an evacuation signal, and simultaneously determine a second distance between the human body that does not receive an evacuation signal and a danger source;

[0026] Determining individual characteristics of the non-receiving human body and thereby determining the receiving capacity;

[0027] Determining the emotional contagion characteristics of the sending person, and then determining the sending capability;

[0028] Based on the first distance, the second distance, the receiving capability, and the sending capability, the nerve response intensity of the corresponding human body is determined.

[0029] Preferably, step 4: constructing an intelligent individual speed function based on the neural response intensity and the undirected graph, including:

[0030] Determining the road area of the pedestrian block and the number of people on each road based on the undirected graph;

[0031] Determine crowd density based on road area and the number of people on the road;

[0032] Obtain the intensity of each person's neural response during road evacuation;

[0033] Based on the least squares method, the crowd density and neural response intensity are fitted to obtain the intelligent individual speed function corresponding to the human body.

[0034] Preferably, step 5: defining panic ratios for personality characteristics of different categories, and constructing corresponding path evacuation models based on the intelligent individual speed function, including:

[0035] Determining a first number of irrational people in the pedestrian block based on different categorized personality characteristics, and simultaneously determining a second number of rational people in the pedestrian block;

[0036] obtaining a panic ratio based on a ratio of the first quantity to the first quantity and the second quantity;

[0037] Based on the panic ratio and in combination with the intelligent individual speed function, a corresponding path evacuation model is constructed.

[0038] Preferably, when the personality characteristics are related to rational individual characteristics, obtaining a corresponding path evacuation model and performing a shortest path search include:

[0039] Determine the undirected graph, and add a super source node to the undirected graph;

[0040] Optimizing the first preset algorithm, obtaining a second preset algorithm, determining the shortest distance from the super source node to the target node in the undirected graph based on the second preset algorithm, and recording the shortest distance;

[0041] Based on the path evacuation model of rational individual characteristics, the number of people evacuated and the available capacity of the roads and intersections corresponding to the target nodes in the undirected graph are updated in real time to determine whether there are any unevacuated people;

[0042] If so, re-determine the shortest distance from the super source node to the target node based on the second preset algorithm and record it;

[0043] Based on all recorded results, complete the search for the shortest path;

[0044] If not present, confirm that the evacuation is complete.

[0045] Preferably, when the personality characteristics are related to irrational individual characteristics, obtaining a corresponding path evacuation model and an improved Viterbi algorithm to obtain the shortest time state sequence includes:

[0046] Determining the adjacency between intersection nodes and roads in the undirected graph, and constructing a state space of the intersection nodes;

[0047] Based on the state space, calculating the velocity model of each intersection node;

[0048] Based on the velocity model, calculating hidden Markov transition probability;

[0049] Based on the hidden Markov transition probability, constructing a state probability transition matrix corresponding to the irrational individuals in the undirected graph;

[0050] Based on the state probability transfer matrix, the Viterbi algorithm is improved and combined with the path evacuation model of irrational individual characteristics to obtain the shortest time state sequence.

[0051] Preferably, the process of expressing vector data based on the first value and the second value to generate an undirected graph of the pedestrian block further includes:

[0052] Determine, based on historical detection information of the pedestrian block by the drone, sites to be constructed and sites currently under construction in the pedestrian block;

[0053] Acquire a current construction configuration image of the site to be constructed, perform a first preliminary analysis on the current construction configuration image, determine an existing construction configuration matrix, and calculate an eigenvector of the construction configuration matrix;

[0054] Based on the vector matching model, the characteristic vector is associated with each row vector in the construction matrix to determine whether there is a construction association. If so, the corresponding row vector is marked as a non-reference vector; otherwise, the corresponding row vector is marked as a reference vector;

[0055] Obtain all calibrated reference vectors, analyze and determine whether there is evacuation possibility, and if so, first determine the maximum number of evacuees allowed in the same row at the minimum evacuation passage of the construction site, and based on the first determination result, perform a first open calibration on the corresponding construction site.

[0056] Otherwise, the first closure calibration will be carried out on the site to be constructed;

[0057] Acquiring a construction progress image of the construction site, and performing a second preliminary analysis on the construction progress image to determine the safety and evacuability of the corresponding construction site;

[0058] Obtaining the safety and evacuation satisfaction standards, and when the satisfaction standard indicates that evacuation is not recommended, performing a second closure calibration on the corresponding construction site;

[0059] Otherwise, based on the satisfaction criteria, the maximum number of people allowed to evacuate in the same row corresponding to the minimum evacuation passageway of the construction site is determined, and a second open calibration is performed;

[0060] Determine a first feasible layout of the site to be constructed based on the pedestrian block based on the first open demarcation and the first closed demarcation; and simultaneously determine a second feasible layout of the site under construction based on the pedestrian block based on the second open demarcation and the second closed demarcation;

[0061] A first revised expression of the first feasible layout and a second revised expression of the second feasible layout are established, and combined with the vector data expression to generate an undirected graph of the pedestrian block.

[0062] Preferably, based on the search results and the obtained results, after performing the emergency evacuation simulation in the virtual scene constructed in the pedestrian street, the method further includes:

[0063] Based on the search results and the obtained results, determining the complete evacuation time of each passable road in the virtual scene constructed by the pedestrian block, and obtaining the total duration of the simulated evacuation based on the virtual scene;

[0064] Based on the total simulated evacuation duration and several complete evacuation times, m1 evacuation routes are screened, and based on the historical population density of each road in the pedestrian block, the route with the highest population density is selected from the m1 evacuation routes as the evaluation route;

[0065] Determining the total number of sub-roads included in the evaluation route and determining the number of simulated populations for each sub-road during the simulation process;

[0066] Set the current simulated population of each sub-road to all irrational individual characteristics, and calculate the sub-evacuation time of each sub-road;

[0067]

[0068] Where T1 represents the sub-evacuation time of the corresponding sub-road; S1 represents the effective evacuation length of the corresponding sub-road; v j1 represents the irrational evacuation speed of the j1th current simulated population; J1 represents the total number of the current simulated population of the corresponding sub-road; γ1 represents the evacuation factor caused by the population evacuated from other roads when the corresponding sub-road reaches the intersection of the next sub-road, and its value range is (0, 0.2);

[0069] Based on all the obtained sub-evacuation times, a total evacuation time of the evaluation route is obtained;

[0070] Comparing the total evacuation time with the total simulated evacuation duration T;

[0071]

[0072] Wherein, Y1 is the comparison result. When Y1=1, a first reminder solution is obtained from the reminder database, and a first installation density of the reminder device is determined;

[0073] When Y1=0.5, a second reminder solution is obtained from the reminder database, and a second installation density of the reminder device is determined;

[0074] Otherwise, when Y1=0, there is no need to set a new reminder device.

[0075] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purposes and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0076] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0078] Figure 1 Flowchart of an improved pedestrian block emergency simulation method according to an embodiment of the present invention;

[0079] Figure 2 This is a flowchart of the shortest path search in an embodiment of the present invention;

[0080] Figure 3 This is a flowchart for obtaining the shortest time state sequence in an embodiment of the present invention. DETAILED DESCRIPTION

[0081] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0082] Example 1:

[0083] The present invention proposes an improved pedestrian block emergency simulation method, such as Figure 1 As shown, including:

[0084] Step 1: Preprocess the pedestrian street network vector data to generate an undirected graph;

[0085] Step 2: Based on the preset theory, classify the personality characteristics of the people corresponding to the pedestrian blocks;

[0086] Step 3: Determine the neural response strength of different individuals based on the distance between people in the walking block and the classified personality characteristics;

[0087] Step 4: Construct an intelligent individual speed function based on the neural response intensity and the undirected graph;

[0088] Step 5: Define panic ratios for different categories of personality characteristics, and construct corresponding path evacuation models based on the intelligent individual speed function;

[0089] Step 6: When the personality characteristics are related to rational individual characteristics, a corresponding path evacuation model is obtained and a shortest path search is performed. Meanwhile, when the personality characteristics are related to irrational individual characteristics, a corresponding path evacuation model and an improved Viterbi algorithm are obtained to obtain the shortest time state sequence.

[0090] Step 7: Based on the search results and the obtained results, an emergency evacuation simulation is performed in the virtual scene constructed in the pedestrian street.

[0091] In this example, the least squares method is used to fit the agent's actual evacuation velocity function based on the state of pedestrians in a real evacuation scenario. For irrational agents, a hidden Markov chain cellular automaton evacuation model based on the velocity function is proposed. This model solves the shortest time state sequence based on an improved Viterbi algorithm. For rational agents, a shortest path search algorithm based on the CCASR algorithm is proposed.

[0092] The beneficial effects of the above technical solution are: by classifying personality characteristics, determining the intensity of neural reactions, and then constructing a speed function; at the same time, by combining it with the panic ratio, constructing an evacuation model, and by analyzing different personality characteristics, improving the efficiency of emergency evacuation.

[0093] Example 2:

[0094] Based on Example 1, step 1: preprocessing the road network vector of the pedestrian block to generate an undirected graph includes:

[0095] Obtaining the road layout of the pedestrian street;

[0096] Preliminary analysis of the road layout is performed based on road connectivity to determine a set of intersections and a set of roads in the pedestrian block;

[0097] Determining an intersection capacity attribute and an initial evacuation number attribute for each intersection in the intersection set;

[0098] Determining a road access attribute and a travel time attribute for each road in the road set;

[0099] Based on the adjacency table, first values related to the intersection capacity attribute and the initial evacuation number attribute are obtained respectively, and second values related to the road channel attribute and the travel time attribute are also obtained respectively;

[0100] Vector data is expressed based on the first value and the second value to generate an undirected graph of the pedestrian block.

[0101] The QGIS open source library is used to analyze road connectivity, the adjusted attribute table is exported, the values are stored using the adjacency table, and the vector data is expressed using an undirected graph.

[0102] In the constructed undirected traffic network graph, the edges and vertices represent the roads and intersections in the evacuation block, respectively. Graph = (intersection, road) is used to represent the road conditions in the evacuation block, where intersection is a set of intersections and road is a set of roads. Each intersection has two attributes: intersection capacity and initial evacuation number; each road has two attributes: road capacity and travel time.

[0103] The beneficial effect of the above technical solution is that by expressing the undirected graph as vector data, it is easy to construct the undirected graph, which makes it easy to ensure the accuracy of the simulation and the efficiency of the subsequent evacuation.

[0104] Example 3:

[0105] Based on Example 1, step 2: classifying the personality characteristics of the people corresponding to the pedestrian street based on a preset theory, including:

[0106] Based on a preset theory, determining the evacuation capability of corresponding people in the pedestrian street in response to evacuation signals during an emergency evacuation transfer process;

[0107] The individual characteristics of the corresponding human body are classified based on the evacuation capability.

[0108] In this embodiment, the preset theory refers to the OCEAN theory, that is, the personality theory.

[0109] In this embodiment, it is considered that during the emergency evacuation process, individuals with different personality characteristics have different evacuation capabilities for evacuation signals. In order to better quantify the impact of evacuation signals in emergency evacuation scenarios, based on the principle of signal propagation, assuming that all individuals have the same ability and method to send evacuation signals, the signal receiving capabilities of different types of individuals are considered. Only the sensitivity of the receiver is classified in order to more realistically restore the impact of this factor on the evacuation results. Based on the OCEAN personality theory, the individual's acceptance ability is divided into three categories, and the parameter settings are shown in Table 1.

[0110] The influence of personality traits on the acceptance of evacuation signals

[0111]

[0112] The beneficial effect of the above technical solution is that it is convenient to classify individual characteristics by determining the evacuation capacity.

[0113] Example 4:

[0114] Based on Example 1, step 3: determining the neural response strength of different individuals based on the distance between people in the pedestrian block and the classified personality characteristics, including:

[0115] Determine a first distance between a human body that does not receive an evacuation signal and a human body that sends an evacuation signal, and simultaneously determine a second distance between the human body that does not receive an evacuation signal and a danger source;

[0116] Determining individual characteristics of the non-receiving human body and thereby determining receiving capacity;

[0117] Determining the emotional contagion characteristics of the sending person, and then determining the sending capability;

[0118] Based on the first distance, the second distance, the receiving capability, and the sending capability, the nerve response intensity of the corresponding human body is determined.

[0119] Drawing on the principles of infectious disease models, the Euclidean distance between individuals who haven't received the evacuation signal and all senders is calculated. If the distance is less than a set threshold, the non-receiver is deemed to have received the evacuation signal and has made the decision to evacuate. This distance and the receiving capabilities of different types of individuals are then substituted into the evacuation signal propagation model to update the individual's evacuation signal reception strength.

[0120] The calculation formula for evacuation signal propagation is:

[0121]

[0122] P i (t) is the neural response intensity of individual i at time t;

[0123]

[0124] and

[0125]

[0126] In the above, G(i, j) represents the influence of individual personality type on signal reception; D(i, d) represents the influence of distance on signal reception; Pi(t) is the neural response intensity of individual i at time t, δ i is the receiving ability of agent i. Individuals with different personalities have different receiving abilities. j is the sending ability of intelligent individual j, intelligent individual j is the individual with infectious emotion among the intelligent agents, m is the number of individuals, d ij represents the Euclidean distance between agents i and j, d max represents the emotional contagion radius, ||r i -r d || represents the distance between agent i and danger source d.

[0127] The beneficial effect of the above technical solution is that by determining the intensity of the neural response, it is convenient to carry out effective simulation of the subsequent emergency response.

[0128] Example 5:

[0129] Based on Example 1, step 4: constructing an intelligent individual speed function based on the neural response intensity and the undirected graph, including:

[0130] Determining the road area of the pedestrian block and the number of people on each road based on the undirected graph;

[0131] Determine crowd density based on road area and the number of people on the corresponding road;

[0132] Obtain the intensity of each person's neural response during road evacuation;

[0133] Based on the least squares method, the crowd density and neural response intensity are fitted to obtain the intelligent individual speed function corresponding to the human body.

[0134] In this embodiment, the individual speed depends on the road crowd density and the individual's neural response strength. i (t) is used to calculate the evacuation speed with the crowd density. The crowd density ρ is related to the road area. A function of the number of people n on the road:

[0135]

[0136] Since the crowd density and the neural response strength of the agent change in real time during the evacuation process, the function of the actual movement speed of agent i is obtained by fitting the road crowd density and the neural response strength of agent i using the least squares method:

[0137] v i (t) = β0 + β1 × ρ + β2 × p i (t)

[0138] v i (t) is the actual movement speed of agent i at time t; β0, β1, β2 are the estimated values of model coefficients; ρ is the crowd density; p i (t) is the neural response intensity of the agent.

[0139] The beneficial effect of the above technical solution is that by determining the speed function of the intelligent individual, it is convenient to carry out simulation effectively.

[0140] Example 6:

[0141] Based on Example 1, step 5: defining panic ratios for different categories of personality characteristics, and constructing corresponding path evacuation models based on the intelligent individual speed function, including:

[0142] Determining a first number of irrational people in the pedestrian block based on different categorized personality characteristics, and simultaneously determining a second number of rational people in the pedestrian block;

[0143] obtaining a panic ratio based on a ratio of the first quantity to the first quantity and the second quantity;

[0144] Based on the panic ratio and in combination with the intelligent individual speed function, a corresponding path evacuation model is constructed.

[0145] In this embodiment, pedestrians in an evacuation situation can be divided into two categories: rational pedestrians and irrational pedestrians. Rational pedestrians are emotionally stable, familiar with the street conditions and evacuation exits, and have independent judgment and decision-making regarding their evacuation routes. Irrational pedestrians are emotionally influenced by their surroundings, are unfamiliar with their environment, and lack rational judgment of their surroundings. They desperately want to escape but lack sufficient information, so they blindly choose the path that provides the fastest evacuation.

[0146] The beneficial effect of the above technical solution is that, by combining the panic ratio with the speed function, it is convenient to effectively construct an evacuation model and facilitate faster evacuation.

[0147] Example 7:

[0148] Based on Example 1, Figure 2As shown, when the personality characteristics are related to rational individual characteristics, the corresponding path evacuation model is obtained and the shortest path search is performed, including:

[0149] Determine the undirected graph, and add a super source node to the undirected graph;

[0150] Optimizing the first preset algorithm, obtaining a second preset algorithm, determining the shortest distance from the super source node to the target node in the undirected graph based on the second preset algorithm, and recording the shortest distance;

[0151] Based on the path evacuation model of rational individual characteristics, the number of people evacuated and the available capacity of the roads and intersections corresponding to the target nodes in the undirected graph are updated in real time to determine whether there are any unevacuated people;

[0152] If so, re-determine the shortest distance from the super source node to the target node based on the second preset algorithm and record it;

[0153] Based on all recorded results, complete the search for the shortest path;

[0154] If not present, confirm that the evacuation is complete.

[0155] In this embodiment, an undirected graph of a traffic network in which the initial capacity of each intersection and road has been processed in advance is first imported; a super source node is added to the undirected graph to connect each source node, and the travel time of the edge connecting the super source node and the source node is 0 and the capacity is infinite.

[0156] Since the first preset algorithm has not yet taken into account the road crowd density at different locations during the evacuation process, a new heuristic function is created instead:

[0157]

[0158] Where: Ln is the length of the nth road, V in is the movement speed of agent i on the nth road (the speed on the nth road).

[0159] According to the first-in, first-out evacuation principle, the number of people on roads and intersections and the remaining capacity are updated until all evacuees are safely evacuated and the shortest evacuation route is obtained to form a plan.

[0160] F(n)=g(n)+h(n);h(n)←i(n)

[0161] F(n)=g(n)+i(n)

[0162] Where: F(n) is the minimum path cost from the source node to the destination node; g(n) is the cost of the path from node n to the source node; h(n) is the heuristic function that estimates the minimum path cost from node n to the destination node; i(n) represents the updated minimum path cost from node n to the destination node.

[0163] The beneficial effect of the above technical solution is that it provides an effective basis for subsequent evacuation simulation by determining the search for the shortest path.

[0164] Example 8:

[0165] Based on Example 1, Figure 3 As shown, when the personality characteristics are related to irrational individual characteristics, the corresponding path evacuation model and the improved Viterbi algorithm are obtained to obtain the shortest time state sequence, including:

[0166] Determining the adjacency between intersection nodes and roads in the undirected graph, and constructing a state space of the intersection nodes;

[0167] Based on the state space, calculating the velocity model of each intersection node;

[0168] Based on the velocity model, calculating hidden Markov transition probability;

[0169] Based on the hidden Markov transition probability, constructing a state probability transition matrix corresponding to the irrational individuals in the undirected graph;

[0170] Based on the state probability transfer matrix, the Viterbi algorithm is improved and combined with the path evacuation model of irrational individual characteristics to obtain the shortest time state sequence.

[0171] In this embodiment, a speed-based hidden Markov chain model is proposed for irrational individuals, and an improved Viterbi algorithm is established to solve the shortest time state sequence. The evacuation problem in an undirected road network graph is modeled and analyzed, and a hidden Markov chain evacuation model and an improved Viterbi algorithm based on this model are proposed. Road intersections in an undirected road network graph correspond to nodes in the undirected graph, and each road corresponds to an edge in the undirected graph. In this hidden Markov chain model, the spatial position state transition probability of pedestrians is calculated. The hidden state space corresponds to the set of edges in the undirected graph, and the observable state is the crowd density on each road. It is assumed that pedestrians are initially located at the road intersection, and the initial state transition probability facing each road is equal. During this transition process, the transition rule is to transfer the pedestrian's spatial position state based on the Markov transition probability of each road. Each pedestrian's state corresponds to the edge on which the pedestrian is located. The improved Viterbi algorithm is used to solve the hidden state sequence of the shortest time path.

[0172] In this embodiment, assuming that the number of channels adjacent to the current node is n and its state space is {1, 2, ..., n}, the individual state transition probability matrix is:

[0173] M(k)=[p1,p2,……,p n ]

[0174] Among them: p1~pn are the probabilities of the individual going to the 1st to nth roads adjacent to the current node intersection.

[0175] Let M(k) be the decision function of the speed model of the road network. The number of adjacent road channels at the current node intersection is n, and the corresponding crowd density has n possible values, which are recorded as:

[0176] ρ=[ρ1,ρ2,……,ρ n ]

[0177] The calculation formula for the transition probability of each road is:

[0178]

[0179] Where: i represents the crowd density of any road, ρ i represents the sum of the crowd density of all roads. The formula means that pedestrians are more inclined to move to roads with lower crowd density.

[0180] The beneficial effect of the above technical solution is that by determining the state sequence, subsequent evacuation simulation is facilitated and performed effectively.

[0181] Example 9:

[0182] Based on Example 2, the process of generating an undirected graph of a pedestrian block by performing vector data expression based on the first value and the second value further includes:

[0183] Determine, based on historical detection information of the pedestrian block by the drone, sites to be constructed and sites currently under construction in the pedestrian block;

[0184] Acquire a current construction configuration image of the site to be constructed, perform a first preliminary analysis on the current construction configuration image, determine an existing construction configuration matrix, and calculate an eigenvector of the construction configuration matrix;

[0185] Based on the vector matching model, the characteristic vector is associated with each row vector in the construction matrix to determine whether there is a construction association. If so, the corresponding row vector is marked as a non-reference vector; otherwise, the corresponding row vector is marked as a reference vector;

[0186] Obtain all calibrated reference vectors, analyze and determine whether there is evacuation possibility, and if so, first determine the maximum number of evacuees allowed in the same row at the minimum evacuation passage of the construction site, and based on the first determination result, perform a first open calibration on the corresponding construction site.

[0187] Otherwise, the first closure calibration will be carried out on the site to be constructed;

[0188] Acquiring a construction progress image of the construction site, and performing a second preliminary analysis on the construction progress image to determine the safety and evacuability of the corresponding construction site;

[0189] Obtaining the safety and evacuation satisfaction standards, and when the satisfaction standard indicates that evacuation is not recommended, performing a second closure calibration on the corresponding construction site;

[0190] Otherwise, based on the satisfaction criteria, the maximum number of people allowed to evacuate in the same row corresponding to the minimum evacuation passageway of the construction site is determined, and a second open calibration is performed;

[0191] Determine a first feasible layout of the site to be constructed based on the pedestrian block based on the first open demarcation and the first closed demarcation; and simultaneously determine a second feasible layout of the site under construction based on the pedestrian block based on the second open demarcation and the second closed demarcation;

[0192] A first revised expression of the first feasible layout and a second revised expression of the second feasible layout are established, and combined with the vector data expression to generate an undirected graph of the pedestrian block.

[0193] In this embodiment, the historical detection information can be the construction information collected by the drone on the pedestrian street several times a day. Therefore, the construction sites to be constructed and the construction sites in progress can be effectively determined. Because, in the process of a series of disasters, the construction sites need to be taken into account during the evacuation of the pedestrian street to avoid the inefficiency of the evacuation simulation due to invalid colleagues or a small number of colleagues.

[0194] In this embodiment, the current construction configuration image refers to the image of the construction site, and the first preset analysis is performed to obtain the currently configured construction materials, such as cement, wood boards, etc. The construction materials may or may not be present, and may also include the setting of fences for the construction site, as well as the construction properties of the construction site, such as ground damage, building maintenance, etc. These parameters are used to construct a construction configuration matrix, and each type of parameter is set in one row, and the elements in the matrix are converted according to the conversion standard, in order to effectively analyze the construction configuration matrix.

[0195] In this embodiment, the vector matching model is obtained by training based on vector samples of several different combinations, and by determining the association match between the feature vector and each corresponding row vector, it is determined whether there is a construction association, that is, whether there is a possibility of construction. If so, the corresponding row vector is regarded as non-referenceable, that is, the row vector is not referenced. This is to ensure that the accuracy of judging whether evacuation is possible is reduced due to the existence of construction association.

[0196] In this embodiment, the correlation between the corresponding type parameter and evacuation is determined by calibrating the reference vector, and then the evacuation possibility is determined. The more the calibrated reference vector is correlated with the evacuation, the higher the corresponding correlation is, and the greater the corresponding evacuation possibility is.

[0197] In this embodiment, the maximum number of people allowed to evacuate in the same row refers to the maximum number of people allowed to be evacuated at one time at the evacuation exit of the construction site.

[0198] In this embodiment, through open and closed marking, it is convenient to effectively identify the evacuable places.

[0199] In this embodiment, the standards are met, for example, the safety reaches 90%, and the corresponding evacuation reaches 100%. At this time, based on the satisfaction, the evacuation number of the minimum evacuation exit is determined, and the satisfaction standard is also set for the minimum evacuation channel. Therefore, since the evacuation exit may be blocked during construction or waiting for construction, the evacuation number of the minimum evacuation channel is determined.

[0200] In this embodiment, the first feasible layout and the second feasible layout are determined comprehensively based on their own position positioning and calibration results. By determining the construction sites that can and cannot be carried out together, the reliability of the undirected graph can be further guaranteed, and the efficiency of evacuation can be guaranteed.

[0201] In this embodiment, modifying the expression refers to modifying the data expression under normal circumstances where construction is in progress.

[0202] The beneficial effects of the above technical solution are: by analyzing the construction situation, the inefficiency of evacuation simulation caused by invalid peers or a low number of peers is avoided. By analyzing the construction and ongoing construction, the corrected expression is effectively obtained and expressed with vector data to obtain an undirected graph, ensuring its accuracy and providing a basis for evacuation simulation.

[0203] Example 10:

[0204] Based on the second embodiment, after performing the emergency evacuation simulation in the virtual scene constructed in the pedestrian street based on the search results and the obtained results, the method further includes:

[0205] Based on the search results and the obtained results, determining the complete evacuation time of each passable road in the virtual scene constructed by the pedestrian block, and obtaining the total duration of the simulated evacuation based on the virtual scene;

[0206] Based on the total simulated evacuation duration and several complete evacuation times, m1 evacuation routes are screened, and based on the historical population density of each road in the pedestrian block, the route with the highest population density is selected from the m1 evacuation routes as the evaluation route;

[0207] Determining the total number of sub-roads included in the evaluation route and determining the number of simulated populations for each sub-road during the simulation process;

[0208] Set the current simulated population of each sub-road to all irrational individual characteristics, and calculate the sub-evacuation time of each sub-road;

[0209]

[0210] Where T1 represents the sub-evacuation time of the corresponding sub-road; S1 represents the effective evacuation length of the corresponding sub-road; v j1 represents the irrational evacuation speed of the j1th current simulated population; J1 represents the total number of the current simulated population of the corresponding sub-road; γ1 represents the evacuation factor caused by the population evacuated from other roads when the corresponding sub-road reaches the intersection of the next sub-road, and its value range is (0, 0.2);

[0211] Based on all the obtained sub-evacuation times, a total evacuation time of the evaluation route is obtained;

[0212] Comparing the total evacuation time with the total simulated evacuation duration T;

[0213]

[0214] Wherein, Y1 is the comparison result. When Y1=1, a first reminder solution is obtained from the reminder database, and a first installation density of the reminder device is determined;

[0215] When Y1=0.5, a second reminder solution is obtained from the reminder database, and a second installation density of the reminder device is determined;

[0216] Otherwise, when Y1=0, there is no need to set a new reminder device.

[0217] In this embodiment, the installation density is such as 30 meters, and a reminder device is set.

[0218] In this embodiment, the reminder database includes reminder schemes and corresponding comparison values.

[0219] The beneficial effect of the above technical solution is: by screening the evaluation route based on the evacuation simulation, and then calculating the corresponding sub-evacuation time when all sub-roads in the evaluation route have irrational individual characteristics, and then obtaining the total evacuation time, by comparing with the simulated evacuation time, effectively determining the installation density of reminder devices on different roads in the pedestrian street, thereby effectively ensuring the effective evacuation of irrational individual characteristics during the evacuation process, and further improving the evacuation efficiency.

[0220] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. An improved pedestrian block emergency simulation method, characterized in that: include: Step 1: Preprocess the pedestrian street network vector data to generate an undirected graph; Step 2: Based on the preset theory, classify the personality characteristics of the people corresponding to the pedestrian blocks; Step 3: Determine the neural response strength of different individuals based on the distance between people in the walking block and the classified personality characteristics, including: Determine a first distance between a human body that does not receive an evacuation signal and a human body that sends an evacuation signal, and simultaneously determine a second distance between the human body that does not receive an evacuation signal and a danger source; Determining individual characteristics of the non-receiving human body and thereby determining receiving capacity; Determining the emotional contagion characteristics of the sending person, and then determining the sending capability; Determining a corresponding human body's nerve response intensity based on the first distance, the second distance, the receiving capability, and the sending capability; Step 4: Based on the neural response intensity and the undirected graph, construct the speed function of the intelligent individual, including: Determining the road area of the pedestrian block and the number of people on each road based on the undirected graph; Determine crowd density based on road area and the number of people on the corresponding road; Obtain the intensity of each person's neural response during road evacuation; Based on the least squares method, the crowd density and the neural response intensity are fitted to obtain the intelligent individual speed function corresponding to the human body; Step 5: Define panic ratios for different categories of personality characteristics, and construct corresponding path evacuation models based on the intelligent individual speed function; Step 6: When the personality characteristics are related to rational individual characteristics, a corresponding path evacuation model is obtained and a shortest path search is performed. Meanwhile, when the personality characteristics are related to irrational individual characteristics, a corresponding path evacuation model and an improved Viterbi algorithm are obtained to obtain the shortest time state sequence. Step 7: Based on the search results and the obtained results, an emergency evacuation simulation is performed in the virtual scene constructed in the pedestrian street.

2. The improved pedestrian block emergency simulation method according to claim 1, characterized in that: Step 1: Preprocess the pedestrian street network vector data to generate an undirected graph, including: Obtaining the road layout of the pedestrian street; Preliminary analysis of the road layout is performed based on road connectivity to determine a set of intersections and a set of roads in the pedestrian block; Determining an intersection capacity attribute and an initial evacuation number attribute for each intersection in the intersection set; Determining a road access attribute and a travel time attribute for each road in the road set; Based on the adjacency table, first values related to the intersection capacity attribute and the initial evacuation number attribute are obtained respectively, and second values related to the road channel attribute and the travel time attribute are also obtained respectively; Vector data is expressed based on the first value and the second value to generate an undirected graph of the pedestrian block.

3. The improved pedestrian block emergency simulation method according to claim 1, characterized in that: Step 2: Based on the pre-set theory, classify the personality characteristics of the people corresponding to the pedestrian street, including: Based on a preset theory, determining the evacuation capability of corresponding people in the pedestrian street in response to evacuation signals during an emergency evacuation transfer process; The individual characteristics of the corresponding human body are classified based on the evacuation capability.

4. The improved pedestrian block emergency simulation method according to claim 1, characterized in that: Step 5: Define panic ratios for different categories of personality traits, and construct corresponding path evacuation models based on the intelligent individual speed function, including: Determining a first number of irrational people in the pedestrian block based on different categorized personality characteristics, and simultaneously determining a second number of rational people in the pedestrian block; obtaining a panic ratio based on a ratio of the first quantity to the first quantity and the second quantity; Based on the panic ratio and in combination with the intelligent individual speed function, a corresponding path evacuation model is constructed.

5. The improved pedestrian block emergency simulation method according to claim 1, characterized in that: When the personality characteristics are related to rational individual characteristics, a corresponding path evacuation model is obtained and a shortest path search is performed, including: Determine the undirected graph, and add a super source node to the undirected graph; Optimizing the first preset algorithm, obtaining a second preset algorithm, determining the shortest distance from the super source node to the target node in the undirected graph based on the second preset algorithm, and recording the shortest distance; Based on the path evacuation model of rational individual characteristics, the number of people evacuated and the available capacity of the roads and intersections corresponding to the target nodes in the undirected graph are updated in real time to determine whether there are any unevacuated people; If so, re-determine the shortest distance from the super source node to the target node based on the second preset algorithm and record it; Based on all recorded results, complete the search for the shortest path; If not present, confirm that the evacuation is complete.

6. The improved pedestrian block emergency simulation method according to claim 1, characterized in that: When the personality characteristics are related to irrational individual characteristics, a corresponding path evacuation model and an improved Viterbi algorithm are obtained to obtain the shortest time state sequence, including: Determining the adjacency between intersection nodes and roads in the undirected graph, and constructing a state space of the intersection nodes; Based on the state space, calculating the velocity model of each intersection node; Based on the velocity model, calculating hidden Markov transition probability; Based on the hidden Markov transition probability, constructing a state probability transition matrix corresponding to the irrational individuals in the undirected graph; Based on the state probability transfer matrix, the Viterbi algorithm is improved and combined with the path evacuation model of irrational individual characteristics to obtain the shortest time state sequence.

7. The improved pedestrian block emergency simulation method according to claim 2, characterized in that: The process of expressing vector data based on the first value and the second value to generate an undirected graph of the pedestrian block further includes: Determine, based on historical detection information of the pedestrian block by the drone, sites to be constructed and sites currently under construction in the pedestrian block; Acquire a current construction configuration image of the site to be constructed, perform a first preliminary analysis on the current construction configuration image, determine an existing construction configuration matrix, and calculate an eigenvector of the construction configuration matrix; Based on the vector matching model, the characteristic vector is associated with each row vector in the construction matrix to determine whether there is a construction association. If so, the corresponding row vector is marked as a non-reference vector; otherwise, the corresponding row vector is marked as a reference vector; Obtain all calibrated reference vectors, analyze and determine whether there is evacuation possibility, and if so, first determine the maximum number of evacuees allowed in the same row at the minimum evacuation passage of the construction site, and based on the first determination result, perform a first open calibration on the corresponding construction site. Otherwise, the first closure calibration will be carried out on the site to be constructed; Acquiring a construction progress image of the construction site, and performing a second preliminary analysis on the construction progress image to determine the safety and evacuability of the corresponding construction site; Obtaining the safety and evacuation satisfaction standards, and when the satisfaction standard indicates that evacuation is not recommended, performing a second closure calibration on the corresponding construction site; Otherwise, based on the satisfaction criteria, the maximum number of people allowed to evacuate in the same row corresponding to the minimum evacuation passageway of the construction site is determined, and a second open calibration is performed; Determine a first feasible layout of the site to be constructed based on the pedestrian block based on the first open demarcation and the first closed demarcation; and simultaneously determine a second feasible layout of the site under construction based on the pedestrian block based on the second open demarcation and the second closed demarcation; A first revised expression of the first feasible layout and a second revised expression of the second feasible layout are established, and combined with the vector data expression to generate an undirected graph of the pedestrian block.

8. The improved pedestrian block emergency simulation method according to claim 1, characterized in that: Based on the search results and the obtained results, after performing an emergency evacuation simulation in the virtual scene constructed in the pedestrian street, the method further includes: Based on the search results and the obtained results, determining the complete evacuation time of each passable road in the virtual scene constructed by the pedestrian block, and obtaining the total duration of the simulated evacuation based on the virtual scene; Based on the total simulated evacuation duration and several complete evacuation times, m1 evacuation routes are screened, and based on the historical population density of each road in the pedestrian block, the route with the highest population density is selected from the m1 evacuation routes as the evaluation route; Determining the total number of sub-roads included in the evaluation route and determining the number of simulated populations for each sub-road during the simulation process; Set the current simulated population of each sub-road to all irrational individual characteristics, and calculate the sub-evacuation time of each sub-road; Where T1 represents the sub-evacuation time of the corresponding sub-road; S1 represents the effective evacuation length of the corresponding sub-road; v j1 represents the irrational evacuation speed of the j1th current simulated population; J1 represents the total number of the current simulated population of the corresponding sub-road; γ1 represents the evacuation factor caused by the population evacuated from other roads when the corresponding sub-road reaches the intersection of the next sub-road, and its value range is (0, 0.2); Based on all the obtained sub-evacuation times, a total evacuation time of the evaluation route is obtained; Comparing the total evacuation time with the total simulated evacuation duration T; Wherein, Y1 is the comparison result. When Y1=1, a first reminder solution is obtained from the reminder database, and a first installation density of the reminder device is determined; When Y1=0.5, a second reminder solution is obtained from the reminder database, and a second installation density of the reminder device is determined; Otherwise, when Y1=0, there is no need to set a new reminder device.

Images