Emergency evacuation path planning method based on HBA optimization ant colony algorithm

Abstract

The invention discloses an emergency evacuation path planning method based on an HBA optimization ant colony algorithm. The method comprises the following steps: firstly, on the basis of a constructed indoor space two-dimensional map, based on a badger optimization algorithm, initializing the number and position of a population, updating a density factor, and obtaining an initial optimal path trajectory xnew; thirdly, optimizing initial pheromones of the ant colony algorithm by using the obtained initial path trajectory xnew; introducing distance estimation and a smoothing function to improve transition probability selection of the ant colony algorithm; and finally, pheromones of the ant colony algorithm are updated, and an optimal path Z output is formed after iteration. The badger optimization algorithm is introduced into the ant colony algorithm, the fast search ability of the ant colony algorithm is further improved, the problem of falling into local optimum is avoided, an improved heuristic function is provided by combining a smooth function and distance estimation, the convergence speed of the ant colony algorithm is further improved, and the algorithm is more efficient. The method can adapt to path planning in a complex scene, and the evacuation efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of emergency management, and particularly relates to an emergency evacuation path planning method based on the HBA optimized ant colony algorithm. Background Technique

[0002] With the rapid development of social economy, the number of complex buildings such as various transportation hubs and large complexes is increasing continuously, such as: subway environment, airport terminal buildings, large shopping malls, etc. However, once an emergency occurs, such as: fire, criminal attack, etc., it will cause great harm to the safety of personnel. Therefore, more and more researchers are conducting research on reasonable evacuation planning paths under emergencies. At present, software such as Pathfinder is generally used to carry out emergency evacuation simulation. Although it can simulate the evacuation scenario to a certain extent, however, due to the complexity of the actual evacuation environment, the uncontrollability of people, and in addition, the evacuation models constructed by these software are relatively simple and cannot fit the actual evacuation scenario, their use is limited in the specific evacuation process and further research is needed.

[0003] In recent years, the emergence of intelligent optimization algorithms has provided new ideas for solving the problem of indoor emergency evacuation path planning under emergencies, including: particle swarm algorithm, ant colony algorithm, sparrow search algorithm, etc. Among many optimization algorithms, the ant colony algorithm has been widely used due to its strong global search ability and parallel search characteristics. However, in the actual emergency evacuation process, the traditional ant colony algorithm has problems such as low search efficiency, many inflection points, and long path finding. Summary of the Invention

[0004] Aiming at the above existing problems, the present invention proposes an emergency evacuation path planning method based on the HBA optimized ant colony algorithm. First, aiming at the problem that the initial pheromone distribution in the ant colony algorithm is unreasonable and it is easy to fall into local optimum, the present invention introduces the honey badger optimization algorithm into the ant colony algorithm. The initial pheromone concentration required by the ant colony algorithm is obtained through the fast optimization ability of the honey badger optimization algorithm, thereby avoiding the problem of the ant colony algorithm falling into local optimum. Secondly, on the basis of the traditional ant colony algorithm, the present invention uses a Manhattan-based distance estimation method to guide the ant colony algorithm to search in the shortest target direction. Finally, on the basis of the above method, a smoothing function is designed to minimize the number of turns during the personnel evacuation process and improve the stability and safety of the personnel evacuation route planning. Compared with the traditional emergency evacuation path planning method based on the ant colony algorithm, the method of the present invention can improve the efficiency and quality of the planning, thereby helping personnel to complete the evacuation process quickly.

[0005] The above object is achieved by the following technical solutions:

[0006] An emergency evacuation path planning method based on the HBA optimized ant colony algorithm, the method comprising the following steps:

[0007] S1. Construct a two-dimensional map of the indoor space, including indoor obstacles, evacuation routes, evacuation exits, and emergency danger areas;

[0008] S2. Based on the two-dimensional map of the indoor space constructed in step S1, initialize the number and position of the population, update the density factor, and obtain the initial optimal path trajectory x new ;

[0009] S3. Use the initial path trajectory x obtained in step S2 new to optimize the initial pheromone of the ant colony algorithm;

[0010] S4. Introduce distance estimation and smoothing functions to improve the transition probability selection of the ant colony algorithm;

[0011] S5. Update the pheromone of the ant colony algorithm;

[0012] S6. After iteration, form the optimal path Z and output it.

[0013] Furthermore, step S2 includes the following sub-steps:

[0014] S2-1. Based on the honey badger optimization algorithm, initialize the number and position of the population. The specific formula is as follows:

[0015] x n = lb n + r 1 × (ub n - lb n )

[0016] where x n represents the position of the nth honey badger, lb n represents the lower bound of the search area of the nth honey badger, ub n represents the upper bound of the search area of the nth honey badger, and r 1 represents a random number between 0 and 1;

[0017] S2-2. First, give the expression formula related to the density factor as follows:

[0018]

[0019] S = (x n - x n+1 ) 2

[0020] d n = x prey - x n

[0021] where, I nDenote the olfactory intensity of the nth honey badger, r 2 is a random number in (0, 1), S is the concentration intensity between the honey badger and the prey, d n denotes the Euclidean distance between the nth honey badger and the prey, x prey represents the current global optimal solution in the mathematical formula and is used to simulate the best position of the prey in the honey badger optimization algorithm, x n+1 represents the position of the (n + 1)th honey badger;

[0022] The following density factor update formula is given:

[0023]

[0024] where α represents the density factor; C = 2 represents a constant; NM represents the number of iterations; NM max represents the maximum number of iterations; exp(·) represents the exponential function;

[0025] S2 - 3. In the excavation stage, the following formula is satisfied:

[0026] x new = x prey + F×β×I×x prey + F×r 3 ×α×d n ×|cos(2πr 4 )×[1 - cos(2πr 5 )]|

[0027] where x new represents the new position of the honey badger, x prey represents the best position of the prey, β = 6 represents the predation ability of the honey badger, F is the search direction, which is used to provide the scanning space and is specifically represented as follows:

[0028]

[0029] where r 6 is a random number between 0 and 1;

[0030] S2 - 4. The movement trajectory of the honey badger following the honey - guiding bird to the honey source in the honey - collecting stage satisfies the following formula:

[0031] x new = x prey + F×r 7 ×α×d n

[0032] where r 7 represents a random number, and α represents the density factor.

[0033] Further, the initial pheromone of the optimized ant colony algorithm in step S3 is as follows after optimization:

[0034] τ ij (x new ) = XR

[0035] Among them, τ ij (x new ) represents the initialized pheromone, X = 1.5 is the adjustment constant; R = 0.3 represents the pheromone on other paths.

[0036] Further, the transfer probability selection of the improved ant colony algorithm by introducing distance estimation and smoothing function in step S4 is as follows after improvement: The formula expression is as follows:

[0037]

[0038] Among them, k represents the selectable node; l represents the heuristic factor of pheromone; m represents the heuristic factor function; τ ij (t) represents the pheromone from node i to node j of the ant at time t; A k represents the set of selectable nodes of the ant; η ij (t) represents the heuristic function from node i to node j of the ant at time t; λ ij (t) represents the introduced smoothing function; and:

[0039]

[0040] Among them, d j is the distance from node j to the target point, ε is the balance factor, d si is the distance from the starting point s to node i, d ij represents the distance from node i to node j, d sj represents the distance from the starting point s to node j, N c and N q are the number of iterations;

[0041]

[0042] Among them, θ = 0.8 represents the smoothness percentage; dir ij (t) represents the transfer direction from the current node i to the next node j at time t; dir iv (t) represents the transfer direction from the previous node v to the current node i at time t.

[0043] Further, the expression of pheromone update in step S5 is as follows:

[0044] τ ij(t + 1) = (1 - ρ)τ ij (t) + Δτ ij (t)

[0045]

[0046] where ρ represents the pheromone value; Δτ ij (t) is the pheromone increment when the ant passes through the path coordinates (i, j) at time t; U is the pheromone enhancement coefficient; L k is the total path length from the starting point to the target point for the ant.

[0047] The beneficial effects of the present invention compared with the prior art are as follows:

[0048] (1) Aiming at the problem that the ant colony algorithm is prone to falling into local optimum, the present invention introduces the honey badger optimization algorithm into the ant colony algorithm, thereby avoiding the problem of the ant colony algorithm falling into local optimum.

[0049] (2) Based on the traditional ant colony algorithm, the present invention uses a Manhattan-based distance estimation method to guide the ant colony algorithm to search in the direction of the shortest target;

[0050] (3) The present invention designs a smoothing function to minimize the number of turns during the personnel evacuation process and improve the stability and safety of the personnel evacuation route planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is the flow chart for implementing the method of the present invention;

[0052] Figure 2 is the optimization result of the personnel evacuation path by the traditional fireworks algorithm;

[0053] Figure 3 is the optimization result of the personnel evacuation path by the method of the present invention;

[0054] Figure 4 is the comparison curve of the optimization results of the two algorithms, namely the traditional fireworks algorithm and the method of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0055] The following further describes the present invention in conjunction with the drawings and specific examples.

[0056] As Figure 1 shown, in an indoor known space, when an emergency occurs, people need to quickly plan an escape route to leave the dangerous area in the shortest time. The present invention provides an emergency evacuation path planning method based on the HBA optimized ant colony algorithm, enabling the evacuees to quickly escape from the dangerous area according to the route calculated by the system. The specific implementation steps are as follows:

[0057] Step 1: Update of Honey Badger Optimization Algorithm

[0058] ①Construct a two-dimensional map of the indoor space, including information such as indoor obstacles, evacuation channels, evacuation exits, and dangerous areas of emergencies.

[0059] ②Based on the above map, initialize the number and position of the population using the Honey Badger Optimization Algorithm. The specific formula is as follows:

[0060] x n = lb n + r 1 ×(ub n - lb n )

[0061] Where x n represents the position of the nth honey badger, lb n represents the lower bound of the search area of the nth honey badger, ub n represents the upper bound of the search area of the nth honey badger, and r 1 represents a random number between 0 and 1.

[0062] ③Give and update the density factor

[0063] In the Honey Badger Optimization Algorithm, considering that the greater the olfactory intensity, the faster the movement speed of the honey badger, and the smaller the olfactory intensity, the slower the movement speed of the honey badger. Therefore, the present invention gives the following expression formula for the density factor:

[0064]

[0065] S = (x n - x n+1 ) 2

[0066] d n = x prey - x n

[0067] Where, I n represents the olfactory intensity of the nth honey badger, r 2 is a random number in (0, 1), S is the concentration intensity between the honey badger and the prey, d n represents the Euclidean distance between the nth honey badger and the prey, and x prey represents the current global optimal solution in the mathematical formula and is used to simulate the best position of the prey in the Honey Badger Optimization Algorithm, and x n+1 represents the position of the (n + 1)th honey badger.

[0068] Since the density factor changes with time, the present invention gives the following density factor update formula:

[0069]

[0070] Among them, α represents the density factor; C = 2 represents a constant; NM represents the number of iterations; NM max represents the maximum number of iterations; exp(·) represents the exponential function.

[0071] ④ Mining stage

[0072] In the mining stage, the following formula is satisfied:

[0073] x new = x prey + F × β × I × x prey + F × r 3 × α × d n × |cos(2πr 4 ) × [1 - cos(2πr 5 )]|

[0074] Among them, x new represents the new position of the honey badger, x prey represents the best position of the prey, β = 6 represents the predation ability of the honey badger, F is the search direction, which is used to provide a scanning space, and is specifically expressed as follows:

[0075]

[0076] Among them, r 6 is a random number between 0 and 1.

[0077] ⑤ Honey collection stage

[0078] The movement trajectory of the honey badger following the honey guide bird to the honey source satisfies the following formula:

[0079] x new = x prey + F × r 7 × α × d n

[0080] Among them, r 7 represents a random number, and α represents the density factor.

[0081] Step 2: Initial pheromone optimization processing of the ant colony algorithm

[0082] The present invention uses the updated result of the above honey badger optimization algorithm to optimize the initial pheromone of the ant colony algorithm. Through the above steps, an optimal path trajectory x new has been obtained. At this time, the pheromone on this path is set to be higher than that on other paths to play the guiding role of the honey badger optimization algorithm in the early stage of the ant colony algorithm. Furthermore, it avoids falling into local optimality and improves the convergence speed of the algorithm. The optimized initial pheromone is expressed as follows:

[0083] τij (x new ) = XR

[0084] Among them, τ ij (x new ) represents the initialized pheromone, X = 1.5 is the adjustment constant; R = 0.3 represents the pheromone on other paths.

[0085] Step 3: Improved transition probability selection

[0086] In the traditional ant colony algorithm, the next position is selected according to the grid and obstacles during the search process. Generally, the roulette method is used to implement the probability of moving from node i to node j at time t. In the present invention, distance estimation and a smoothing function are introduced based on the traditional ant colony algorithm's transition probability selection formula, and the improved transition probability selection is expressed by the following formula:

[0087]

[0088] Among them, k represents the selectable nodes; l represents the heuristic factor of the pheromone; m represents the heuristic factor function; τ ij (t) represents the pheromone from node i to node j of the ant at time t; A k represents the set of selectable nodes of the ant; η ij (t) represents the heuristic function from node i to node j of the ant at time t; λ ij (t) represents the introduced smoothing function.

[0089] The following explains two improved parts of the present invention involved above.

[0090] ① Among them, η ij (t) is the heuristic function introducing distance estimation in the present invention. The estimated distance uses the Manhattan distance. On the basis of introducing the estimated value, a balance factor is added to accelerate the convergence of the algorithm. The improved heuristic function in the present invention is expressed as follows:

[0091]

[0092] Among them, d j is the distance from node j to the target point, ε is the balance factor, d si is the distance from the starting point s to node i, d ij represents the distance from node i to node j, d sj represents the distance from the starting point s to node j, N c and N q are the number of iterations.

[0093] ② λ ij (t) represents the smoothing function introduced in the present invention, specifically as follows:

[0094] For safety considerations during the evacuation process, the number of turns is minimized during evacuation. The present invention incorporates a smoothing function λ ij (t) to improve the stability of the algorithm and the safety during the evacuation process, and the specific expression is as follows:

[0095]

[0096] where θ = 0.8 represents the smoothing percentage; dir ij (t) represents the transfer direction from the current node i to the next node j at time t; dir iv (t) represents the transfer direction from the previous node v to the current node i at time t.

[0097] Step Four: Pheromone Update

[0098] The pheromone update expression in the traditional ant colony algorithm is adopted here as follows:

[0099] τ ij (t + 1) = (1 - ρ)τ ij (t) + Δτ ij (t)

[0100]

[0101] where ρ represents the pheromone value; Δτ ij (t) is the pheromone increment when the ant passes through the path coordinates (i, j) at time t; U is the pheromone reinforcement coefficient; L k is the total path length from the starting point to the target point of the ant.

[0102] Step Five: Optimal Path Output

[0103] In the above continuous iteration process, the optimal path Z is finally formed.

[0104] Experimental Verification:

[0105] To verify the effectiveness of an emergency evacuation path planning method based on the HBA - optimized ant colony algorithm proposed by the present invention, relevant experiments are designed. In the experiment, a certain range of grid area is set as the evacuation space, in which a certain number of irregular obstacles are set. The starting point of the evacuation is the upper - left corner of the grid area, and the evacuation exit / end point is located in the lower - right corner of the grid area. Under the above conditions, the evacuation path planning experiments are respectively carried out using the traditional ant colony algorithm and the method of the present invention.

[0106] Results: Figure 1 Flowchart of the method of the present invention. Figure 2 Emergency evacuation planning route based on the traditional ant colony algorithm. Figure 3Emergency evacuation planning route of the method of the present invention. Figure 4 Comparison curve of path lengths planned by different methods. By Figure 4 It can be seen that the method of the present invention achieves the shortest path length with the fewest number of iterations. Among them, the traditional ant colony algorithm takes 5.176471 seconds in the entire route planning process, while the method of the present invention takes 4.784792 seconds, indicating that the method of the present invention is more efficient. In terms of the average path length, the path length planned by the traditional ant colony algorithm is 34.0933 meters, and the path length planned by the method of the present invention is 29.9120 meters, indicating that the planning of the present invention is more reasonable.

Claims

1. An emergency evacuation path planning method based on HBA optimized ant colony algorithm, characterized in that: The method comprises the following steps: S1. Construct a two-dimensional map of the indoor space, including indoor obstacles, evacuation passages, evacuation exits, and emergency danger areas; S2. Based on the two-dimensional map of the indoor space constructed in step S1, the number and position of the population are initialized, the density factor is updated, and the initial optimal path trajectory x is obtained based on the honey badger optimization algorithm. new ; S3. Use the initial path trajectory x obtained in step S2 new Optimize the initial pheromone of the ant colony algorithm; S4. Introducing distance estimation and smoothing function to improve the transition probability selection of ant colony algorithm; S5.Pheromone update of ant colony algorithm; S6. After iteration, the optimal path Z is output.

2. The method for emergency evacuation path planning based on HBA optimized ant colony algorithm according to claim 1, characterized in that: Step S2 includes the following sub-steps: S2-1. Based on the honey badger optimization algorithm, initialize the number and location of the population. The specific formula is as follows: x n =lb n +r1×(ub n -lb n ) where x n Indicates the position of the nth honey badger, lb n represents the lower bound of the search area of ​​the nth honey badger, ub n represents the upper bound of the search area of ​​the nth honey badger, and r1 represents a random number between 0 and 1; S2-2. First, the density factor related expression formula is given as follows: S=(x n -x n+1 ) 2 d n =x prey -x n Among them, I n represents the olfactory intensity of the nth honey badger, r2 is a random number between (0,1), S is the concentration intensity between the honey badger and its prey, and d n represents the Euclidean distance between the nth honey badger and its prey, x prey In mathematical formulas, it represents the current global optimal solution, which is used to simulate the optimal position of prey in the honey badger optimization algorithm. n+1 Indicates the position of the n+1th honey badger; Then give the following density factor update formula: Where α represents the density factor; C = 2 represents a constant; NM represents the number of iterations; NM max represents the maximum number of iterations; exp(·) represents the exponential function; S2-3. In the data mining stage, the following formula is satisfied: x new =x prey +F×β×I×x prey +F×r3×α×d n ×|cos(2πr4)×[1-cos(2πr5)]| Among them, x new represents the new position of the honey badger, x prey represents the best position of the prey, β=6 represents the predation ability of the honey badger, and F is the search direction, which is used to provide the scanning space. The specific expression is as follows: Among them, r6 is a random number between 0 and 1; S2-4. During the honey collection stage, the movement trajectory of the honey badger following the honey guide bird to the nectar source satisfies the following formula: x new =x prey +F×r7×α×d n Where r7 represents a random number and α represents a density factor.

3. The method for emergency evacuation path planning based on HBA optimized ant colony algorithm according to claim 2, characterized in that: In step S3, the initial pheromone of the ant colony algorithm is optimized, and the optimized initial pheromone is expressed as follows: τ ij (x new )=XR Among them, τ ij (x new ) represents the initialization pheromone, X=1.5 is the adjustment constant; R=0.3 represents the pheromone on other paths.

4. The method for emergency evacuation path planning based on HBA optimized ant colony algorithm according to claim 3 is characterized in that: Step S4 introduces distance estimation and smoothing function to improve the transition probability selection of the ant colony algorithm. The improved transition probability selection The formula is as follows: Where k represents the selectable node; l represents the heuristic factor of pheromone; m represents the heuristic factor function; τ ij (t) represents the pheromone from the ant from node i to node j at time t; A k represents the set of nodes that the ant can choose; η ij (t) represents the heuristic function of the ant from node i to node j at time t; ij (t) represents the introduced smoothing function; and: Among them, d j is the distance from node j to the target point, ε is the balance factor, d si is the distance from the starting point s to the node i, d ij represents the distance from node i to node j, d sj Indicates the distance from the starting point s to the node j, N c and N q is the number of iterations; Among them, θ = 0.8 represents the smoothness percentage; dir ij (t) represents the transfer direction from the current node i to the next node j at time t; dir iv (t) represents the transfer direction from the previous node v to the current node i at time t.

5. The method for emergency evacuation path planning based on HBA optimized ant colony algorithm according to claim 4 is characterized in that: The expression of pheromone update in step S5 is as follows: t ij (t+1)=(1-ρ)τ ij (t)+Δτ ij (t) Where ρ represents the pheromone value; Δτ ij (t) is the pheromone increment when the ant passes through the path coordinate (i, j) at time t; U is the pheromone reinforcement coefficient; L k is the total length of the ant's path from the starting point to the target point.

Images