Emergency evacuation path planning method considering dynamic inclination of passenger ship, program, equipment and storage medium

Abstract

The invention discloses an emergency evacuation path planning method, program and equipment considering dynamic inclination of a passenger ship, and a storage medium, and belongs to the field of personnel evacuation simulation modeling. The method comprises the following steps: 1, obtaining a passenger ship inclination angle and a pedestrian moving direction, and constructing a pedestrian speed attenuation factor reasoning model; 2, analyzing and modeling the influence of pedestrian mass movement on ship body inclination; 3, introducing a congestion penalty factor, a safety penalty coefficient and an attenuation factor based on a path cost function of a D * Lite algorithm, and obtaining a pre-planned path according to the starting point and the exit point of the leader; and 4, presetting a re-planning trigger condition, and when the preset trigger condition is met, taking the current node as a new starting point, and repeating the steps 1-4 to re-plan the evacuation route until the whole evacuation process is completed. According to the method, when an emergency occurs, the evacuation strategy can be adjusted according to the inclination angle, the exit congestion and the passenger distribution, and the evacuation efficiency and safety in a passenger ship evacuation scene are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of personnel evacuation simulation modeling, and in particular to an emergency evacuation path planning method, program, equipment and storage medium taking into account the dynamic tilt of a passenger ship. Background Art

[0002] Passenger ships have attracted more and more tourists and experienced vigorous development due to their integrated features of transportation, accommodation and entertainment services. However, ensuring the safety of passengers and crew members in the event of a maritime accident remains a major challenge. Emergency evacuation route planning, as an important part of the passenger ship evacuation management system, helps provide safe and effective evacuation guidance in emergency situations and ensures the safety of passengers in emergency situations.

[0003] Path planning technology improves the efficiency and safety of pedestrians by determining the best traffic routes for them. Typically, path planning involves solving a multi-objective optimization problem that requires comprehensive consideration of multiple factors such as path length, safety, congestion, and other key factors. Compared with land-based building environments, passenger ships pose unique challenges to the design of path planning algorithms due to their complex internal layout and narrow passages. In addition, the hull tilt caused by factors such as wind and waves makes the movement of pedestrians more complicated, and it is critical that ship evacuations meet strict time limits as delays may compromise the personal safety of passengers. By leveraging the high-precision monitoring and real-time communication capabilities of wireless sensor networks (WSNs), environmental changes can be effectively detected in real time, followed by evacuation strategy formulation by the evacuation management platform. Leaders equipped with intelligent navigation devices guide pedestrians to complete the evacuation after receiving instructions. This approach greatly improves the efficiency and safety of the passenger ship evacuation process.

[0004] The tilt of the ship is essentially a dynamic process, and its impact on the movement of pedestrians and the interaction of pedestrian movement with the tilt of the ship need to be considered more comprehensively, because the speed of pedestrians on the tilted plane will be attenuated, and the uneven mass distribution will aggravate the tilt of the ship. Therefore, it is very important to develop a reasonable path planning strategy for the dynamic tilt of passenger ships, which should not only consider the efficiency of evacuation, but also minimize the impact of the uneven distribution of pedestrian mass on the tilt of passenger ships, reduce risks, and extend the available evacuation time. Summary of the invention

[0005] In view of the shortcomings of the existing methods, the present invention provides an emergency evacuation path planning method taking into account the dynamic tilt of a passenger ship, which helps leaders guide pedestrians to complete the evacuation more safely and efficiently.

[0006] The present invention provides an emergency evacuation path planning method considering the dynamic tilt of a passenger ship, comprising the following steps:

[0007] Step 1: Obtain the tilt angle of the passenger ship and the moving direction of the pedestrian, and build an inference model of the pedestrian speed attenuation factor;

[0008] Step 2: Analyze and model the effect of pedestrian mass movement on the hull tilt;

[0009] Step 3: Design an evacuation path planning algorithm based on the improved D*Lite algorithm; introduce congestion penalty factor, safety penalty coefficient and attenuation factor based on the path cost function of the D*Lite algorithm, and obtain the pre-planned path through the improved D*Lite algorithm according to the starting point of the leader and the exit point of the passenger ship;

[0010] Step 4: Preset re-planning trigger conditions. When the preset trigger conditions are met, the current node is used as the new starting point, and steps 1 to 3 are repeated to re-plan the evacuation route, and evacuate according to the new evacuation route until the entire evacuation process is completed.

[0011] Furthermore, the step 1 comprises the following steps:

[0012] Step 1: Substitute the tilt angle θ and the pedestrian movement direction with the tilt direction angle It is input into the fuzzy controller as an input variable; the speed attenuation factor δ is used as the output variable of the fuzzy controller;

[0013] The fuzzy set of input variables is:

[0014] Tilt angle θ = {S θ : Small tilt angle, M θ : Middle tilt angle, L θ : Large tilt angle}; S θ and L θ The membership function of M is the generalized Bell membership function. θ The membership function of selects Gaussian membership function;

[0015] The angle between the pedestrian's movement direction and the tilt direction

[0016] The membership functions of are all generalized Bell membership functions;

[0017] Step 2: After fuzzification, a fuzzy control rule base is formulated based on if-then rules and 15 control rules are obtained and fuzzy reasoning is performed; the fuzzy set of output variables is: attenuation factor δ' = {VVS δ' : The attenuation factor is very small, VS δ' : The attenuation factor is small, S δ' The attenuation factor is small, M δ' The attenuation factor is moderate, L δ' Large attenuation factor, VL δ'The attenuation factor is large, VVL δ' The attenuation factor is large}; VVS δ' and VVL δ' The membership function of VS is trapezoidal membership function; δ' , S δ' , M δ' , L δ' , VL δ' The membership functions of all select the triangle membership function;

[0018] Step 3: Use the centroid defuzzification method to defuzzify the attenuation factor δ' and convert the control decision obtained by fuzzy reasoning into an accurate attenuation factor δ.

[0019] Furthermore, the step 2 comprises the following steps:

[0020] Step 1: Calculate the position G1 of the center of gravity of the ship after the pedestrian moves:

[0021]

[0022] Among them, w g is the mass of the object; d is the distance the object moves; w G is the draft of the ship; G is the center of gravity of the ship before the pedestrian moves;

[0023] Step 2: Calculate the tilting force arm GZ' of the hull after the pedestrian moves:

[0024]

[0025] Among them, GZ is the forward tilt arm of the pedestrian;

[0026] The area of ​​the GZ curve of the ship tilt arm before the pedestrian moves must be greater than or equal to 3.151m.deg when the ship tilt angle is 0° to 30°, and the area of ​​the GZ curve of the ship tilt arm before the pedestrian moves must be greater than or equal to 5.157m.deg when the ship tilt angle is 0° to 40°. The GZ curve of the ship tilt arm before the pedestrian moves after fitting is:

[0027] GZ=0.44267θ-0.299356θ 3

[0028] Furthermore, the step 3 comprises the following steps:

[0029] Step 1: When the g(v) value and rhs(v) value of node v are not equal, the state of node v is in a local inconsistent state. It needs to be included in the detection queue, and the priority of node v is determined according to the key value key(v);

[0030]

[0031] Among them, g(v) and rhs(v) are the total costs from node v to the target point; Δ m is a key value corrector, Δ m =0 is the initial value; the heuristic value h(v,v start ) represents the node v to the starting node v start The estimated cost of; c(v,v′) represents the actual cost from node v to its successor node v′;

[0032] Step 2: Based on the distance cost, the calculation method of the modified cost is as follows, taking into account the speed reduction degree, congestion degree and safety degree:

[0033] c(v,v')=(1+ρ)·(1+γ·ψ·sinθ)·dis(v,v') / δ

[0034] Where dis(v,v′) is the distance from node v to the next node v′; (1+γ·ψ·sinθ) is the safety penalty coefficient; γ is 1 when the direction from node v to the next node v′ is downhill, and γ is 0 when the direction from node v to the next node v′ is downhill; ψ is the proportional factor, which is related to the number of hull layers, and the node of the kth layer takes the value of 2 k ;δ is the speed attenuation factor of the road section (v,v′); (1+ρ) is the congestion penalty factor;

[0035] The congestion penalty factor (1+ρ) is calculated using the pheromone concentration in the ant colony algorithm, and the pheromone concentration represents the congestion level at the next node v′;

[0036] ρ=a g ·ρ+(1-a g )·ρ l

[0037] ρ l =a l ·ρ l +(1-a l )·ε l

[0038] Where ρ is the global pheromone concentration at the next node; a g is the global volatility factor; ρ l It is a local pheromone; a l is the local volatility factor; ε l is the amount of pheromone left behind by a pedestrian each time he moves.

[0039] Furthermore, in step 4, the triggering condition for re-planning is:

[0040] if ΔGZ>1.5andθ≥15

[0041] ifVar(ρ e )>0.3

[0042]

[0043] Where ΔGZ is the change of the ship tilting force arm caused by the movement of the ship's center of gravity when the pedestrian moves; Var(ρ e ) is the standard deviation of each outlet pheromone concentration;

[0044] When the hull tilt arm change ΔGZ and the hull tilt angle θ satisfy if ΔGZ>1.5andθ≥15, it means that the evacuation process is in a potentially dangerous state due to the movement of pedestrians, and steps 1 to 3 should be repeated to replan the evacuation route; when Var(ρ e ) satisfies ifVar(ρ e )>0.3, the pheromone concentration at the exit is too high and congestion occurs at the exit, indicating that the pedestrian distribution at each exit is unbalanced and the exit is not fully utilized. Steps 1 to 3 should be repeated to replan the evacuation route.

[0045] Furthermore, the leader is a security officer on the passenger ship, and the security officer guides the movement of pedestrians around.

[0046] The present invention also provides a computer device / equipment / system, comprising a memory, a processor and a computer program stored in the memory, wherein when the processor executes the computer program, the steps of any of the above-mentioned methods for emergency evacuation path planning that takes into account the dynamic tilt of a passenger ship are implemented.

[0047] The present invention also provides a computer-readable storage medium having a computer program / instruction stored thereon, which, when executed by a processor, implements the steps of any of the above-mentioned emergency evacuation path planning methods taking into account the dynamic tilt of a passenger ship.

[0048] The present invention also provides a computer program product, comprising a computer program / instruction, which, when executed by a processor, implements the steps of any of the above-mentioned emergency evacuation path planning methods considering the dynamic tilt of a passenger ship.

[0049] The beneficial effects of the present invention are:

[0050] 1. The method of the present invention is applicable to emergency situations where a passenger ship tilts due to wind and waves, water ingress into the hull, etc.

[0051] 2. The method of the present invention provides a speed attenuation factor quantification method based on fuzzy logic reasoning, which can effectively evaluate the movement of pedestrians on an inclined plane.

[0052] 3. The method of the present invention introduces a pheromone-based congestion degree calculation method, which can effectively evaluate the congestion situation at the node.

[0053] 4. The method of the present invention provides a path planning algorithm that comprehensively considers the impact of the passenger ship's tilt on pedestrians, the impact of congestion, and the impact of pedestrian movement on the hull tilt, which can improve the efficiency and safety of evacuation. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a flow chart of the emergency evacuation route planning method of the present invention;

[0055] Figure 2 It is the inclination angle membership curve diagram required in the fuzzy reasoning method of the present invention;

[0056] Figure 3 It is the moving direction membership curve diagram required in the fuzzy reasoning method of the present invention;

[0057] Figure 4 It is a speed attenuation factor membership curve diagram required in the fuzzy reasoning method of the present invention;

[0058] Figure 5 It is the output accurate value diagram required in the fuzzy inference method of the present invention;

[0059] Figure 6 The evacuation scene structure and the initial distribution map of pedestrians of the present invention;

[0060] Figure 7 This is a comparison diagram of evacuation time at different tilt angles of the present invention;

[0061] Figure 8 It is a graph showing the change of the remaining number of people over time at different tilt angles of the present invention;

[0062] Fig. 9 It is the evacuation density map of the algorithm of the present invention without considering the degree of congestion;

[0063] Fig.10 An improved algorithm evacuation density map considering the degree of congestion according to the present invention;

[0064] Fig.11 It is a comparison diagram of the remaining number of people changing with time under the dynamic inclination angle of the present invention (starting to change at 0°);

[0065] Fig.12 This is a comparison chart of the remaining number of people changing over time under the dynamic inclination angle of the present invention (starting to change at 10°). DETAILED DESCRIPTION

[0066] The present invention will be further described below in conjunction with the accompanying drawings:

[0067] The present invention provides an emergency evacuation path planning method taking into account the dynamic tilt of a passenger ship. Figure 1 As shown, the following steps are included:

[0068] Step S1, considering the coupling effect of pedestrian movement direction and tilt angle, construct a pedestrian speed attenuation factor inference model.

[0069] In the fuzzy inference system, the first input is the tilt angle θ, which is divided into a fuzzy set {S θ ,M θ ,L θ}, where S θ 、M θ , L θ Represent small, medium and large tilt angles respectively, and their membership curves are as follows Figure 2 As shown. The membership functions of S and L are generated by the generalized Bell membership function shown in equation (1), while the membership function of M is generated by the Gaussian membership function shown in equation (2). In equations (1) to (4), x is an independent variable, and a, b, c, e, f, g, h, i, j, k, m, n are constants.

[0070]

[0071]

[0072] The second input variable is the angle between the pedestrian's movement direction and the tilt direction From the smallest to the largest, the fuzzy set is established as The membership of each fuzzy element is generated by the generalized Bell membership function, and its membership curve is as follows: Figure 3 shown.

[0073] After fuzzification, it is necessary to establish fuzzy rules based on if-then rules to form the reasoning process of the attenuation factor δ. Since the first input has 3 fuzzy elements and the second input has 5 fuzzy elements, 15 fuzzy rules need to be established. In this invention, we define the output attenuation factor δ' fuzzy set as {VVS δ' ,VS δ' ,S δ' ,M δ' ,L δ' ,VL δ' ,VVL δ'}, its membership curve is as follows Figure 4 As shown. VVS δ' and VVL δ' The membership function of is represented by the trapezoidal membership function in equation (3), and the other elements are represented by the triangular membership function in equation (4).

[0074] Finally, the fuzzy representation of the output must be converted to an exact value, a process called defuzzification. This paper uses the centroid defuzzification method to achieve this goal, and establishes a Mamdani fuzzy logic system to derive the specific value of the attenuation factor. The output under different inputs is as follows Figure 5 shown.

[0075] Step S2, analyzing and modeling the effect of pedestrian mass movement on the hull tilt.

[0076] Considering the passenger ship as a rigid body, when the objects on the ship move in different directions, the center of gravity of the ship will move accordingly. The direction of this movement is parallel to the direction of the object's movement. The position of the center of gravity of the ship before and after the object moves is recorded as G and G1 respectively. The position after the center of gravity changes can be calculated using the following formula.

[0077]

[0078] Among them, w g is the mass of the object, d is the distance the object moves, and w G is the draft of the ship. The stability of the ship depends on the direction in which the object is moved. Usually, moving in the direction of the tilt will increase the tilt of the hull.

[0079] Considering that the tilting arm of the ship before the object moves is recorded as GZ, the new effective arm can be determined by considering the weight transfer. The calculation method is as follows.

[0080]

[0081] According to the stability criteria of the International Maritime Organization (IMO), the relationship between the lever arm and the inclination angle can be expressed as a nonlinear function, usually expressed as the following formula.

[0082] GZ=π1θ+π3θ 3 +π5θ 5 +π7θ 7 +... (7)

[0083] To simplify matters, we have selected the first two criteria from the IMO's stability guidelines:

[0084] (1) The area under the GZ curve from 0° to 30° must be greater than or equal to 3.151 m.deg.

[0085] (2) The area under the GZ curve from 0° to 40° must be greater than or equal to 5.157 m.deg.

[0086] Based on these criteria, we constructed a third-order estimating equation and the fitting curve is shown below.

[0087] GZ=0.442267θ-0.299356θ 3 (8)

[0088] Step S3, designing an evacuation path planning algorithm based on the D*Lite algorithm.

[0089] Considering that there are a large number of passengers on a passenger ship, it is impractical to plan an evacuation path for each person individually. In the present invention, this problem is solved by considering the presence of security personnel on board, who are regarded as leaders, plan evacuation paths for them, guide pedestrians around, and ensure an orderly evacuation process.

[0090] The D*Lite algorithm is a variant of the A* algorithm, which combines the incremental search characteristics of the LPA* algorithm and the reverse search capability of the D* algorithm, and has the ability to dynamically plan paths. Therefore, the present invention makes improvements based on the D*Lite algorithm and designs an evacuation path planning algorithm for passenger ships under dynamic inclination conditions.

[0091] In the D*Lite algorithm, the concept of priority queue (denoted as U) is introduced to store nodes that need to be explored. When the g(v) value and rhs(v) value of node v are not equal, the state of node v is locally inconsistent and needs to be included in the detection queue. Node v will be stored in U and the priority of node v will be determined according to the key value key(v). The expression of key(v) is as follows.

[0092]

[0093] Where g(v) and rhs(v) both represent the total cost from node v to the destination, and since g(v) will become outdated due to environmental changes, rhs(v) is always updated first and then assigned to g(v). m is a key-value corrector that is updated every time the agent moves to a new node.

[0094]

[0095]

[0096] Typically, the heuristic value h(v,v start ) represents the estimated cost from node v to the starting node, which is usually calculated using Euclidean distance or Manhattan distance, and c(v,v′) represents the actual cost from node v to its successor node v′.

[0097] The present invention considers several additional factors on the basis of distance cost, including the influence of passenger ship tilt, node congestion, and the influence of passenger distribution on passenger ship tilt. On this basis, an improved D*Lite algorithm is proposed, and its path cost is modified as follows.

[0098] The cost value of each door is established according to the distribution of pedestrians and the distance between the leader and the exit. The leader selects the exit according to the size of the cost value. The cost value of the door is expressed as follows:

[0099] c(v,v′)=(1+ρ)·(1+γ·ψ·sinθ)·dis(v,v′) / δ (12)

[0100] Where dis(v,v′) is the distance from v to v′, (1+γ·ψ·sinθ) is the safety penalty coefficient. When the direction from v to v′ is downhill, γ is 1, otherwise γ is 0. ψ is a proportional factor that controls the penalty based on the number of decks. The higher the number of decks, the larger the value, because the movement of the center of gravity on the higher deck has a greater impact on the ship's tilt. The node value of the kth layer is 2 k ; For a three-deck hull, the values ​​of ψ are set to 2, 4, and 8 for the nodes on decks 1, 2, and 3, respectively. δ is the velocity attenuation factor of the segment (v, v′), which is calculated using the above fuzzy inference method.

[0101] (1+ρ) represents the congestion penalty factor. The present invention borrows the concept of pheromone in the ant colony algorithm and uses the pheromone concentration to quantify the congestion on a node. ρ is the global pheromone concentration at the node v′, which is expressed as follows.

[0102] ρ=a g ·ρ+(1-a g )·ρ l (13)

[0103] ρ l =a l ·ρ l +(1-a l )·ε l (14)

[0104] Among them, a g is the global volatility factor, ρ l It is a local pheromone, a l is the local volatility factor, ε l is the number of pheromones left by the pedestrian each time he moves. Each time the pedestrian chooses a node v′ as a target, a local pheromone update is performed, while the global pheromone is updated once per second. g 、a l and ε l The values ​​of are set to 0.7, 0.7, and 2 respectively.

[0105] In order to better adapt to the dynamic characteristics of the passenger ship evacuation environment, the present invention proposes the following re-planning triggering conditions.

[0106] if ΔGZ>1.5andθ≥15 (15)

[0107] ifVar(ρ e )>0.3 (16)

[0108]

[0109] Among them, ΔGZ represents the change of the force arm caused by the movement of the center of gravity of the ship when the pedestrian moves, Var(ρ e ) represents the standard deviation of the pheromone concentration at each exit. When condition (15) is met, it indicates that the evacuation process is in a potentially dangerous state due to the movement of pedestrians, and the route needs to be replanned to enhance evacuation safety. On the other hand, when condition (16) is met, it indicates that the distribution of pedestrians at each exit is unbalanced, some exits are not fully utilized, and the evacuation route should be replanned to optimize efficiency.

[0110] Step S4, establishing an evacuation simulation environment and verifying the performance of the path planning algorithm.

[0111] The evacuation environment was simulated using AnyLogic software. Figure 6 As shown, the ship is 130 meters long, 20 meters wide, with a total tonnage of 15,000 tons and a maximum passenger capacity of 650 people. A total of 600 passengers are evenly distributed on each deck, with 200 people on each deck. The first and second decks each contain 10 leaders, while the third deck has 4 leaders.

[0112] The motion of pedestrians is driven by a social force model. For simplicity, we ignore the heterogeneity of pedestrians. Under normal circumstances, it is difficult for pedestrians to maintain balance when the deck tilt exceeds 35 degrees. In addition, in order to calculate the change of the moment arm caused by the movement of pedestrian weight, the position of each pedestrian must be known. Therefore, we make the following reasonable assumptions.

[0113] (1) If the tilt angle exceeds 35 degrees, the evacuation ends and pedestrians who fail to reach the exit are considered to have failed to evacuate.

[0114] (2) The location of pedestrians is monitored in real time by wireless sensor networks (WSNs) with sufficiently high accuracy.

[0115] (3) The central control platform makes reasonable evacuation route decisions based on the acquired data, and leaders can receive instructions and guide evacuation through smart devices such as mobile phones.

[0116] The evacuation simulation settings for the effect of ship tilt on pedestrians are as follows:

[0117] When pedestrians move perpendicular to the inclined direction, the speed decay is less than when they move horizontally. Therefore, in evacuation path planning, using roads with lower speed decay can significantly improve evacuation efficiency. In this paper, an enhanced D*Lite algorithm, called AD*Lite algorithm, is proposed, which incorporates the speed decay factor into the edge cost calculation.

[0118] In order to evaluate the effectiveness of the AD*Lite algorithm, simulation experiments were conducted at different tilt angles and directions. The tilt angles were 10° and 20°, respectively. The tilt direction α represents the uphill direction. 0° (left pitch), 90° (lateral pitch), and 180° (right pitch) were considered, resulting in a total of 6 different experimental conditions. For each experimental condition, 10 simulations were performed to calculate the average evacuation time.

[0119] like Figure 7 As shown in the figure, it can be seen from the experimental results that when the ship is heeling, the evacuation time under the two path planning algorithms is similar. This is because the evacuation exits are distributed on the left and right sides of the hull. During most of the evacuation process, pedestrians will move horizontally, so the influence of the trim on the evacuation process is greater than that of the heel.

[0120] In order to more clearly compare the evacuation efficiency of the two algorithms under the left pitch condition, we draw a curve representing the variation of the remaining number of people in the room over time, such as Figure 8 The results show that under the conditions of left pitch of 10 degrees and 20 degrees, the AD*Lite algorithm completes evacuation in a shorter time than the D*Lite algorithm. This proves the superiority of the AD*Lite algorithm in evacuation efficiency.

[0121] The evacuation simulation setting considering the congestion level is as follows:

[0122] The path planning algorithm that integrates both speed attenuation factor and congestion penalty factor is called ACD*Lite algorithm.

[0123] To verify the balancing effect of the ACD*Lite algorithm on congestion at each exit, we conducted an evacuation simulation test under a 10-degree left pitch condition and used the density map at 130 seconds for comparison.

[0124] like Fig. 9 As shown in Figure 1, the evacuation path planned by the AD*Lite algorithm causes severe congestion at the lower left exit, while the upper left exit remains relatively empty - an undesirable situation during evacuation. In contrast, the ACD*Lite algorithm solves this problem by updating the pheromone levels on the exit nodes in real time and performing replanning. Fig.10As shown in Figure 1, ACD*Lite effectively relieves congestion at the lower left exit by directing some pedestrians to the upper left exit. This helps balance the number of evacuees at each exit, improves the overall utilization of the exits, and improves evacuation efficiency.

[0125] The evacuation simulation setting that comprehensively considers the effects of speed decay, congestion, and weight shift is as follows:

[0126] Taking the above factors into consideration, the present invention proposes an enhanced algorithm based on the D*Lite algorithm, called the ACGD*Lite algorithm.

[0127] In order to evaluate the ability of the ACGD*Lite algorithm to improve evacuation safety, an evacuation simulation is performed with a dynamic change in the left trim angle. Assume that the hull begins to tilt at a constant speed due to damage or other factors. When the pedestrian load is more concentrated on the left side, it will generate an additional moment arm, accelerating the sinking of the hull. On the contrary, distributing more pedestrians on the right side helps to slow down the tilting speed of the hull. The tilting speed is defined as follows:

[0128]

[0129] exist Fig.11 In the experiment, the initial tilt angle of the ship was zero. Both the ACGD*Lite and ACD*Lite algorithms were able to complete the evacuation task within 400 seconds. Under the ACD*Lite algorithm, the tilt angle of the passenger ship reached 22 degrees, while under the ACGD*Lite algorithm, the tilt angle was reduced to 18 degrees, thereby improving the safety of the evacuation. In addition, the path planned by the ACGD*Lite algorithm helped to avoid a large number of pedestrians from concentrating on the low side of the hull, thereby slowing down the sinking speed of the hull. This resulted in faster movement of pedestrians at a smaller tilt angle, thereby improving the evacuation efficiency. The total evacuation time of the ACGD*Lite algorithm was 340 seconds, which was 42 seconds faster than the evacuation time of the ACD*Lite algorithm.

[0130] In another experiment, the hull tilted from 10 degrees, and the simulation results are as follows: Fig.12 As shown. Obviously, the ACD*Lite algorithm did not consider the impact of pedestrian movement on the ship's tilt, and the tilt angle reached 35 degrees at 440 seconds, when there were still 103 passengers who had not evacuated. The ACGD*Lite algorithm dynamically adjusted the evacuation path according to the distribution of pedestrian weight, thereby slowing down the sinking speed of the ship. The evacuation was completed within 693 seconds, and the maximum tilt angle was 32 degrees, which was lower than the safety threshold of 35 degrees. This shows that the evacuation path planned by the ACGD*Lite algorithm provides higher safety.

[0131] In summary, the method proposed in the present invention provides valuable support for effectively formulating evacuation strategies for passenger ships. In the event of an emergency, the evacuation strategy can be adjusted according to the tilt angle, exit congestion, and passenger distribution, which significantly improves the evacuation efficiency and safety in the passenger ship evacuation scenario.

[0132] In particular, in some preferred embodiments of the present invention, a computer device is also provided, including a memory and a processor and a computer program stored in the memory, and when the processor executes the computer program, the steps of the emergency evacuation path planning method considering the dynamic tilt of the passenger ship described in any of the above embodiments are implemented.

[0133] In some other preferred embodiments of the present invention, a computer-readable storage medium is provided, on which a computer program / instructions are stored. When the computer program is executed by a processor, the steps of the emergency evacuation path planning method considering the dynamic tilt of the passenger ship described in any of the above embodiments are implemented.

[0134] A person of ordinary skill in the art may understand that all or part of the processes in the above-mentioned embodiment method may be implemented by instructing related hardware through a computer program. The computer program may be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it may include the processes of the above-mentioned embodiment of the emergency evacuation path planning method considering the dynamic tilt of the passenger ship, which will not be repeated here.

[0135] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for emergency evacuation path planning considering dynamic tilt of a passenger ship, characterized in that: The following steps are involved: Step 1: Obtain the tilt angle of the passenger ship and the moving direction of the pedestrian, and build an inference model of the pedestrian speed attenuation factor; Step 2: Analyze and model the effect of pedestrian mass movement on the hull tilt; Step 3: Design an evacuation path planning algorithm based on the improved D*Lite algorithm; introduce congestion penalty factor, safety penalty coefficient and attenuation factor based on the path cost function of the D*Lite algorithm, and obtain the pre-planned path through the improved D*Lite algorithm according to the starting point of the leader and the exit point of the passenger ship; Step 4: Preset re-planning trigger conditions. When the preset trigger conditions are met, the current node is used as the new starting point, and steps 1 to 3 are repeated to re-plan the evacuation route, and evacuate according to the new evacuation route until the entire evacuation process is completed.

2. The method for planning an emergency evacuation path considering the dynamic tilt of a passenger ship according to claim 1, characterized in that: The step 1 comprises the following steps: Step 1: Substitute the tilt angle θ and the pedestrian movement direction with the tilt direction angle It is input into the fuzzy controller as an input variable; the speed attenuation factor δ is used as the output variable of the fuzzy controller; The fuzzy set of input variables is: Tilt angle θ = {S θ : Small tilt angle, M θ : Middle tilt angle, L θ : Large tilt angle}; S θ and L θ The membership function of M is the generalized Bell membership function. θ The membership function of selects Gaussian membership function; The angle between the pedestrian's movement direction and the tilt direction The membership functions of are all generalized Bell membership functions; Step 2: After fuzzification, a fuzzy control rule base is formulated based on if-then rules and 15 control rules are obtained and fuzzy reasoning is performed; the fuzzy set of output variables is: attenuation factor δ' = {VVS δ' : The attenuation factor is very small, VS δ' : The attenuation factor is small, S δ' The attenuation factor is small, M δ' The attenuation factor is moderate, L δ' Large attenuation factor, VL δ' The attenuation factor is large, VVL δ' The attenuation factor is large}; VVS δ' and VVL δ' The membership function of VS is trapezoidal membership function; δ' , S δ' , M δ' , L δ' , VL δ' The membership functions of all select the triangle membership function; Step 3: Defuzzify the attenuation factor δ' using the centroid defuzzification method and convert the control decision obtained by fuzzy reasoning into an accurate attenuation factor δ'.

3. The method for emergency evacuation path planning considering dynamic tilt of a passenger ship according to claim 1, characterized in that: The step 2 comprises the following steps: Step 1: Calculate the position G1 of the center of gravity of the ship after the pedestrian moves: Among them, w g is the mass of the object; d is the distance the object moves; w G is the draft of the ship; G is the center of gravity of the ship before the pedestrian moves; Step 2: Calculate the tilting force arm GZ' of the hull after the pedestrian moves: Among them, GZ is the forward tilt arm of the pedestrian; The area of ​​the GZ curve of the ship tilt arm before the pedestrian moves must be greater than or equal to 3.151m.deg when the ship tilt angle is 0° to 30°, and the area of ​​the GZ curve of the ship tilt arm before the pedestrian moves must be greater than or equal to 5.157m.deg when the ship tilt angle is 0° to 40°. The GZ curve of the ship tilt arm before the pedestrian moves after fitting is: GZ=0.44267θ-0.299356θ 3 4. The method for emergency evacuation path planning considering dynamic tilt of a passenger ship according to claim 1, characterized in that: The step 3 comprises the following steps: Step 1: When the g(v) value and rhs(v) value of node v are not equal, the state of node v is in a local inconsistent state. It needs to be included in the detection queue, and the priority of node v is determined according to the key value key(v); Among them, g(v) and rhs(v) are the total costs from node v to the target point; Δ m is a key value corrector, Δ m =0 is the initial value; the heuristic value h(v,v start ) represents the node v to the starting node v start The estimated cost of; c(v,v′) represents the actual cost from node v to its successor node v′; Step 2: Based on the distance cost, the calculation method of the modified cost is as follows, taking into account the speed reduction degree, congestion degree and safety degree: c(v,v')=(1+ρ)·(1+γ·ψ·sinθ)·dis(v,v') / δ Where dis(v,v′) is the distance from node v to the next node v′; (1+γ·ψ·sinθ) is the safety penalty coefficient; γ is 1 when the direction from node v to the next node v′ is downhill, and γ is 0 when the direction from node v to the next node v′ is downhill; ψ is the proportional factor, which is related to the number of hull layers, and the node of the kth layer takes the value of 2 k ;δ is the speed attenuation factor of the road section (v,v′); (1+ρ) is the congestion penalty factor; The congestion penalty factor (1+ρ) is calculated using the pheromone concentration in the ant colony algorithm, and the pheromone concentration represents the congestion level at the next node v′; p=a g ·p+(1-a g )·r l r l =a l ·r l +(1-a l )·e l Where ρ is the global pheromone concentration at the next node; a g is the global volatility factor; ρ l It is a local pheromone; a l is the local volatility factor; ε l is the amount of pheromone left behind by a pedestrian each time he moves.

5. The method for planning an emergency evacuation path considering the dynamic tilt of a passenger ship according to claim 1, characterized in that: In step 4, the triggering condition for re-planning is: if ΔGZ>1.5andθ≥15 ifVar(ρ e )>0.3 Where ΔGZ is the change of the ship tilting force arm caused by the movement of the ship's center of gravity when the pedestrian moves; Var(ρ e ) is the standard deviation of each outlet pheromone concentration; When the hull tilt arm change ΔGZ and the hull tilt angle θ satisfy if ΔGZ>1.5andθ≥15, it means that the evacuation process is in a potentially dangerous state due to the movement of pedestrians, and steps 1 to 3 should be repeated to replan the evacuation route; when Var(ρ e ) satisfies ifVar(ρ e )>0.3, it means that the pedestrian distribution at each exit is unbalanced and the exits are not fully utilized. Steps 1 to 3 should be repeated to replan the evacuation route.

6. The method for emergency evacuation path planning considering dynamic tilt of a passenger ship according to claim 1, characterized in that: In step 3, the leader is a security officer on the passenger ship, and the security officer guides the surrounding pedestrians to move.

7. A computer device / equipment / system comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer program product comprising a computer program / instructions, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images