A method and apparatus for selecting maritime search and rescue resources

Abstract

The application provides a maritime search and rescue resource selection method and device, the method comprising: determining a first characteristic parameter of each search and rescue resource, the search and rescue resource at least comprising an airplane and a ship, the first characteristic parameter being used for indicating a search and rescue capability of the search and rescue resource; determining a second characteristic parameter of a search and rescue task; determining a response probability of each search and rescue resource for executing the search and rescue task based on the first characteristic parameter and the second characteristic parameter, the response probability representing a probability of the search and rescue resource for executing the search and rescue task; and determining a target search and rescue resource for executing the search and rescue task from a plurality of search and rescue resources based on the response probability and an ant colony algorithm. The maritime search and rescue resource selection method can efficiently and accurately match and determine a search and rescue resource for executing a search and rescue task.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of computer technology, and in particular to a method and apparatus for selecting maritime search and rescue resources. Background Technology

[0002] Since the selection of maritime search and rescue resources is mostly based on human decision-making, and when information is multi-sourced and complex, and decision-makers are unable to make rapid and scientific decisions under emergency psychological conditions, it is essential to use intelligent optimization algorithms to assist decision-makers in making quick and scientific decisions. Considering resource needs to better suit the selection of maritime search and rescue resources has significant theoretical value and practical guiding significance for improving maritime search and rescue capabilities.

[0003] Currently, domestic and international research has been conducted on maritime search and rescue methods. A representative example is the "International Handbook of Air and Maritime Search and Rescue," jointly published in 1998 by the International Maritime Organization (IMO) and the International Civil Aviation Organization (ICAO). This handbook is a summary of Parts I, II, and III of "The Theory of Search," published by Koopman and his colleagues' Anti-Submarine Military Operations Research Group. It utilizes classification analysis, modern decision theory, and fuzzy mathematics evaluation methods to study optimal models and methods for rescue vessels; it also investigates how to calculate the drift of distressed targets at sea due to wind and current, determine search areas and ranges, and determine the optimal search mode.

[0004] The Ant Colony Algorithm (ACA) is a swarm intelligence algorithm proposed by Italian scholar Dorigo M. et al. in the 1990s, inspired by the foraging behavior of ants. It possesses strong robustness and is easily combined with other optimization algorithms. With its development, the ACA has been applied to various fields such as job scheduling and path planning. However, the standard ACA does not adequately consider the resource requirements of maritime search and rescue. While it can solve multi-objective optimization models, it suffers from slow convergence, susceptibility to local optima, and excessively long search times. Summary of the Invention

[0005] This invention provides a method and apparatus for selecting maritime search and rescue resources that can efficiently and accurately match and determine search and rescue resources for carrying out search and rescue missions.

[0006] To address the aforementioned technical problems, embodiments of the present invention provide a method for selecting maritime search and rescue resources, comprising: A first characteristic parameter is determined for each search and rescue resource, which includes at least aircraft and ships, and the first characteristic parameter is used to indicate the search and rescue capability of the search and rescue resource; Determine the second characteristic parameter of the search and rescue mission; The response probability of each search and rescue resource executing the search and rescue mission is determined based on the first feature parameter and the second feature parameter, wherein the response probability characterizes the probability that the search and rescue resource executes the search and rescue mission; Based on the response probability and the ant colony algorithm, the target search and rescue resource for performing the search and rescue mission is determined from a plurality of search and rescue resources.

[0007] As an optional embodiment, it also includes: Establish a response threshold model; Determining the response probability of each search and rescue resource executing the search and rescue mission based on the first feature parameter and the second feature parameter includes: The first feature parameter and the second feature parameter are processed based on the response threshold model to obtain the response probability of each search and rescue resource executing the search and rescue mission. As an optional embodiment, establishing the response threshold model includes: A first response function is established based on historical first characteristic parameters to determine the stimulus quantity of the search and rescue resource, wherein the stimulus quantity characterizes the ability of the search and rescue resource to perform the search and rescue mission; Determine the degree of cooperation among different search and rescue resources when performing search and rescue missions based on historical data; A second response function is established based on the cooperation degree to determine the response threshold of the search and rescue mission to the search and rescue resources; The response threshold model is established based on the first response function and the second response function.

[0008] As an optional embodiment, the step of processing the first feature parameter and the second feature parameter based on the response threshold model to obtain the response probability of each search and rescue resource executing the search and rescue mission includes: Based on the response threshold, the first feature parameter and the second feature parameter are processed to determine the stimulus amount and response threshold of the search and rescue resource; The response probability is calculated and determined based on the stimulus amount and the response threshold. The greater the stimulus intensity, the smaller the response threshold and the greater the response probability.

[0009] As an optional embodiment, the target search and rescue resource includes multiple search and rescue resources, and the method further includes: Determine the level of coordination among target search and rescue resources; Determine the execution result of the search and rescue mission by the target search and rescue resources; The parameters of the response threshold model are adjusted based on the cooperation degree and the execution result. The parameters include the response threshold, and the adjustment method includes lowering or raising the response threshold.

[0010] As an optional embodiment, the step of determining the target search and rescue resource for performing the search and rescue mission from a plurality of search and rescue resources based on the response probability and ant colony algorithm includes: The state transition probability for selecting each search and rescue resource is determined based on the roulette wheel method in the ant colony algorithm. The target search and rescue resource for performing the search and rescue mission is determined from a plurality of search and rescue resources based on the response probability and state transition probability.

[0011] Another embodiment of the present invention also provides a maritime search and rescue resource selection device, comprising: The first determining module is used to determine a first characteristic parameter for each search and rescue resource, wherein the search and rescue resources include at least aircraft and ships, and the first characteristic parameter is used to indicate the search and rescue capability of the search and rescue resource. The second determining module is used to determine the second characteristic parameters of the search and rescue mission; The calculation module is used to determine the response probability of each of the search and rescue resources performing the search and rescue mission based on the first feature parameter and the second feature parameter, wherein the response probability represents the probability that the search and rescue resource performs the search and rescue mission; The selection module is used to calculate and determine the target search and rescue resource for performing the search and rescue mission from a plurality of search and rescue resources based on the response probability and the ant colony algorithm.

[0012] As an optional embodiment, it also includes: Establish a module for building the response threshold model; Determining the response probability of each search and rescue resource executing the search and rescue mission based on the first feature parameter and the second feature parameter includes: The first feature parameter and the second feature parameter are processed based on the response threshold model to obtain the response probability of each search and rescue resource executing the search and rescue mission. As an optional embodiment, establishing the response threshold model includes: A first response function is established based on historical first characteristic parameters to determine the stimulus quantity of the search and rescue resource, wherein the stimulus quantity characterizes the ability of the search and rescue resource to perform the search and rescue mission; Determine the degree of cooperation among different search and rescue resources when performing search and rescue missions based on historical data; A second response function is established based on the cooperation degree to determine the response threshold of the search and rescue mission to the search and rescue resources; The response threshold model is established based on the first response function and the second response function.

[0013] As an optional embodiment, the step of processing the first feature parameter and the second feature parameter based on the response threshold model to obtain the response probability of each search and rescue resource executing the search and rescue mission includes: Based on the response threshold, the first feature parameter and the second feature parameter are processed to determine the stimulus amount and response threshold of the search and rescue resource; The response probability is calculated and determined based on the stimulus amount and the response threshold. The greater the stimulus intensity, the smaller the response threshold and the greater the response probability.

[0014] Based on the disclosure of the above embodiments, it can be understood that the beneficial effects of the embodiments of the present invention include the ability to effectively solve the problems of slow convergence speed and easy getting trapped in local optima by combining the response threshold model used to solve the response probability with the ant colony algorithm. At the same time, since the characteristics of various search and rescue resources and search and rescue tasks are combined, it is possible to select better and more suitable resources in the process of selecting search and rescue resources to perform search and rescue tasks, which significantly improves the solution speed of resource selection and avoids the omission of better resources.

[0015] Other features and advantages of the invention will be set forth in the following description. The objects and other advantages of the invention can be realized and obtained by means of the structures particularly pointed out in the written description and drawings.

[0016] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart of the method for selecting maritime search and rescue resources in an embodiment of the present invention.

[0018] Figure 2 This is a flowchart of the model algorithm in an embodiment of the present invention.

[0019] Figure 3 This is a flowchart of a method for selecting maritime search and rescue resources according to another embodiment of the present invention.

[0020] Figure 4 This is a structural block diagram of the maritime search and rescue resource selection device in an embodiment of the present invention. Detailed Implementation

[0021] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings, but these are not intended to limit the scope of the invention.

[0022] It should be understood that various modifications can be made to the embodiments disclosed herein. Therefore, the following description should not be considered as limiting, but merely as an example of embodiments. Other modifications within the scope and spirit of this disclosure will be apparent to those skilled in the art.

[0023] The accompanying drawings, which are included in and form part of this specification, illustrate embodiments of the present disclosure and, together with the general description of the disclosure given above and the detailed description of the embodiments given below, serve to explain the principles of the disclosure.

[0024] These and other features of the invention will become apparent from the following description of preferred forms of embodiments given as non-limiting examples, with reference to the accompanying drawings.

[0025] It should also be understood that although the invention has been described with reference to some specific examples, those skilled in the art can certainly implement many other equivalent forms of the invention.

[0026] The above and other aspects, features and advantages of this disclosure will become more apparent when taken in conjunction with the accompanying drawings and in view of the following detailed description.

[0027] Specific embodiments of the present disclosure are described thereafter with reference to the accompanying drawings; however, it should be understood that the disclosed embodiments are merely examples of the present disclosure and can be implemented in various ways. Well-known and / or repeated functions and structures are not described in detail to avoid unnecessary or redundant details that could obscure the present disclosure. Therefore, the specific structural and functional details disclosed herein are not intended to be limiting, but are merely representative of the present disclosure and are intended to teach those skilled in the art to use the present disclosure in a variety of substantially any suitable detailed structures.

[0028] This specification may use the phrases “in one embodiment,” “in another embodiment,” “in yet another embodiment,” or “in still another embodiment,” all of which may refer to one or more of the same or different embodiments according to this disclosure.

[0029] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0030] like Figure 1 As shown, this embodiment of the invention provides a method for selecting maritime search and rescue resources, including: Determine the first characteristic parameter for each search and rescue resource, which includes at least aircraft and ships. The first characteristic parameter is used to indicate the search and rescue capability of the search and rescue resource. Determine the second characteristic parameter of the search and rescue mission; The response probability of each search and rescue resource executing a search and rescue mission is determined based on the first feature parameter and the second feature parameter. The response probability represents the probability that the search and rescue resource will execute a search and rescue mission. The target search and rescue resource for performing the search and rescue mission is determined from multiple search and rescue resources based on response probability and ant colony algorithm.

[0031] For example, this embodiment is carried out under the following assumptions: ① All the people in distress were evenly distributed within the search area; ②This embodiment only considers the scenarios of aircraft search and ship rescue; ③ Assume that the time it takes for all search and rescue aircraft to cover the designated search area is the time of the search phase; ④ Assume that the performance information of various types of aircraft under different sea conditions is already known.

[0032] Maritime search and rescue operations are typically conducted under highly complex conditions, influenced by numerous factors, thus requiring consideration of a wide range of model input conditions. This embodiment addresses the problem under the premise that the basic information that should be known at the time of a maritime disaster is already determined. This basic information includes: information related to the distressed persons; information related to the marine environment at the time of the search and rescue; the impact of the environment on the search and rescue operation; and the potential dangers faced by the search and rescue personnel.

[0033] Furthermore, when executing the method of this embodiment, a first characteristic parameter is first obtained for each search and rescue resource. This search and rescue resource includes at least aircraft and ships, and may also include other search and rescue resources; the specific type is not unique. The first characteristic parameter indicates the search and rescue capability of the resource, such as the maximum passenger capacity of a ship, the average salvage time of a ship, search and rescue sea state capability, aircraft search capability, aircraft detection probability, etc. Each parameter reflects the performance of different resources. Next, a second characteristic parameter for the search and rescue mission is determined. This second characteristic parameter indicates the type of search and rescue mission, the difficulty of the search and rescue, the sea area where the mission is located, the number of people to be rescued, etc. For example, refer to the following table:

[0034] Once the first and second feature parameters are obtained, the system can determine the response probability of each search and rescue resource in executing the search and rescue mission, which is equivalent to determining the probability of each search and rescue resource being selected to execute the mission. This is similar to determining the matching degree; if the first and second feature parameters match, the corresponding search and rescue resource has a higher response probability, and vice versa. After determining the response probabilities of all search and rescue resources, the system can use both response probabilities and the ant colony algorithm to calculate and determine the target search and rescue resource for executing the mission. This target search and rescue resource can include not only one resource but also multiple resources, such as one, two, or three ships, or even a ship and an aircraft forming the target search and rescue resource. The specific combination and quantity are variable.

[0035] Based on the disclosure of the above embodiments, it can be understood that the beneficial effects of this embodiment include the ability to effectively solve the problems of slow convergence speed and easy getting trapped in local optima by combining response probability and ant colony algorithm. At the same time, since the characteristics of various search and rescue resources and search and rescue tasks are combined, it is possible to select better and more suitable resources in the process of selecting search and rescue resources to perform search and rescue tasks, which significantly improves the solution speed of resource selection and avoids the omission of better resources.

[0036] Furthermore, the maritime search and rescue resource selection scheme in this embodiment includes a description of the types and quantities of resources. Therefore, the mathematical expression of the maritime search and rescue resource scheme is: (1) Where, x i This represents the quantity of resource i in the scheme. If x... i A value of 0 indicates that resource i in the maritime search and rescue plan did not participate.

[0037] In this embodiment, the method needs to achieve multiple objectives during the selection and formulation of the corresponding calculation function. For example, objective 1 is to minimize the search time: the total search time refers to the maximum duration required for various resources to work to cover the area to be searched. To calculate the total search time T_s, relevant information needs to be explained. The total search time includes the time it takes for resources to arrive at the search area and the time spent searching for resources. For aircraft, the time spent on the road needs to be considered. Therefore, the time spent on the road by the i-th aircraft is t. i : (2) The actual search time for the i-th aircraft within the total search time is: (3) Since the sum of the areas searched by all participating search and rescue aircraft should be equal to the total area to be searched, we have: (4) (5) Maximizing average resource utility: Average resource utility refers to the average utility generated by resources involved in search and rescue. The utility generated by resources can be measured by the probability of success (POS).

[0038] Since the search area is known in this embodiment, it is assumed that all the people in distress are within the search area. POC =1 The probability of discovery refers to the probability that a person in distress is found within the search area of ​​a search subject (in this example, the search subject is an aircraft).

[0039] (6) According to formula (6), the average utility of the resource can be obtained as follows: (7) This embodiment transforms the multi-objective problem involving the two objectives mentioned above into a single-objective problem, resulting in the following model: (8) in and The targets are respectively and target The weights, in later examples, are determined by... and Parameters are adjusted to optimize the objective of the single-objective problem.

[0040] Furthermore, due to limitations in real-world environmental conditions and the inherent properties of resources, there are certain constraints in the process of optimizing the selection of maritime search and rescue resources. The following are the constraints summarized in this embodiment based on actual circumstances.

[0041] (1) Resource quantity constraints There is an upper limit to the number of available resources. Therefore, in the solution, x i The following constraints should be met: 0≤x i ≤n i ,n i Let i be the number of available resources, i=1,2,3. M (9) (2) Maximum sea state constraints for safe navigation of resources To ensure the safety of the search and rescue resources themselves, it is necessary to ensure that the resources carrying out the search and rescue mission operate at sea within the maximum permissible sea state constraints. Therefore: B≥B,B i ∈{1,2, ,9},i=1,2, M (3) Maximum passenger capacity of ships To ensure safe navigation, different ships have different numbers of passengers they can accommodate.

[0042] N≤∑c i i∈The set of resources participating in the search and rescue (4) The time for search and rescue resources to reach the designated search area shall not exceed the total search and rescue time constraint. Each search and rescue resource must arrive at the scene before the search mission is completed in order to have a chance to participate in the search operation; therefore, there are: t j <T s and t k <T r (5) Survival requirement for the rescued person A positive POL (Purpose Level of Survival) indicates that the person in distress is still alive when rescued. The POL represents the survival rate of the person in distress upon successful rescue and is directly related to search and rescue time and sea conditions. Therefore, this embodiment uses search and rescue time and the longest waiting time for the target to be rescued to measure the POL. The expression for POL is as follows:

[0043] Furthermore, the method in this embodiment also includes: Establish a response threshold model; The response probability of each search and rescue resource to perform a search and rescue mission is determined based on the first feature parameter and the second feature parameter, including: The response threshold model is used to process the first and second feature parameters to obtain the response probability of each search and rescue resource executing a search and rescue mission. Among them, such as Figure 3 As shown, a response threshold model is established, including: A first response function is established based on the historical first characteristic parameter to determine the stimulus quantity of search and rescue resources. The stimulus quantity characterizes the ability of search and rescue resources to perform search and rescue missions. Determine the degree of cooperation among different search and rescue resources when performing search and rescue missions based on historical data; A second response function is established based on the degree of cooperation to determine the response threshold of the search and rescue mission to search and rescue resources; A response threshold model is established based on the first response function and the second response function.

[0044] Furthermore, the response threshold model is used to process the first and second feature parameters to obtain the response probability of each search and rescue resource executing a search and rescue mission, including: The stimulus amount and response threshold of search and rescue resources are determined based on the first feature parameter and the second feature parameter processed by the response threshold. The response probability is determined based on the stimulus amount and the response threshold. The greater the stimulus intensity, the smaller the response threshold, and the greater the response probability.

[0045] For example, response threshold models are closely related to two factors: one is the stimulus intensity *s* associated with a specific task, which can be represented as the number of encounters, the concentration of a chemical substance, or other quantifiable cues that the individual can perceive in the task. The other is the response threshold *θ*, which describes the individual's internal tendency to perform the task; the smaller the threshold, the stronger the tendency. Specifically, when *s*... The response probability is low at θ, and low at s The response probability is high at θ. A family of response functions that satisfy the above requirements. Defined as: (14) Here, n is a constant that determines the slope of the response threshold curve.

[0046] By incorporating resource performance information into the ant colony algorithm through the stimulus and threshold values ​​in the response threshold model, the convergence speed of the ant colony algorithm can be accelerated based on the actual information of the problem, avoiding getting trapped in local optima.

[0047] In this embodiment, when using an ant colony algorithm based on a response threshold model to solve the model, each resource performing the task is modeled as an intelligent individual. The response threshold model determines the probability of selecting a resource to execute the task. Based on resource performance, the more suitable a resource is for performing a task, the more likely it is to be selected, meaning a higher probability of responding to the task. The resource response probability obtained from the response threshold model can replace the heuristic information in the ant colony algorithm. The pheromone and threshold values ​​are updated based on the quality of the solutions (solutions) obtained by the ant colony algorithm, further updating the response probability. The better the solution, the higher the pheromone concentration and response probability of the selected resource. In the next iteration, the better resource is given priority. This allows resource performance and actual environmental requirements to be considered in the ant colony algorithm, leading to a faster and better search for the optimal solution.

[0048] The probability of a resource responding to a task is determined by two response functions in the response threshold model. Therefore, two response functions need to be constructed in the response threshold model. One function determines the stimulus quantity of the resource; for resource aircraft, this response function should be modeled as a multivariate function of factors such as search capability, probability of target discovery, time to reach the search area, and sea state; for resource vessels, it should be modeled as a multivariate function of maximum manpower, average salvage time, time to reach the search area, and sea state. The other response function determines the task's response threshold to the resource; this function should be modeled as a recursive function based on the resource's suitability in the best solution during the iteration process.

[0049] like Figure 2 As shown, to incorporate the aforementioned factors reflecting resource performance into the ant colony algorithm and to select superior resources for maritime search and rescue operations, this embodiment uses a response threshold model to determine the initial response probability of resources to maritime search and rescue missions. In other words, resource performance determines the initial probability of a resource participating in the search and rescue operation. During the iterative calculation of the model, the better the resource performance and the lower its corresponding task response threshold, the greater its probability of participating in the task, i.e., the greater its response probability, and the greater the probability of a superior resource being selected. Therefore, using the response threshold model can better meet resource requirements and select superior resources.

[0050] Employing a response threshold model requires determining the stimulus quantity *s* of the resource and the response threshold *θ* of the resource to the task. Based on the preceding modeling analysis of maritime search and rescue and the introduction of the response threshold model, two functional relationships (models) need to be determined.

[0051] Specifically, the stimulus quantity model reflects the resource's ability to perform tasks. The larger the stimulus quantity, the better the resource's ability to perform tasks. As shown in Equation (14), with a fixed response threshold, the larger the stimulus quantity, the greater the probability of the resource responding to the task. Therefore, the higher the probability of a better resource being selected, which meets the requirements for selecting marine resources.

[0052] Different resources (aircraft, ships) require different performance parameters due to the different tasks they perform, and therefore the corresponding models for calculating their stimulus quantities are also different.

[0053] For ships, better resource performance is indicated by a larger maximum passenger capacity, shorter average salvage time, higher search and rescue sea state, and shorter time to reach the search area. The corresponding stimulus quantity is s. j The larger the value of the ship resource j, the greater the stimulus to the rescue mission, and the better the rescue mission can be completed. It should have a higher probability of participating in the rescue operation. The stimulus model is shown in equation (15). (15) For aircraft, better performance is indicated by stronger search capabilities, a higher probability of target detection, ability to navigate high sea states, and shorter time to reach the search area. j The larger the value of the aircraft resource j, the greater the stimulus to the search mission, the better it can complete the search mission and the higher its probability of participating in the search operation; its stimulus model is shown in equation (16): (16) Furthermore, the target search and rescue resources include multiple search and rescue resources, and the methods also include: Determine the level of coordination among target search and rescue resources; Determine the outcome of the search and rescue mission by identifying the target search and rescue resources; The parameters of the response threshold model are adjusted based on the degree of cooperation and the execution results. The parameters include the response threshold, and the adjustment methods include lowering or raising the response threshold.

[0054] Specifically, the resource matching degree (i.e., the matching degree mentioned above) refers to the minimum capability requirement or threshold of the task for resources. As shown in equation (14), under the condition of a certain stimulus, the smaller the threshold, that is, the lower the task threshold, the greater the probability of the resource responding to the task. Therefore, the higher the probability of the resource that is better at performing the task being selected, which meets the requirements of marine resource selection.

[0055] Cooperation of maritime search and rescue resources This represents the probability that resource j also participates when resource i participates in the search and rescue mission. During the iteration process, the more times resource i or resource j appears in the optimal solution, the higher the probability of resource j participating. The smaller the value, the more resources it represents. or resources The lower the response threshold for a task, the better. Because of resources... or resources A high frequency of occurrence in the optimal solution indicates abundant resources. and resources Good coordination, and resources or resources If a person can better complete a search and rescue mission, then the threshold for participating in the search and rescue mission should be lower, and the corresponding search and rescue mission threshold should also be lower, resulting in a higher probability of participation. That is, under the group reward and punishment mechanism, the threshold will decrease when an individual successfully completes a mission, and increase when the individual fails to complete a mission. Representing the threshold by resource coordination can avoid the situation where the solution gets trapped in a local optimum due to the prominence of a certain resource. The threshold model is shown in equation (17): (17) in Indicates the first g Resources in the -1 generation optimal solution With resources Number of times it appears; Indicates the first g The number of resources in the -1 generation optimal solution; It is an adjustment parameter, controlling The rate of change.

[0056] Furthermore, the Ant Colony Optimization (ACO) algorithm works by simulating the behavior of ants in the process of searching for food: when ants are searching for food, they release a pheromone that is unique to ants along their paths, which other ants can detect and influence their behavior within a certain range (tending to choose the path with the highest pheromone concentration). The more ants that walk on a certain path, the more pheromone is left on that path, and the higher the pheromone concentration, thus increasing the probability that subsequent ants will choose this path. In the process of continuous repetition, the path with the highest pheromone concentration is eventually the optimal foraging path.

[0057] Maritime search and rescue, especially major maritime search and rescue operations, involves a large variety and quantity of resources. Resource selection in maritime search and rescue is an NP-hard problem, and the timeliness requirements are very high. It is difficult to quickly formulate an efficient resource selection plan using ordinary methods such as manual methods. Therefore, this embodiment adopts an ant colony algorithm based on a response threshold model. Besides making intelligent optimization algorithms more suitable for the actual needs of solving maritime search and rescue resource problems, it also overcomes the shortcomings of ordinary ant colony algorithms, such as slow convergence and susceptibility to getting trapped in local optima.

[0058] Ant colony optimization (ACO) algorithms draw on and incorporate the foraging behavior characteristics of real ant colonies in nature. As a distributed, multi-intelligence system, it possesses strong robustness. However, when the solution space has a complex, multi-modal shape, it is prone to getting trapped in local maxima and exhibiting premature convergence, thus failing to generate high-quality marine resource selection solutions.

[0059] As mentioned above, in order to solve the problems of the ant colony algorithm, this embodiment combines the response threshold model with the ant colony algorithm. The relevant formulas and some parameters after the combination are shown below. The state transition probability of ant k from node i to node j at time t is shown in (18): (18) in The response model serves as heuristic information in the ant colony algorithm. For pheromones, This is the pheromone adjustment coefficient. This is the adjustment coefficient for heuristic information.

[0060] The pheromone left by the k-th ant on the path (i,j) is shown in equation (19): (19) The total amount of pheromone left by m ants on the path (i,j) is shown in Equation (20): (20) in This represents the pheromone increment along path (i,j) in the current iteration, at the initial time. =0. This represents the amount of information left by the k-th ant on the path (i,j) in this iteration.

[0061] After all the ants have completed one cycle, the remaining pheromones on the path need to be updated, i.e., the pheromones evaporate. The pheromones on each path (i,j) are adjusted according to formula (21): (twenty one) in It is the pheromone evaporation coefficient.

[0062] Furthermore, in this embodiment, the target search and rescue resource for performing the search and rescue mission is determined from multiple search and rescue resources based on response probability and ant colony algorithm, including: The state transition probability for selecting each search and rescue resource is determined based on the roulette wheel method in the ant colony algorithm. The target search and rescue resource for performing the search and rescue mission is determined from multiple search and rescue resources based on response probability and state transition probability.

[0063] For example, in ant colony optimization, there are multiple ways for ants to choose the next resource. This embodiment chooses the classic roulette wheel method. Specifically, the probability of an ant choosing the next resource uses the state transition probability described above. Although we obtain the probability of choosing each resource (state transition probability), due to the diversity of ant colony optimization, it is necessary to ensure that other resources also have a chance of being selected. Otherwise, if only resources with high probabilities are selected, it would become a greedy algorithm. Therefore, the roulette wheel method is chosen. The specific operation is as follows: obtain the probability value of each resource, sum these probabilities to obtain the total probability value, then randomly generate a random number between 0 and the total probability value, and then subtract the probability value of each resource from it until a negative number is obtained to determine the next resource to choose. Repeat the same operation until the loop termination condition is met (the number of people found in the search phase has been rescued).

[0064] For a maritime search and rescue resource selection scheme, all resources can be accessed exactly once. Therefore, accessed resources need to be stored in a taboo list to avoid repeated access. Among all accessible resources, the performance of each resource must also be considered. The probability of state transition is used to determine whether a resource can be added to the taboo list. The resources in the final taboo list are the resources that will ultimately participate in the mission. In other words, the taboo list is the obtained maritime search and rescue resource selection scheme.

[0065] To better understand the generation of maritime search and rescue resource selection schemes, the table below shows the basic theoretical names of the resource selection methods and their mapping relationships with maritime search and rescue resources.

[0066]

[0067] The following description is based on specific embodiments: An accident occurred in a certain sea area. A merchant ship carrying 70 people ran aground, lost power, and was in danger of capsizing. The crew abandoned ship and actively attempted self-rescue, deploying life-saving equipment. However, due to the rough sea conditions, the crew members were floating over a wide area with complex injuries, requiring urgent external rescue. The nearest professional search and rescue center and naval search and rescue center are 90 and 120 nautical miles away, respectively. There are 5 available fishing and merchant ships in the vicinity. The professional search and rescue center has some specialized rescue vessels and aircraft conducting routine patrols, while the naval search and rescue center has limited equipment under repair and is also unavailable. The area to be searched is measured at 800 nautical miles (n mile²), and the current sea state is 4. The maximum waiting time for people in sea state 4 is 5 hours. The performance information of the professional search and rescue forces is shown in the table.

[0068] The performance information of the military's search and rescue forces is shown in the table.

[0069] The performance information of the search and rescue capabilities of passing vessels is shown in the table.

[0070] The distance between professional search and rescue forces and the accident site is 90 (n mile), while the distance between military search and rescue forces and the accident site is 120 (n mile).

[0071] This embodiment uses an ant colony algorithm based on a response threshold model, so the experimental parameters of the basic ant colony algorithm and the parameters of the threshold model will be used in this experiment. Substituting the above information into the ant colony algorithm based on the response threshold model yields the following table showing the selection scheme for maritime search and rescue resources:

[0072] Substituting the information into the basic ant colony algorithm yields the resource selection schemes shown in the table below.

[0073]

[0074] The comparison results show that the basic ant colony algorithm reaches its optimal value around 120 generations, while the ant colony algorithm based on the response threshold model reaches a stable state around 55 generations, quickly converging to the ideal value. Furthermore, the target value of 1.631 obtained by the algorithm in this embodiment is better than the target value of 1.636 obtained by the basic ant colony algorithm. Additionally, based on the comparison chart of the iterative average resource utility of the algorithm in this embodiment and the basic ant colony algorithm, the average resource utility value of 0.0931 obtained by the algorithm in this embodiment is better than the average resource utility value of 0.0772 obtained by the basic ant colony algorithm. Furthermore, based on the comparison chart of iterative search time and search success rate of the algorithm in this embodiment and the basic ant colony algorithm, although the search time of 1.512h obtained by the algorithm in this embodiment is longer than the search time of 1.041h obtained by the basic ant colony algorithm, the search success rate of 0.931 obtained by the algorithm in this embodiment is higher than the search success rate of 0.927 obtained by the basic ant colony algorithm. Moreover, the search time obtained by the algorithm in this embodiment is much shorter than the longest waiting time of 5h for humans in sea state 4. Therefore, the above comparative analysis shows that the ant colony algorithm based on the response threshold model converges faster and is closer to the ideal solution than the basic ant colony algorithm. Under the condition that the search time meets the requirements of maritime search and rescue, the ant colony algorithm based on the response threshold model improves the average resource utility by 20.6% compared with the basic ant colony algorithm.

[0075] like Figure 4 As shown, another embodiment of the present invention also provides a maritime search and rescue resource selection device, comprising: The first determining module is used to determine a first characteristic parameter for each search and rescue resource, wherein the search and rescue resources include at least aircraft and ships, and the first characteristic parameter is used to indicate the search and rescue capability of the search and rescue resource. The second determining module is used to determine the second characteristic parameters of the search and rescue mission; The calculation module is used to determine the response probability of each of the search and rescue resources performing the search and rescue mission based on the first feature parameter and the second feature parameter, wherein the response probability represents the probability that the search and rescue resource performs the search and rescue mission; The selection module is used to calculate and determine the target search and rescue resource for performing the search and rescue mission from a plurality of search and rescue resources based on the response probability and the ant colony algorithm.

[0076] As an optional embodiment, it also includes: Establish a module for building the response threshold model; Determining the response probability of each search and rescue resource executing the search and rescue mission based on the first feature parameter and the second feature parameter includes: The first feature parameter and the second feature parameter are processed based on the response threshold model to obtain the response probability of each search and rescue resource executing the search and rescue mission. As an optional embodiment, establishing the response threshold model includes: A first response function is established based on historical first characteristic parameters to determine the stimulus quantity of the search and rescue resource, wherein the stimulus quantity characterizes the ability of the search and rescue resource to perform the search and rescue mission; Determine the degree of cooperation among different search and rescue resources when performing search and rescue missions based on historical data; A second response function is established based on the cooperation degree to determine the response threshold of the search and rescue mission to the search and rescue resources; The response threshold model is established based on the first response function and the second response function.

[0077] As an optional embodiment, the step of processing the first feature parameter and the second feature parameter based on the response threshold model to obtain the response probability of each search and rescue resource executing the search and rescue mission includes: Based on the response threshold, the first feature parameter and the second feature parameter are processed to determine the stimulus amount and response threshold of the search and rescue resource; The response probability is calculated and determined based on the stimulus amount and the response threshold. The greater the stimulus intensity, the smaller the response threshold and the greater the response probability.

[0078] As an optional embodiment, the target search and rescue resource includes multiple search and rescue resources, and the method further includes: Determine the level of coordination among target search and rescue resources; Determine the execution result of the search and rescue mission by the target search and rescue resources; The parameters of the response threshold model are adjusted based on the cooperation degree and the execution result. The parameters include the response threshold, and the adjustment method includes lowering or raising the response threshold.

[0079] As an optional embodiment, the step of determining the target search and rescue resource for performing the search and rescue mission from a plurality of search and rescue resources based on the response probability and ant colony algorithm includes: The state transition probability for selecting each search and rescue resource is determined based on the roulette wheel method in the ant colony algorithm. The target search and rescue resource for performing the search and rescue mission is determined from a plurality of search and rescue resources based on the response probability and state transition probability.

[0080] Another embodiment of the present invention also provides an electronic device, comprising: One or more processors; Memory, configured to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors shall implement the methods described above.

[0081] An embodiment of the present invention also provides a storage medium storing a computer program thereon, which, when executed by a processor, implements the method described above. It should be understood that the various solutions in this embodiment have the corresponding technical effects in the above method embodiments, and will not be repeated here.

[0082] Another embodiment of the present invention also provides a computer program product tangibly stored on a computer-readable medium and including computer-readable instructions, which, when executed, cause at least one processor to perform methods such as those described in the embodiments above. It should be understood that the various solutions in this embodiment have the corresponding technical effects in the above method embodiments, and will not be repeated here.

[0083] It should be noted that the computer storage medium in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access storage medium (RAM), a read-only storage medium (ROM), an erasable programmable read-only storage medium (EPROM or flash memory), an optical fiber, a portable compact disk read-only storage medium (CD-ROM), an optical storage medium, a magnetic storage medium, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program configured for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, antenna, optical fiber, RF, etc., or any suitable combination thereof.

[0084] It should be understood that although this application is described according to various embodiments, not every embodiment contains only one independent technical solution. This way of describing the specification is only for clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

[0085] The above embodiments are merely exemplary embodiments of the present invention and are not intended to limit the present invention. The scope of protection of the present invention is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to the present invention within its spirit and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of the present invention.

Claims

1. A method for selecting maritime search and rescue resources, characterized in that, include: A first characteristic parameter is determined for each search and rescue resource, which includes at least aircraft and ships, and the first characteristic parameter is used to indicate the search and rescue capability of the search and rescue resource; Determine the second characteristic parameter of the search and rescue mission; The response probability of each search and rescue resource executing the search and rescue mission is determined based on the first feature parameter and the second feature parameter, wherein the response probability characterizes the probability that the search and rescue resource executes the search and rescue mission; Based on the response probability and the ant colony algorithm, the target search and rescue resource for performing the search and rescue mission is determined from a plurality of search and rescue resources. Also includes: Establish a response threshold model; Determining the response probability of each search and rescue resource executing the search and rescue mission based on the first feature parameter and the second feature parameter includes: The first feature parameter and the second feature parameter are processed based on the response threshold model to obtain the response probability of each search and rescue resource executing the search and rescue task; The step of processing the first feature parameter and the second feature parameter based on the response threshold model to obtain the response probability of each search and rescue resource executing the search and rescue mission includes: Based on the response threshold model, the first feature parameter and the second feature parameter are processed to determine the stimulus amount and response threshold of the search and rescue resources; The response probability is calculated and determined based on the stimulus amount and the response threshold. The greater the stimulus intensity, the smaller the response threshold and the greater the response probability.

2. The method according to claim 1, characterized in that, The establishment of the response threshold model includes: A first response function is established based on historical first characteristic parameters to determine the stimulus quantity of the search and rescue resource, wherein the stimulus quantity characterizes the ability of the search and rescue resource to perform the search and rescue mission; Determine the degree of cooperation among different search and rescue resources when performing search and rescue missions based on historical data; A second response function is established based on the cooperation degree to determine the response threshold of the search and rescue mission to the search and rescue resources; The response threshold model is established based on the first response function and the second response function.

3. The method according to claim 1, characterized in that, The target search and rescue resources include multiple search and rescue resources, and the method further includes: Determine the level of coordination among target search and rescue resources; Determine the execution result of the search and rescue mission by the target search and rescue resources; The parameters of the response threshold model are adjusted based on the cooperation degree and the execution result. The parameters include the response threshold, and the adjustment method includes lowering or raising the response threshold.

4. The method according to claim 1, characterized in that, The step of determining the target search and rescue resource for performing the search and rescue mission from a plurality of search and rescue resources based on the response probability and ant colony algorithm includes: The state transition probability for selecting each search and rescue resource is determined based on the roulette wheel method in the ant colony algorithm. The target search and rescue resource for performing the search and rescue mission is determined from a plurality of search and rescue resources based on the response probability and state transition probability.

5. A maritime search and rescue resource selection device, characterized in that, include: The first determining module is used to determine a first characteristic parameter for each search and rescue resource, wherein the search and rescue resources include at least aircraft and ships, and the first characteristic parameter is used to indicate the search and rescue capability of the search and rescue resource. The second determining module is used to determine the second characteristic parameters of the search and rescue mission; The calculation module is used to determine the response probability of each of the search and rescue resources performing the search and rescue mission based on the first feature parameter and the second feature parameter, wherein the response probability represents the probability that the search and rescue resource performs the search and rescue mission; The selection module is used to calculate and determine the target search and rescue resource for performing the search and rescue mission from a plurality of search and rescue resources based on the response probability and the ant colony algorithm. Also includes: Establish a module for building the response threshold model; Determining the response probability of each search and rescue resource executing the search and rescue mission based on the first feature parameter and the second feature parameter includes: The first feature parameter and the second feature parameter are processed based on the response threshold model to obtain the response probability of each search and rescue resource executing the search and rescue task; The step of processing the first feature parameter and the second feature parameter according to the response threshold model to obtain the response probability of each search and rescue resource executing the search and rescue mission includes: Based on the response threshold, the first feature parameter and the second feature parameter are processed to determine the stimulus amount and response threshold of the search and rescue resource; The response probability is calculated and determined based on the stimulus amount and the response threshold. The greater the stimulus intensity, the smaller the response threshold and the greater the response probability.

6. The apparatus according to claim 5, characterized in that, The establishment of the response threshold model includes: A first response function is established based on historical first characteristic parameters to determine the stimulus quantity of the search and rescue resource, wherein the stimulus quantity characterizes the ability of the search and rescue resource to perform the search and rescue mission; Determine the degree of cooperation among different search and rescue resources when performing search and rescue missions based on historical data; A second response function is established based on the cooperation degree to determine the response threshold of the search and rescue mission to the search and rescue resources; The response threshold model is established based on the first response function and the second response function.

Images