An ant colony optimization based multi-unmanned surface vehicle target search algorithm

Abstract

The application belongs to the field of sea unmanned vehicle search, and particularly discloses a multi-unmanned vehicle target search algorithm based on ant colony optimization. The algorithm mainly uses the ant colony optimization method, so that the unmanned vehicle cluster can search for the lost target in the shortest possible time, and ensure the optimal search trajectory. The application updates the target probability graph composite target moving route based on the Markov prediction model, guides the unmanned vehicle cluster to move to the area with high target existence probability for target search task, which not only can guarantee the best solution, but also can accelerate the convergence speed of the algorithm and improve the efficiency of the target search task. Compared with other algorithms, the target search algorithm based on ant colony optimization can effectively avoid obstacles for the unmanned vehicle cluster, and ensure the safety of the unmanned vehicle cluster in the search process. In addition, when the multi-unmanned vehicle cooperatively searches for unknown dynamic targets, it has the advantages of information sharing and regional search division, can expand the search range and shorten the time to search for the target.

Description

Technical Field

[0001] This invention belongs to the field of marine target search, and specifically designs a multi-unmanned surface vessel target search algorithm based on ant colony optimization. Background Technology

[0002] Search theory was first proposed by Bernard O. Koopman. Currently, search theory has been successfully applied in many important fields, including resource and terrain exploration, maritime rescue, and the search and strike of suspicious targets. As an NP-hard problem, the search problem is typically solved using approximate optimization algorithms or natural heuristics, such as cross-entropy optimization, particle swarm optimization, and ant colony optimization. Facing the complex marine environment, maritime target search missions need to consider the influence of ocean currents and surface obstacles on the target to predict its trajectory. Simultaneously, target search missions have specific time constraints, aiming for unmanned surface vessel swarms to successfully locate the target in the shortest possible time. Summary of the Invention

[0003] To address the problems and shortcomings of existing technologies, the present invention aims to provide a multi-unmanned surface vessel target search algorithm based on ant colony optimization.

[0004] To achieve the objectives of this invention, the technical solution adopted is as follows:

[0005] The first aspect of this invention provides a multi-unmanned surface vessel (USV) target search algorithm based on ant colony optimization, comprising the following steps:

[0006] Step 1: Create a search environment map and rasterize the map. Based on the target's location information at the last moment before it was lost, create an initial target probability map.

[0007] Step 2: Based on the search environment map, establish a Markov prediction model and a target movement probability model;

[0008] Step 3: After initializing the parameters of the ant colony optimization algorithm, the iteration begins. The ant colony searches in parallel, initializes the positions of the unmanned surface vessel and the target, and sets a movement taboo table; the parameters include pheromone concentration.

[0009] Step 4: Based on the Markov prediction model, predict the target probability map at the current moment to obtain the predicted target probability map at the current moment;

[0010] Step 5: Based on the movement taboo table and obstacle information, update the grid areas that the unmanned surface vessel can access at its current position; then, based on the predicted target probability map, calculate the heuristic information of the grid areas that the unmanned surface vessel can access.

[0011] Step 6: Combining the pheromone concentration and heuristic information, calculate the transition probability of the unmanned surface vessel (USV) from its current position to each grid cell, and determine the grid cell region that the USV will go to in the next moment based on the transition probability using a roulette wheel algorithm.

[0012] Step 7: The unmanned surface vessel moves to the next grid area to search, obtains the search results at the current moment, and updates the target probability map at the current moment;

[0013] Step 8: The ant colony search ends. Obtain the shortest search time and the corresponding unmanned surface vessel path solution for the current iteration, and update the pheromone concentration.

[0014] Step 9: Continue to the next iteration until the maximum number of iterations is allowed, and output the obtained global optimal solution, which includes the shortest time spent on target search and the optimal search path of the unmanned surface vessel swarm.

[0015] Based on the above-mentioned multi-unmanned surface vessel target search algorithm based on ant colony optimization, preferably, the target movement probability model in step 2 is:

[0016]

[0017] Among them, v t-1 Indicates the possible location of the target at the previous moment, v t This indicates the current location of the target.

[0018] Based on the above-described ant colony-optimized multi-unmanned surface vessel (USV) target search algorithm, preferably, the predicted target probability map at the current moment in step 4 is as follows:

[0019]

[0020] Among them, v t This indicates the position of the target at time t. This represents the probability map of the predicted target at time t. This represents the target probability map at the previous time step.

[0021] Based on the above-described ant colony-optimized multi-unmanned surface vessel target search algorithm, preferably, the method for calculating the heuristic information in step 5 is as follows:

[0022]

[0023] in, Indicates the position of the unmanned surface vessel at time t. Heuristic information for moving in a certain direction 'a'. This represents all the grid cells in this direction. This represents the distance between grid j and the unmanned surface vessel's position. This represents the predicted target probability value corresponding to grid j.

[0024] According to the above-mentioned multi-unmanned surface vessel target search algorithm based on ant colony optimization, preferably, the method for calculating the transition probability in step 6 is as follows:

[0025]

[0026] in Indicates that the unmanned surface vessel is in The probability of choosing action 'a' at position . unmanned surface vessel The action selection 'a' corresponds to the pheromone concentration within the grid, where α represents the pheromone importance factor and β represents the heuristic importance factor.

[0027] Based on the above-described ant colony-optimized multi-unmanned surface vessel (USV) target search algorithm, preferably, the method for updating the target probability map at the current moment in step 7 is as follows:

[0028]

[0029]

[0030] in This represents the target probability map updated at time t. Represents the sensor model, v t This represents the position of the target at time t. This represents the position information of the u-th unmanned surface vessel at time t. This represents the search result for the u-th unmanned surface vessel at time t.

[0031] Based on the above-described ant colony-optimized multi-unmanned surface vessel target search algorithm, preferably, the method for updating the pheromone concentration in step 8 is as follows:

[0032] τ (t) =(1-ρ)τ (t-1) +Δτ

[0033]

[0034] Where τ (t) τ represents the pheromone concentration in all grid cells at the current time. (t-1) This represents the pheromone concentration in all grid cells at the previous time step, Δτ represents the increase in pheromone concentration, and ρ is the evaporation coefficient τ(route) of the pheromone concentration. ib ), representing the pheromone concentration corresponding to the optimal path solution in the current iteration. This represents the reciprocal of the shortest search time obtained in the current iteration.

[0035] According to the above-mentioned multi-unmanned surface vessel (USV) target search algorithm based on ant colony optimization, preferably, in step 7, during the process of the USV moving to the next grid area for searching, the optimization objective function is also set, wherein the optimization objective function is:

[0036]

[0037] in This represents the search time corresponding to the search trajectory of the U unmanned surface vessel cluster within N time steps. This represents the target probability map updated at time t.

[0038] A second aspect of the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement any step of the multi-unmanned surface vessel target search algorithm based on ant colony optimization as described in the first aspect.

[0039] A third aspect of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a computer processor, implements any step of the multi-unmanned surface vessel target search algorithm based on ant colony optimization as described in the first aspect.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0041] (1) This invention uses an ant colony optimization algorithm to perform target search for multiple unmanned surface vessels. Compared with other algorithms, the target search algorithm based on ant colony optimization can enable the unmanned surface vessel cluster to effectively avoid obstacles and ensure the safety of the unmanned surface vessel cluster during the search process.

[0042] (2) The target search algorithm of this invention based on ant colony optimization is based on the target probability map updated by Markov prediction model and the movement route of composite targets, thereby guiding the unmanned surface vessel cluster to move towards the area with a high probability of target existence to carry out the target search task. This not only ensures that the best solution is obtained, but also speeds up the convergence speed of the algorithm and improves the efficiency of the target search task.

[0043] (3) When multiple unmanned surface vessels (USVs) are searching for unknown dynamic targets in a coordinated manner, they have the advantages of information sharing and regional search, which can expand the search range and shorten the time to find the target. Attached Figure Description

[0044] Figure 1 This is the initial target probability map of the present invention;

[0045] Figure 2 This is a search environment map for the present invention;

[0046] Figure 3 This is a diagram of the Markov prediction model of the present invention;

[0047] Figure 4 The target movement probability diagram of the present invention;

[0048] Figure 5 This is a description of the calculation of the predicted target probability map of the present invention;

[0049] Figure 6 This is a diagram illustrating the movement rules of the unmanned surface vessel (USV) according to the present invention.

[0050] Figure 7 This is a heuristic information illustration diagram for the present invention;

[0051] Figure 8 This is the updated target probability map of the present invention;

[0052] Figure 9 This refers to the search environment map in Embodiment 2 of the present invention;

[0053] Figure 10 This is the optimal search path diagram in Embodiment 2 of the present invention;

[0054] Figure 11 This is the result diagram of combining the target probability map with the unmanned surface vessel's target search in Embodiment 2 of the present invention. Detailed Implementation

[0055] To enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention will be described in detail below with reference to specific embodiments.

[0056] Example 1

[0057] A multi-unmanned surface vessel (USV) target search algorithm based on ant colony optimization includes the following steps:

[0058] Step 1: Create a search environment map and rasterize the map. Based on the target's location information at the last moment before it was lost, create an initial target probability map.

[0059] First, construct a search environment map Ω, and then discretize the search environment map Ω into Ω... x *Ω y The map is rasterized using individual grids. Then, an initial target probability map is built based on the target's location information at the last moment before it was lost. For example... Figure 1 The figure shown is the initial target probability map of this invention, centered on the position of the target at the last moment before it was lost. Since the initial target probability map satisfies a two-dimensional normal distribution model, it is denoted as b(v0). If the target must exist in the search environment map Ω, then... That is, the sum of the probabilities of the target existing in all grid cells is 1.

[0060] Step 2: Based on the search environment map, establish a Markov prediction model and a target movement probability model;

[0061] like Figure 2 This is the search environment map of the present invention. Figure 2 The irregular black areas represent sea surface obstacles, and the black arrows indicate ocean current directions. The target's movement trajectory will be affected by the marine environment. The target's movement directions within the map grid area are defined as east, south, west, and north. A Markov prediction model is established to predict the target's movement trajectory using a Markov decision process. Figure 3 The diagram shown is a Markov prediction model diagram of a preferred embodiment of the present invention. Each grid cell in the diagram contains the predicted movement direction of the target at that grid cell position.

[0062] Then, a target movement probability model is established based on the Markov prediction model. For example... Figure 4 The diagram shown is a target movement probability map of the present invention, which is composed of... Figure 4 It can be seen that the probability of the target moving in the direction predicted by the Markov prediction model is 0.7, while the probability of moving in the other three directions is 0.1. Therefore, the target movement probability model can be expressed by formula (1).

[0063]

[0064] In equation (1), v t-1 Indicates the possible location of the target at the previous moment, v t This indicates the current location of the target.

[0065] Step 3: After initializing the parameters of the ant colony optimization algorithm, start the iteration. The ant colony searches in parallel, initializes the positions of the unmanned surface vessel and the target, and sets the movement taboo table at the same time.

[0066] First, the relevant parameters of the ant colony optimization algorithm are initialized. These parameters include the maximum number of iterations allowed (NC), the ant population size (M), the initial pheromone concentration (τ) and its evaporation coefficient (ρ), as well as the pheromone importance factor (α) and the heuristic importance factor (β). The initial pheromone concentration is the same in all cells. After each iteration, the pheromone concentration in each cell is updated based on the optimal path solution of the current iteration, thus better guiding the ant colony in the target search task during the next iteration.

[0067] Then, the iteration begins, with the ant colony starting a parallel search. Each iteration requires initialization of the unmanned surface vessel (USV) and the target location. A movement taboo list is set in the search environment map for the USV, which refers to the grid areas that the USV is not allowed to pass through, including obstacle areas and areas that the USV passed through in the previous moment.

[0068] Step 4: Based on the Markov prediction model, predict the target probability map at the current moment to obtain the predicted target probability map at the current moment;

[0069] In ant colony optimization algorithms, besides pheromone concentration serving as a primary mechanism for decision-making and cooperation within the ant colony during the search process, heuristic information is also a crucial mechanism guiding the ant colony's search. Heuristic information is obtained based on problem characteristics and prior knowledge; in target search tasks, heuristic information depends on the target probability map.

[0070] Based on the target movement probability model established in step 2 and the target probability map of the previous time step, the predicted target probability map of the current time step is obtained, as shown in equation (2):

[0071]

[0072] Among them, v t This indicates the position of the target at time t. This represents the probability map of the predicted target at time t. This represents the target probability map from the previous time step. The predicted target probability map is a key reference in heuristic information.

[0073] To better illustrate the calculation process of the predicted target probability map, this invention will demonstrate it through an example, such as... Figure 5 The diagram illustrates the calculation of the predicted target probability map according to a preferred embodiment of the present invention. Based on the Markov prediction model at the current moment, the predicted probability value within grid 5 is calculated. Since the target can only move from grids 2, 4, 6, and 8 to grid 5, the predicted probability value of grid 5 is calculated from the target probability values ​​of these four grids at the previous moment. Combining the target movement probability model, i.e., formula (1) and... Figure 4 We can obtain:

[0074]

[0075] Step 5: Based on the taboo list and obstacle information, update the grid areas that the unmanned surface vessel can access at its current location, and then calculate the heuristic information of the grid areas that the unmanned surface vessel can access based on the predicted target probability map.

[0076] First, based on the taboo list and obstacle information, determine the grid area that each unmanned surface vessel can move within. For example... Figure 6 The diagram shows the movement rules of the unmanned surface vessel of the present invention. The unmanned surface vessel can move arbitrarily to the surrounding 8 grids, but it needs to exclude obstacles and grid areas in the taboo table.

[0077] Secondly, in this invention, the heuristic information is a spatial function that depends on the current position of the unmanned surface vessel. and the probability map of the predicted target like Figure 7 The diagram shown is an illustration of the heuristic information of the present invention. The heuristic calculation method for the unmanned surface vessel in a certain movable direction is shown in equation (3):

[0078]

[0079] In equation (3), Indicates the position of the unmanned surface vessel at time t. Heuristic information for moving in a certain direction 'a'. This represents all the grid cells in this direction. This represents the distance between grid j and the unmanned surface vessel's position. The greater the distance, the smaller the heuristic effect on the unmanned surface vessel's movement. This represents the predicted target probability value corresponding to grid j.

[0080] Step 6: Combining the pheromone concentration and heuristic information, calculate the transition probability of the unmanned surface vessel (USV) from its current position to each grid cell, and determine the grid cell region that the USV will go to in the next moment based on the transition probability using a roulette wheel algorithm.

[0081] Combining the pheromone concentration at the current moment and heuristic information, the transition probability of the unmanned surface vessel from its current position to each grid is calculated, as shown in equation (4):

[0082]

[0083] In equation (4), Indicates that the unmanned surface vessel is in The probability of choosing action 'a' at position . unmanned surface vessel The action to be performed is selected based on the pheromone concentration within the corresponding grid.

[0084] Step 7: The unmanned surface vessel moves to the next grid area to search, obtains the search results at the current moment, and updates the target probability map at the current moment;

[0085] The unmanned surface vessel moves to the next grid area to search, obtains the search results at the current moment, and updates the target probability map at the current moment based on the search results, as shown in Equation (5):

[0086]

[0087] In formula (5) This represents the target probability map updated at time t, which is updated based on the search results and the predicted target probability map at time t. Decide. It is a sensor model, v t This represents the position of the target at time t. This represents the position information of the u-th unmanned surface vessel at time t. This represents the search result of the u-th unmanned surface vessel at time t. If the target is found, then... This indicates that if the target is not found, then... Once the unmanned surface vessel (USV) detects a target, the target probability map stops updating; therefore, the target probability map is only updated when no target is detected. This invention focuses on trajectory optimization of the USV's target search path. Therefore, the sensor model is set as an ideal sensor model, as shown in formula (6):

[0088]

[0089] The sensor model (6) is used to determine whether a target has been detected. If no target is detected, the probability of the target being present in the grid cell where the unmanned surface vessel cluster is located in the predicted target probability map is first set to 0, and then the target probability map is updated according to formula (5). In a preferred embodiment of the present invention, the updated target probability map is as follows: Figure 8 As shown.

[0090] Meanwhile, the time required for this step is accumulated according to formula (7). If the target is found, the current search task is exited, and the time spent searching for the ants is output.

[0091]

[0092] In equation (7), This represents the time required for the target search task up to time t-1. This represents the time spent searching at time t. This indicates the time currently spent on the ant search.

[0093] During the search process, the target search time is often used as the objective function for optimization, and the quality of the unmanned surface vessel's search trajectory is evaluated through the search time. Therefore, the objective function of the algorithm described in this invention can be expressed by formula (8):

[0094]

[0095] In equation (8), This represents the search trajectory of a swarm of U unmanned surface vessels over N time steps, and is expressed in terms of... This represents the search time corresponding to this search trajectory. Let represent the probability that the unmanned surface vessel swarm has not yet detected the target at time t, and we can obtain . Therefore, the target search time actually depends on the target probability map, as expressed by formula (9):

[0096]

[0097] Step 8: The ant population search ends. Obtain the shortest search time for the current iteration in the population and its corresponding unmanned surface vessel path solution, and update the pheromone concentration.

[0098] The ant colony ends its target search, and this iteration ends. The ant that takes the shortest time to search for the target among all ant colonies is the optimal solution for the current iteration, and the time required for its target search is denoted as Time_ib. It is compared with the shortest search time in the previous iteration, and the shorter search time is taken as the global optimal search time, denoted as Time_gb.

[0099] The pheromone update process includes two stages: enhancement and evaporation. The enhancement of the pheromone is related to the optimal solution in the current iteration. The pheromone in the grid corresponding to the optimal path route_ib obtained in the current iteration will be enhanced. The enhancement magnitude is the reciprocal of the search time corresponding to this path solution. The smaller the search time, the greater the pheromone concentration of the corresponding search path will be, as shown in Equation (10).

[0100]

[0101] In equation (10), Δτ represents the increase in pheromone concentration, route_ib represents the path solution corresponding to this shortest search time, and τ(route_ib) represents the path solution corresponding to this shortest search time. ib () represents the pheromone concentration corresponding to the optimal path solution in this iteration, and pheromone enhancement will be performed on this path, with the concentration increased by (). This represents the reciprocal of the shortest search time obtained in the current iteration. The enhancement of pheromone concentration is achieved by increasing the pheromone concentration τ(route) of this path. ib Enhancement.

[0102] Meanwhile, the pheromone concentration in all grids will evaporate with each iteration. Given the evaporation coefficient ρ, the updated pheromone concentration is as shown in formula (11).

[0103] τ (t) =(1-ρ)τ (t-1) +Δτ(11)

[0104] In equation (11), τ (t) τ represents the pheromone concentration in all grid cells at the current time. (t-1) This represents the pheromone concentration in all cells at the previous time step. The pheromone update first evaporates the original pheromone concentration and then enhances the pheromone concentration in the cells corresponding to the optimal path solution.

[0105] Step 9: Continue to the next iteration until the maximum number of iterations is allowed, and output the obtained global optimal solution, which includes the shortest time spent on target search and the optimal search path of the unmanned surface vessel swarm.

[0106] Example 2

[0107] Based on the ant colony optimization-based multi-unmanned surface vessel (USV) target search algorithm described in Example 1, this example demonstrates the practical application of the algorithm. First, a 50*50 grid map is created to simulate the search area, and the corresponding map environment is set. The number of USVs is 3. For example... Figure 9 As shown in the diagram, obstacles in the environment are represented by white irregular shapes, the target is represented by a red asterisk, and the three unmanned surface vessels (USVs) are represented by triangles of three different colors. Within the search area, the three USVs entered from the edge of the search area to perform the search task, ultimately obtaining the following results: Figure 10 The optimal search path is shown.

[0108] from Figure 10 As can be seen, the three unmanned surface vessels (USVs) performed target search tasks using an ant colony optimization algorithm based on a Markov prediction model. Once the target was found on the blue trajectory corresponding to USV1, the target search task ceased, and the other USVs also stopped searching. Furthermore, the search paths of each USV demonstrate that the ant colony optimization-based target search algorithm effectively enabled the USV swarm to avoid obstacles, thus ensuring the safety of the USV swarm during the search process.

[0109] Figure 11 This is a result of combining the target probability map with the target search by the unmanned surface vessel. Figure 11 This more intuitively illustrates the target search routes initiated by unmanned surface vessels (USVs) inspired by the target probability map. The USV swarm's search routes are represented by green lines, while the target's movement routes are represented by red trajectories. It can be seen that the target probability map updated based on the Markov prediction model matches the target's movement route, thus guiding the USV swarm towards areas with a high probability of target presence for target search, effectively improving target search efficiency.

[0110] Example 3

[0111] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement any step in a multi-unmanned surface vessel target search algorithm based on ant colony optimization as described in Embodiment 1.

[0112] Furthermore, the ant colony-optimized multi-unmanned surface vessel (USV) target search algorithm described in Embodiment 1 can be implemented as a computer software program. For example, this embodiment includes a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the method. In such an embodiment, the computer program can be downloaded and installed from a network, and / or installed from a removable medium. When the computer program is executed by a processor, it performs the functions defined in the method of this application.

[0113] Example 4

[0114] A computer-readable storage medium storing a computer program that, when executed by a processor, implements any step in a multi-unmanned surface vessel target search algorithm based on ant colony optimization as described in Embodiment 1.

[0115] The computer-readable medium described in this application may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may 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, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium may 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 may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may 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 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, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0116] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0117] The above embodiments are specific implementations of the present invention, but the implementation of the present invention is not limited to the above embodiments. Any other combination, change, modification, substitution, or simplification that does not exceed the design concept of the present invention shall fall within the protection scope of the present invention.

Claims

1. A multi-unmanned surface vessel (USV) target search algorithm based on ant colony optimization, characterized in that, Includes the following steps: Step 1: Create a search environment map and rasterize the map. Based on the target's location information at the last moment before it was lost, create an initial target probability map. Step 2: Based on the search environment map, establish a Markov prediction model and a target movement probability model; Step 3: After initializing the parameters of the ant colony optimization algorithm, the iteration begins. The ant colony searches in parallel. Then, the unmanned surface vessel's position and the target position are initialized, and a movement taboo table is set. The parameters include pheromone concentration. Step 4: Based on the Markov prediction model, predict the target probability map at the current moment to obtain the predicted target probability map at the current moment: in, This indicates the position of the target at time t. This represents the probability map of the predicted target at time t. This represents the target probability map from the previous time step. Step 5: Based on the movement taboo table and obstacle information, update the grid areas that the unmanned surface vessel (USV) can access from its current position; then, based on the predicted target probability map, calculate the heuristic information of the grid areas that the USV can access. The calculation method is as follows: in, Indicates the position of the unmanned surface vessel at time t. Heuristic information for moving in a certain direction 'a'. This represents all the grid cells in this direction. This represents the distance between grid j and the unmanned surface vessel's position. This represents the predicted target probability value corresponding to grid j; Step 6: Combining the pheromone concentration and heuristic information, calculate the transition probability of the unmanned surface vessel (USV) from its current position to each grid cell, and determine the grid cell region the USV will visit at the next moment using a roulette wheel algorithm based on the transition probability; the calculation method for the transition probability is as follows: in Indicates that the unmanned surface vessel is in The probability of choosing action 'a' at position . unmanned surface vessel The action to be performed corresponds to the pheromone concentration within the corresponding grid cell. Factors indicating the importance of pheromones Indicates the heuristic importance factor; Step 7: The unmanned surface vessel moves to the next grid area to search, obtains the search results at the current moment, and updates the target probability map at the current moment; Step 8: The ant colony search ends. Obtain the shortest search time and the corresponding unmanned surface vessel path solution for the current iteration, and update the pheromone concentration. Step 9: Continue to the next iteration until the maximum number of iterations is allowed, and output the obtained global optimal solution, which includes the shortest time spent on target search and the optimal search path of the unmanned surface vessel swarm.

2. The multi-unmanned surface vessel target search algorithm based on ant colony optimization according to claim 1, characterized in that, The target movement probability model mentioned in step 2 is: in, This indicates the possible location of the target at the previous moment. This indicates the current location of the target.

3. The multi-unmanned surface vessel target search algorithm based on ant colony optimization according to claim 2, characterized in that, The method for updating the target probability map at the current time step 7 is as follows: in This represents the target probability map updated at time t. Representing the sensor model, This represents the position of the target at time t. This represents the position information of the u-th unmanned surface vessel at time t. This represents the search result for the u-th unmanned surface vessel at time t.

4. The multi-unmanned surface vessel target search algorithm based on ant colony optimization according to claim 3, characterized in that, The specific method for updating the pheromone concentration in step 8 is as follows: = in This represents the pheromone concentration in all cells at the current time. This represents the pheromone concentration in all cells at the previous time step. This indicates the increase in pheromone concentration. The evaporation coefficient of pheromone concentration. This represents the pheromone concentration corresponding to the optimal path solution in the current iteration. This represents the reciprocal of the shortest search time obtained in the current iteration.

5. The multi-unmanned surface vessel target search algorithm based on ant colony optimization according to claim 4, characterized in that, Step 7, during the process of the unmanned surface vessel moving to the next grid area for searching, also includes setting an optimization objective function, which is: in This represents the search time corresponding to the search trajectory of the U unmanned surface vessel cluster within N time steps. This represents the target probability map updated at time t.

6. An electronic device, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements any step in the multi-unmanned surface vessel target search algorithm based on ant colony optimization as described in any one of claims 1-5.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a computer processor, implements any step of the multi-unmanned surface vessel target search algorithm based on ant colony optimization as described in any one of claims 1-5.

Images